Textbook of Uncommon Cancer Third Edition
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
Textbook of Uncommon Cancer Third Edition Editors
Derek Raghavan Cleveland Clinic Taussig Cancer Center, Cleveland, OH, USA
Martin L. Brecher Roswell Park Cancer Institute, Buffalo, NY, USA
David H. Johnson Vanderbilt University Medical Center, Nashville, TN, USA
Neal J. Meropol Fox Chase Cancer Center, Philadelphia, PA, USA
Paul L. Moots Vanderbilt University Medical Center, Nashville, TN, USA
Peter G. Rose Cleveland Clinic Taussig Cancer Center, Cleveland, OH, USA Associate Editor
Ingrid A. Mayer Vanderbilt University Medical Center, Nashville, TN, USA
First published 1988, Editors: Williams, Krikorian, Green and Raghavan (Reprinted 1991, 1994); Second Edition published 1999 (Reprinted 2000); Third Edition published 2006. Copyright 1988, 1999, 2006 John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, England Telephone: (+44) 1243 779777 Email (for orders and customer service enquiries):
[email protected] Visit our Home Page on www.wiley.com All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to
[email protected], or faxed to (+44) 1243 770620. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. Cover images kindly supplied by Dan Meropol. Other Wiley Editorial Offices John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809 John Wiley & Sons Canada Ltd, 5353 Dundas Street West, Suite 400, Etobicoke, Ontario, Canada M9B 6H8 Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Library of Congress Cataloguing-in-Publication Data Textbook of uncommon cancer. – 3rd ed. / editors, Derek Raghavan . . . [et al.] p. ; cm. Includes bibliographical references and index. ISBN-13: 978-0-470-01202-4 (cloth : alk. paper) ISBN-10: 0-470-01202-1 (cloth : alk. paper) 1. Cancer. I. Raghavan, Derek. [DNLM: 1. Neoplasms. 2. Rare Diseases. QZ 200 T3556 2006] RC262.T432 2006 616.99 4–dc22 2005036102 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN-13: 978-0-470-01202-4 ISBN-10: 0-470-01202-1 Typeset in 10/11.5 pt Times Roman by Laserwords Private Limited, Chennai, India. Printed and bound by Grafos SA, Barcelona, Spain. This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production.
Contents List of Contributors . . . . . . . . . . . . . . . . . . . ix
Section 3
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix
10 Neoplastic Disorders of the Adrenal Glands . . . . . . . . . . . . . . . . . . . . . . . . . 143
Section 1 Genitourinary Cancer 1 Uncommon Tumors of the Kidney . . . . . . 1 Sujith R. Kalmadi, Ming Zhou, Andrew Novick and Ronald M. Bukowski
2 Uncommon Cancers of the Bladder . . . . 18
Endocrine Tumors
K. Oberg, A. Goldhirsch and A. Munro Neville
11 Uncommon Cancers of the Thyroid . . . 165 Mark Bloomston and Manisha H. Shah
12 Parathyroid Carcinoma . . . . . . . . . . . . . 174 Alliric I. Willis and John A. Ridge
Arlene O. Siefker-Radtke, Bogdan A. Czerniak, Colin P. Dinney and Randall E. Millikan
3 Urethral Cancer . . . . . . . . . . . . . . . . . . . . 27 Oscar E. Streeter Jr and David I. Quinn
4 Uncommon Cancers of the Prostate . . . . 38 Scott T. Tagawa, Omid Hamid, Eila Skinner and Parvesh Kumar
5 Rare Tumors of the Testis and Paratesticular Tissues . . . . . . . . . . . . . . . 66 Vedang Murthy, Cyril Fisher and Alan Horwich
Section 2 Head and Neck Cancer 6 Uncommon Tumors of the Oral Cavity and Adjacent Structures . . . . . . . . . . . . . 87 A. Robert Kagan, Stephen I. Shibata, Michael P. McNicoll and Najeeb S. Alshak
7 Rare Tumors of the Larynx . . . . . . . . . . 102 Samir S. Khariwala and Marshall Strome
8 Nasopharyngeal Carcinoma in Non-endemic Populations . . . . . . . . . . . 113 June Corry and Bonnie Glisson
Section 4
Breast Cancer
13 Metaplastic Breast Carcinoma . . . . . . . . 181 Helenice Gobbi, Ingrid A. Mayer and A. Bapsi Chakravarthy
14 Adenoid Cystic Carcinoma of the Breast 187 Melinda E. Sanders, Masako Kasami, Julie Means-Powell and David L. Page
15 Non-Hodgkin Lymphoma of the Breast . . . . . . . . . . . . . . . . . . . . . . . . . . 194 David S. Morgan and Jean F. Simpson
16 Male Breast Cancer . . . . . . . . . . . . . . . . 201 Ian K. Komenaka, Kathy D. Miller and George W. Sledge, Jr
17 Phyllodes Tumor of the Breast . . . . . . . 209 Ian Ellis, Elinor J. Sawyer, Raj Rampaul and Carlos G. Pineda
18 Carcinosarcoma of the Breast . . . . . . . . 218 B.T. Hennessy, M.Z. Gilcrease, G. Babiera, W. Yang, V. Valero and G.N. Hortobagyi
9 Esthesioneuroblastoma . . . . . . . . . . . . . 133 Barbara A. Murphy, Joseph M. Aulino, Christine H. Chung, Kim Ely, Robert Sinard and Anthony Cmelak
19 Tubular Carcinoma . . . . . . . . . . . . . . . . 230 Melinda E. Sanders, Ingrid A. Mayer and David L. Page
vi
CONTENTS
Section 5
Thoracic Tumors
20 Thymoma and Thymic Carcinoma . . . . 237 Annette M. Moore, Christopher J. Sweeney, Mark R. Wick and Patrick J. Loehrer
21 Primary Lymphomas of the Lung . . . . . 257 Francis C. Nichols and Stephen D. Cassivi
22 Primary Sarcomas of the Lung . . . . . . . 264
34 Cancer of the Small Bowel . . . . . . . . . . 391 Robert R. McWilliams, Thomas C. Smyrk and Axel Grothey
35 Unusual Tumors of the Colon, Rectum and Anus . . . . . . . . . . . . . . . . . 401 William P. Tew and Leonard B. Saltz
36 Cancer of the Appendix . . . . . . . . . . . . 410 Matthew H. Kulke and Charles S. Fuchs
Rachel E. Sanborn, Adriana L. Gonzalez, Thomas M. Ulbright, Guru Sonpavde and Alan B. Sandler
37 Gastrointestinal Stromal Tumors . . . . . . 418
23 Mesotheliomas . . . . . . . . . . . . . . . . . . . 279
38 Small Cell Carcinomas of the Gastrointestinal Tract . . . . . . . . . . . . . . 430
Giuseppe Giaccone and Paul Baas
24 Primary Melanoma of the Lung . . . . . . 293
Margaret von Mehren and Douglas Flieder
Alexandria T. Phan and Paulo M. Hoff
Richard A. Scolyer, James F. Bishop and John F. Thompson
25 Large Cell Neuroendocrine Carcinoma . . . . . . . . . . . . . . . . . . . . . . 298 William D. Travis, Lee M. Krug and Valerie Rusch
26 Carcinoid Tumors of the Lung . . . . . . . 307 Simon Chowdhury, Paul Cane, James F. Spicer and Peter G. Harper
27 Bronchioloalveolar Carcinoma of the Lung . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 Gregory J. Riely and Vincent A. Miller
28 Primary Adenoid Cystic Carcinoma of the Lung . . . . . . . . . . . . . . . . . . . . . . 321 John G. Devlin and Corey J. Langer
29 Mucoepidermoid Tumors of the Lung . . . . . . . . . . . . . . . . . . . . . . 329 Tracey L. Evans and Thomas J. Lynch
Section 6
Gastrointestinal Tumors
30 Uncommon Cancers of the Esophagus . . . . . . . . . . . . . . . . . . . . . . 337 John G. Devlin, Robert D. Odze and Jonathan D. Cheng
31 Uncommon Cancers of the Stomach . . . 352 Jordan D. Berlin and Mary K. Washington
32 Unusual Pancreatic Tumors . . . . . . . . . . 367 Ann Wexler, Roger J. Waltzman and John S. Macdonald
33 Uncommon Hepatobiliary Tumors . . . . 383 Steven J. Cohen and Natalie E. Joseph
Section 7
Gynecological Cancers
39 Extra-ovarian Primary Peritoneal Carcinomas . . . . . . . . . . . . . . . . . . . . . . 437 Alberto E. Selman and Larry J. Copeland
40 Borderline Tumors and Other Rare Epithelial Tumors of the Ovary . . . . . . 447 Teresa P. D´ıaz-Montes, Russell Vang, Deborah K. Armstrong and Robert E. Bristow
41 Stromal Tumors of the Ovary . . . . . . . . 455 Jubilee Brown, Anuja Jhingran, Michael Deavers and Maurie Markman
42 Germ Cell Tumors of the Ovary . . . . . . 467 Daniela E. Matei, Jeanne M. Schilder and Helen Michael
43 Fallopian Tube Cancer . . . . . . . . . . . . . 477 Destin Black and Richard R. Barakat
44 Uterine Sarcomas and Unusual Endometrial Carcinomas . . . . . . . . . . . 485 Peter G. Rose, Pedro F. Escobar, Peter Fleming and Charles Biscotti
45 Tumors of the Cervix . . . . . . . . . . . . . . 501 Krishnansu S. Tewari and Bradley J. Monk
46 Tumors of the Vulva and Vagina . . . . . 521 Jonathan E. Tammela, Wainwright Jaggernauth, Paulette Mhawech-Fauceglia and Shashikant B. Lele
47 Gestational Trophoblastic Diseases . . . . 532 Emily Berry and John R. Lurain
CONTENTS
Section 8 Hematological Malignancies 48 Rare Leukemias . . . . . . . . . . . . . . . . . . 543 Attaphol Pawarode and Maria R. Baer
49 Rare Lymphomas . . . . . . . . . . . . . . . . . 555 Graham A.R. Young
50 Uncommon Presentations of Plasma Cell Dyscrasias . . . . . . . . . . . . . . . . . . . 569 Rachid Baz and Mohamad A. Hussein
Section 9 Cutaneous Malignancies 51 Unusual Cutaneous Malignancies . . . . . 577 Toni K. Choueiri, Thomas Olencki, Wolfram Samlowski, Scott Florell, Sancy Leachman, Martin Majer and Allison Vidimos
52 Dermatofibrosarcoma Protuberans . . . . . 589 Michael D. Alvarado, Jane L. Messina and Vernon K. Sondak
53 Merkel Cell Carcinoma . . . . . . . . . . . . . 594 Wolfram Goessling and Robert J. Mayer
Section 10
Neurological Malignancies
54 Melanotic Lesions of the Meninges . . . 605 Paul L. Moots and Michael L. Edgeworth
55 Langerhans’ Cell Histiocytosis of the Central Nervous System . . . . . . . . . . . . 610 Rima F. Jubran and Jonathan Finlay
56 Chordomas . . . . . . . . . . . . . . . . . . . . . . 614 Herbert B. Newton
57 Meningeal Sarcomas . . . . . . . . . . . . . . . 626 Nicholas G. Avgeropoulos and John W. Henson
58 Atypical and Malignant Meningiomas . . . . . . . . . . . . . . . . . . . . 638 Samer E. Kaba and Athanassios P. Kyritsis
59 Primary Intracranial Germ Cell Tumors . . . . . . . . . . . . . . . . . . . . . . . . . 649 Jan Drappatz and Jay S. Loeffler
60 Primary Central Nervous System Lymphoma . . . . . . . . . . . . . . . . . . . . . . 657 Lisa M. DeAngelis
61 Choroid Plexus Papilloma and Carcinoma . . . . . . . . . . . . . . . . . . . . . . 667 Michael L. Edgeworth and Julie E. Hammack
vii
62 Glioma and Other Neuroepithelial Neoplasms . . . . . . . . . . . . . . . . . . . . . . 674 Paul L. Moots, Mahlon D. Johnson, Mark T. Jennings and Anthony T. Cmelak
63 Medulloblastoma and CNS Primitive Neuroectodermal Tumors . . . . . . . . . . . 695 Paul L. Moots and Mark T. Jennings
64 Craniopharyngiomas . . . . . . . . . . . . . . . 705 Gene H. Barnett and John Park
65 Ophthalmic Cancers . . . . . . . . . . . . . . . 712 Arun D. Singh, William J. Dupps, Jr and Sophie Bakri
Section 11
Pediatric Malignancies
66 Rare Pediatric Malignancies of the Head and Neck . . . . . . . . . . . . . . . . . . . 721 Ted A. James, Larry L. Myers, Nestor Rigual, Janet S. Winston, Thom R. Loree and Wesley L. Hicks
67 Uncommon Pediatric Tumors of the Thorax . . . . . . . . . . . . . . . . . . . . . . . . . 732 Joanne M. Hilden, Sharon O. Meerbaum and Louis P. Dehner
68 Uncommon Tumors of the Gastrointestinal Tract in Children . . . . . . . . . . . . . . . . . 749 Christopher L. Moertel, Jan Watterson and Louis P. Dehner
69 Uncommon Pediatric Genitourinary Tumors . . . . . . . . . . . . . . . . . . . . . . . . . 760 Barbara Bambach
70 Uncommon Endocrine Tumors in Children and Adolescents . . . . . . . . . . . 775 Raul C. Ribeiro, Carlos Rodriguez-Galindo, Gerald P. Zambetti, Bonald C. Figueiredo, Karel Pacak, Andrew Bauer and Constantine A. Stratakis
71 Uncommon Pediatric Brain Tumors . . . 798 Sharon H. Smith
72 Malignant Tumors of the Skin and Subcutaneous Tissue in Children . . . . . 810 Ilene L. Rothman, Joyce B. Farah and Thomas N. Helm
Acknowledgments . . . . . . . . . . . . . . . . . . . . 819 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 821
Contributors Najeeb S. Alshak Department of Pathology and Laboratory Medicine, Southern California Kaiser Permanente Medical Group, Los Angeles, CA, USA Michael D. Alvarado Division of Cutaneous Oncology, H Lee Moffitt Cancer Center, Tampa, FL, USA Deborah K. Armstrong Department of Medical Oncology, The Johns Hopkins Medical Institutions, Baltimore, MD, USA Joseph M. Aulino Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USA Nicholas G. Avgeropoulos Neuro-Oncology Center, Florida Hospital Cancer Institute at Florida Hospital, Orlando, FL, USA
Richard R. Barakat Gynecology Service at Memorial Sloan-Kettering Cancer Center, New York, NY, USA Gene H. Barnett Brain Tumor Institute & Department of Neurosurgery, Cleveland Clinic Taussig Cancer Center, The Cleveland Clinic and Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, OH, USA Andrew Bauer Pediatric Endocrinology, Walter Reed Army Medical Center, Washington, DC and Section on Endocrinology & Genetics (SEGEN), Developmental Endocrinology Branch (DEB), National Institutes of Health, Bethesda, MD, USA Rachid Baz Myeloma Research Program, Cleveland Clinic Foundation, Cleveland, OH, USA
Paul Baas The Netherlands Cancer Institute, Amsterdam, The Netherlands
Jordan D. Berlin Department of Medicine, Vanderbilt University Medical Center, Vanderbilt-Ingram Cancer Center, Nashville, TN, USA
G. Babiera Department of Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA
Emily Berry Department of Obstetrics and Gynecology, Northwestern University, Chicago, IL, USA
Maria R. Baer Department of Medicine, Roswell Park Cancer Institute, University at Buffalo School of Medicine and Biomedical Sciences, Buffalo, NY, USA
Charles Biscotti Section of Anatomic Pathology, Cleveland Clinic, Cleveland, OH, USA
Sophie Bakri Department of Ophthalmology, Mayo Clinic, Rochester, MN, USA
James F. Bishop New South Wales Cancer Institute, Eveleigh and Faculty of Medicine, The University of Sydney, Sydney, NSW, Australia
Barbara Bambach Department of Pediatrics, Roswell Park Cancer Institute, Buffalo, NY, USA
Destin Black Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
x
CONTRIBUTORS
Mark Bloomston Department of Surgery, The Ohio State University, Columbus, OH, USA Robert E. Bristow Department of Gynecology and Obstetrics, The Johns Hopkins Medical Institutions, Baltimore, MD, USA Jubilee Brown Department of Gynecologic Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA
Larry J. Copeland Division of Gynecologic Oncology, James Cancer Hospital and Solove Research Institute and the Comprehensive Cancer Center, The Ohio State University, Columbus, OH, USA June Corry Peter MacCallum Cancer Institute, Victoria, Australia Bogdan A. Czerniak Department of Anatomic Pathology, University of Texas M.D. Anderson Cancer Center, Houston, TX, USA
Ronald M. Bukowski Department of Experimental Therapeutics, Taussig Cancer Center, The Cleveland Clinic Foundation, Cleveland, OH, USA
Lisa M. DeAngelis Department of Neurology, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
Paul Cane Medical Oncology, Guy’s Hospital, London, UK
Michael Deavers Department of Pathology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA
Stephen D. Cassivi Division of General Thoracic Surgery, Mayo Clinic College of Medicine, Rochester, MN, USA A. Bapsi Chakravarthy Radiation Oncology, Vanderbilt University School of Medicine, Nashville, TN, USA Jonathan D. Cheng Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, PA, USA Toni K. Choueiri Taussig Cancer Center, The Cleveland Clinic Foundation, Cleveland, OH, USA Simon Chowdhury Medical Oncology, Guy’s Hospital, London, UK Christine H. Chung Division of Hematology and Oncology, Vanderbilt Ingram Cancer Center, Nashville, TN, USA Anthony T. Cmelak Department of Radiation Oncology and Vanderbilt-Ingram Cancer Center, Vanderbilt University Medical Center, Nashville, TN, USA Steven J. Cohen Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, PA, USA
Louis P. Dehner Lauren V. Ackerman Laboratory of Surgical Pathology, Barnes-Jewish and St. Louis Children’s Hospital, Washington University Medical Center, St. Louis, MO, USA John G. Devlin Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, PA, USA Teresa P. D´ıaz-Montes Department of Gynecology and Obstetrics, The Johns Hopkins Medical Institutions, Baltimore, MD, USA Colin P. Dinney Department of Urology, University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Jan Drappatz Center for Neuro-Oncology, Dana Farber/Brigham and Women’s Cancer Center, Harvard Medical School, Boston, MA, USA William J. Dupps, Jr Cole Eye Institute, Cleveland Clinic Foundation, Cleveland, OH, USA Michael L. Edgeworth Department of Neurology, Division of Neuro-Oncology, Vanderbilt University Medical Center, Nashville, TN, USA
CONTRIBUTORS
Ian Ellis Molecular Medical Sciences, University of Nottingham and Department of Histopathology, Nottingham City Hospital, Nottingham, UK Kim Ely Department of Pathology, Vanderbilt University Medical Center, Nashville, TN, USA Pedro F. Escobar Section of Gynecologic Oncology, Cleveland Clinic, Cleveland, OH, USA Tracey L. Evans Department of Medicine, Division of Hematology/Oncology, University of Pennsylvania, Philadelphia, PA, USA Joyce B. Farah Department of Dermatology, State University of New York at Buffalo, Buffalo, NY, USA Bonald C. Figueiredo Center for Molecular Genetics and Childhood Cancer Research (CEGEMPAC), Federal University of Paran´a, Curitiba, Brazil Jonathan Finlay Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Cyril Fisher Clinical Anatomic Pathology Unit, Royal Marsden NHS Foundation Trust, London, UK Peter Fleming Section of Radiation Oncology, Cleveland Clinic, Cleveland, OH, USA Douglas Flieder Department of Surgical Pathology, Fox Chase Cancer Center, Philadelphia, PA, USA Scott Florell Multidisciplinary Melanoma Program, Huntsman Cancer Institute and Department of Dermatology, University of Utah, Salt Lake City, UT, USA Charles S. Fuchs Department of Medical Oncology, Dana-Farber Cancer Institute and Channing Laboratory, Brigham and Women’s Hospital, Boston, MA, USA
xi
Giuseppe Giaccone Department of Medical Oncology, Vrije Universiteit Medical Center, Amsterdam, The Netherlands M.Z. Gilcrease Department of Pathology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Bonnie Glisson The University of Texas M.D. Anderson Cancer Center, Houston, TX ,USA Helenice Gobbi Anatomic Pathology, School of Medicine, Federal University of Minas Gerais, Belo Horizonte, Brazil Wolfram Goessling Department of Medical Oncology, Dana-Farber Cancer Institute and Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School and Hematology/Oncology, Children’s Hospital and Gastrointestinal Unit, Massachusetts General Hospital, Boston, MA, USA A. Goldhirsch Department of Medicine, European Institute of Oncology, Milan, Italy Adriana L. Gonzalez Department of Pathology, Vanderbilt University Medical Center, Nashville, TN, USA Axel Grothey Department of Oncology, Division of Medical Oncology, Mayo Clinic College of Medicine, Rochester, MN, USA Omid Hamid Department of Medicine, University of Southern California, Los Angeles, CA, USA Julie E. Hammack Department of Neurology, Division of Neuro-Oncology, Mayo Clinic, Rochester, MN, USA Peter G. Harper Medical Oncology, Guy’s Hospital, London, UK Thomas N. Helm Department of Dermatology, State University of New York at Buffalo, Buffalo, NY, USA
xii
CONTRIBUTORS
B.T. Hennessy Department of Breast Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA John W. Henson Pappas Center for Neuro-oncology, Division of Neuroradiology, Massachusetts General Hospital, Boston, MA, USA Wesley L. Hicks, Jr Department of Head and Neck Surgery, Roswell Park Cancer Institute, Buffalo, NY, USA Joanne M. Hilden Department of Hematology and Oncology, The Children’s Hospital, The Cleveland Clinic, Cleveland, OH, USA Paulo M. Hoff Department of Gastrointestinal Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA G.N. Hortobagyi Department of Breast Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Alan Horwich Academic Unit of Radiotherapy & Oncology, Institute of Cancer Research & Royal Marsden NHS Foundation Trust, London, UK Mohamad A. Hussein Myeloma Research program, Cleveland Clinic Foundation, Cleveland, OH, USA Wainwright Jaggernauth Department of Radiation Oncology, Roswell Park Cancer Institute, Buffalo, NY, USA
Mahlon D. Johnson Department of Pathology, University of Tennessee Graduate School of Medicine, Knoxville, TN, USA Natalie E. Joseph Department of Surgical Oncology, Fox Chase Cancer Center, Philadelphia, PA, USA Rima F. Jubran Division of Pediatric Hematology and Oncology, University of Southern California, Los Angeles, CA, USA Samer E. Kaba Department of Neurology, Emory University, Atlanta, GA, USA A. Robert Kagan Department of Radiation Oncology, Southern California Kaiser Permanente Medical Group, Los Angeles, CA, USA Sujith R. Kalmadi Department of Hematology and Medical Oncology, Taussig Cancer Center, The Cleveland Clinic Foundation, Cleveland, OH, USA Masako Kasami Department of Pathology, Shizuoka Cancer Center, Shizuoka, Japan Samir S. Khariwala Head and Neck Institute, Cleveland Clinic Foundation, Cleveland, OH, USA Ian K. Komenaka Indiana University Cancer Pavilion, Indianapolis, IN, USA
Ted A. James Department of Surgery, University of Vermont, Burlington, VT, USA
Lee M. Krug Department of Medicine, Thoracic Oncology Service, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
Mark T. Jennings Division of Pediatric Neurology, Children’s Hospital of Illinois, Peoria, IL, USA
Matthew H. Kulke Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
Anuja Jhingran Department of Radiation Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA
Parvesh Kumar Department of Radiation Oncology, University of Southern California, Los Angeles, CA, USA
CONTRIBUTORS
xiii
Athanassios P. Kyritsis Department of Neurology, University of Ioannina, Ioannina, Greece
Daniela E. Matei Division of Hematology and Oncology, Indiana University School of Medicine, Indianapolis, IN, USA
Corey J. Langer Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, PA, USA
Ingrid A. Mayer Medical Oncology, Vanderbilt University School of Medicine, Nashville, TN, USA
Sancy Leachman Multidisciplinary Melanoma Program, Huntsman Cancer Institute and Department of Dermatology, University of Utah, Salt Lake City, UT, USA
Robert J. Mayer Department of Medical Oncology, Dana-Farber Cancer Institute and Department of Medicine Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA
Shashikant B. Lele Department of Gynecologic Oncology, Roswell Park Cancer Institute, Buffalo, NY, USA Jay S. Loeffler Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA Patrick J. Loehrer Department of Hematology and Oncology, Indiana University School of Medicine, Indianapolis, IN, USA Thom R. Loree Department of Head and Neck Surgery, Roswell Park Cancer Institute, Buffalo, NY, USA
Michael P. McNicoll Department of Head & Neck Surgery, Southern California Kaiser Permanente Medical Group, Los Angeles, CA, USA Robert R. McWilliams Department of Oncology, Division of Medical Oncology, Mayo Clinic College of Medicine, Rochester, MN, USA Julie Means-Powell Department of Medicine, Vanderbilt University School of Medicine, Nashville, TN, USA Sharon O. Meerbaum Department of Hematology and Oncology, The Children’s Hospital, The Cleveland Clinic, Cleveland, OH, USA
John R. Lurain Department of Obstetrics and Gynecology, Northwestern University, Chicago, IL, USA
Jane L. Messina Department of Pathology, University of South Florida College of Medicine, Tampa, FL, USA
Thomas J. Lynch Division of Hematology/Oncology, Massachusetts General Hospital, Boston, MA, USA
Paulette Mhawech-Fauceglia Department of Pathology and Lab Medicine, Roswell Park Cancer Institute, Buffalo, NY, USA
John S. Macdonald Saint Vincent’s Comprehensive Cancer Center, New York, NY, USA
Helen Michael Department of Pathology, Indiana University School of Medicine, Indianapolis, IN, USA
Martin Majer Multidisciplinary Melanoma Program, Huntsman Cancer Institute and Department of Internal Medicine (Oncology), University of Utah, Salt Lake City, UT, USA
Kathy D. Miller Indiana University Cancer Pavilion, Indianapolis, IN, USA
Maurie Markman The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA
Vincent A. Miller Thoracic Oncology Service, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
xiv
CONTRIBUTORS
Randall E. Millikan Department of Genitourinary Medical Oncology, University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Christopher L. Moertel The Theodora Lang Hematology and Oncology Clinic, Children’s Hospitals and Clinics of Minnesota, St. Paul, MN, USA Bradley J. Monk The Division of Gynecologic Oncology, The Chao Family Comprehensive Cancer Center, University of California, Irvine Medical Center, Orange, CA, USA Annette M. Moore Department of Hematology and Oncology, Indiana University School of Medicine, Indianapolis, IN, USA Paul L. Moots Department of Neurology and Vanderbilt-Ingram Cancer Center, Vanderbilt University Medical Center, Nashville, TN, USA David S. Morgan Department of Medicine, Vanderbilt University School of Medicine, Nashville, TN, USA Barbara A. Murphy Division of Hematology and Oncology, Vanderbilt Ingram Cancer Center, Nashville, TN, USA Vedang Murthy Academic Unit of Radiotherapy & Oncology, Institute of Cancer Research & Royal Marsden NHS Foundation Trust, London, UK Larry L. Myers Department of Otolaryngology, University of Texas Southwestern Medical Center at Dallas, Dallas, TX, USA A. Munro Neville Department of Oncology, Imperial College of Medicine, London, UK Herbert B. Newton Dardinger Neuro-Oncology Center and Division of Neuro-Oncology, Department of Neurology, The Ohio State University Medical Center and James Cancer Hospital & Solove Research Institute, Columbus, OH, USA
Francis C. Nichols Division of General Thoracic Surgery, Mayo Clinic College of Medicine, Rochester, MN, USA Andrew Novick Glickmann Urologic Institute, The Cleveland Clinic Foundation, Cleveland, OH, USA K. Oberg Department of Medical Sciences, Uppsala University, University Hospital, Uppsala, Sweden Robert D. Odze Gastrointestinal Pathology Service, Brigham and Women’s Hospital, Boston, MA, USA Thomas Olencki Division of Hematology and Oncology, The Ohio State University College of Medicine and Public Health, Columbus, OH, USA Karel Pacak Reproductive Biology and Medicine Branch (RBMB), National Institute of Child Health and Human Development (NICHD), National Institutes of Health, Bethesda, MD, USA David L. Page Department of Pathology, Vanderbilt University School of Medicine, Nashville, TN, USA John Park Brain Tumor Institute & Department of Neurosurgery, Cleveland Clinic Taussig Cancer Center, The Cleveland Clinic, Cleveland, OH, USA Attaphol Pawarode Department of Medicine, Roswell Park Cancer Institute, University at Buffalo School of Medicine and Biomedical Sciences, Buffalo, NY, USA Alexandria T. Phan Department of Gastrointestinal Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Carlos G. Pineda Department of Histopathology, Nottingham City Hospital, Nottingham, UK David I. Quinn Department of Medicine, Division of Medical Oncology, Keck School of Medicine University of Southern
CONTRIBUTORS
California, USC Norris Comprehensive Cancer Center, Los Angeles, CA, USA Raj Rampaul Department of Surgery, Nottingham City Hospital, Nottingham, UK Raul C. Ribeiro Department of Pediatrics, University of Tennessee, Memphis, TN, USA John A. Ridge Section of Head and Neck Surgery, Fox Chase Cancer Center, Philadelphia, PA, USA Gregory J. Riely Thoracic Oncology Service, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY, USA Nestor Rigual Department of Head and Neck Surgery, Roswell Park Cancer Institute, Buffalo, NY, USA Carlos Rodriguez-Galindo Department of Pediatrics, University of Tennessee, Memphis, TN, USA Peter G. Rose Section of Gynecologic Oncology, Cleveland Clinic, Cleveland, OH, USA Ilene L. Rothman Department of Dermatology, State University of New York at Buffalo, Buffalo, NY, USA
Melinda E. Sanders Department of Pathology, Vanderbilt University School of Medicine, Nashville, TN, USA Alan B. Sandler The Vanderbilt-Ingram Cancer Center, Nashville, TN, USA Elinor J. Sawyer Department of Molecular and Population Genetics, Cancer Research UK, London, UK Jeanne M. Schilder Department of Obstetrics and Gynecology, Indiana University School of Medicine, Indianapolis, IN, USA Richard A. Scolyer Sydney Melanoma Unit, Sydney Cancer Centre, Royal Prince Alfred Hospital and Department of Anatomical Pathology, Royal Prince Alfred Hospital, Camperdown, NSW, Australia Alberto E. Selman Division of Gynecologic Oncology, Clinical Hospital, Universidad de Chile, Santiago, Chile Manisha H. Shah Department of Internal Medicine, The Ohio State University, Columbus, OH, USA Stephen I. Shibata Division of Medical Oncology, City of Hope Medical Center, Duarte, CA, USA
Valerie Rusch Department of Surgery, Thoracic Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
Arlene O. Siefker-Radtke Department of Genitourinary Medical Oncology, University of Texas M.D. Anderson Cancer Center, Houston, TX, USA
Leonard B. Saltz Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
Jean F. Simpson Department of Pathology, Vanderbilt University School of Medicine, Nashville, TN, USA
Wolfram Samlowski Multidisciplinary Melanoma Program, Huntsman Cancer Institute and Department of Internal Medicine (Oncology), University of Utah, Salt Lake City, UT, USA
Robert Sinard Department of Otolaryngology, Vanderbilt University Medical Center, Nashville, TN, USA
Rachel E. Sanborn Department of Medicine, Oregon Health and Science University Cancer Institute, Portland, OR, USA
xv
Arun D. Singh Department of Ophthalmic Oncology, Cole Eye Institute and Taussig Cancer Center, Cleveland Clinic Foundation, Cleveland, OH, USA
xvi
CONTRIBUTORS
Eila Skinner Department of Urology, University of Southern California, Los Angeles, CA, USA
Jonathan E. Tammela Department of Gynecologic Oncology, Roswell Park Cancer Institute, Buffalo, NY, USA
George W. Sledge, Jr Indiana University Cancer Pavilion, Indianapolis, IN, USA
William P. Tew Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
Sharon H. Smith Children’s Center for Cancer and Blood Diseases, St. Vincent Children’s Hospital, Indianapolis, IN, USA
Krishnansu S. Tewari The Division of Gynecologic Oncology, The Chao Family Comprehensive Cancer Center, University of California, Irvine Medical Center, Orange, CA, USA
Thomas C. Smyrk Department of Pathology, Division of Anatomic Pathology, Mayo Clinic College of Medicine, Rochester, MN, USA Vernon K. Sondak Division of Cutaneous Oncology, H. Lee Moffitt Cancer Center, Tampa, FL, USA Guru Sonpavde Deke Slayton Cancer Center and Baylor College of Medicine, Houston, TX, USA James F. Spicer Medical Oncology, Guy’s Hospital, London, UK Constantine A. Stratakis Section on Endocrinology & Genetics (SEGEN), Developmental Endocrinology Branch (DEB), National Institute of Child Health and Human Development (NICHD), National Institutes of Health, Bethesda, MD, USA
John F. Thompson Sydney Melanoma Unit, Sydney Cancer Centre, Royal Prince Alfred Hospital, Camperdown and Faculty of Medicine, The University of Sydney, Sydney, NSW, Australia William D. Travis Department of Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY, USA Thomas M. Ulbright Department of Pathology, Indiana University, Indianapolis, IN, USA V. Valero Department of Breast Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Russell Vang Department of Pathology, The John Hopkins University School of Medicine, Baltimore, MD, USA
Oscar E. Streeter, Jr Department of Radiation Oncology, Keck School of Medicine University of Southern California, USC Norris Comprehensive Cancer Center, Los Angeles, CA, USA
Allison Vidimos Department of Dermatology, Section of Dermatologic Surgery and Cutaneous Oncology, The Cleveland Clinic Foundation, Cleveland, OH, USA
Marshall Strome Head and Neck Institute, Cleveland Clinic Foundation, Cleveland, OH, USA
Margaret von Mehren Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, PA, USA
Christopher J. Sweeney Department of Hematology and Oncology, Indiana University School of Medicine, Indianapolis, IN, USA
Roger J. Waltzman Saint Vincent’s Comprehensive Cancer Center, New York, NY, USA
Scott T. Tagawa Department of Medicine, Mount Sinai School of Medicine, New York, NY and Department of Medicine, University of Southern California, Los Angeles, CA, USA
Mary K. Washington Department of Pathology, Vanderbilt University Medical Center, Vanderbilt-Ingram Cancer Center, Nashville, TN, USA
CONTRIBUTORS
Jan Watterson The Theodora Lang Hematology and Oncology Clinic, Children’s Hospitals and Clinics of Minnesota, St. Paul, MN, USA Ann Wexler Saint Vincent’s Comprehensive Cancer Center, New York, NY, USA Mark R. Wick Department of Pathology, University of Virginia Health System, Charlottesville, VA, USA Alliric I. Willis Fox Chase Cancer Center, Philadelphia, PA, USA Janet S. Winston Department of Pathology, Potomac Hospital, Woodbridge, VA, USA
xvii
W. Yang Department of Diagnostic Radiology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Graham A. R. Young Institute of Haematology, Royal Prince Alfred Hospital and University of Sydney, Sydney, NSW, Australia Gerald P. Zambetti Department of Biochemistry, University of Tennessee, Memphis, TN, USA Ming Zhou Department of Anatomic Pathology, The Cleveland Clinic Foundation, Cleveland, OH, USA
Preface More than 15 years after the first edition of Textbook of Uncommon Cancer, there is still a demand for current information on the biology and management of a broad range of uncommon tumors. In 2004, the American Society of Clinical Oncology devoted an educational symposium to this topic, a session that was very well attended, illustrating the interest of oncologists in attempting to systematize their approach to this set of problems. Although much of the available information is still quite anecdotal, an increasing body of information is becoming available regarding the management of many of these tumors, and it seems timely to update this textbook. We have asked our authors to provide, whenever possible, their own approach to management, even in the absence of defined guidelines, simply to help the clinician inexperienced in treating specific rare tumors. In order to maintain the dynamic nature of the book, we have evolved changes in the Editorial Board, and have also modified the structure of the book. In some instances, classic chapters have merely been updated, and for other topics, new authors have been recruited, either completely rewriting or updating and extending the contributions from earlier editions. We have deleted the chapters on uncommon presentations of common malignancies as they overlapped
with other sections, and as there was little to update in those domains. In this edition, there has been some duplication as certain histologies present at multiple sites and we believe that it is appropriate to provide extensive site-specific management guidelines where appropriate. John Wiley & Sons, Ltd have provided new editorial support, consequent upon the well-earned retirement of our initial guide, Dr. Richard Edelstein. We are most grateful for the active assistance and guidance from Deborah Russell and Layla Paggetti, and the staff of John Wiley. As always, our families and our support staff have made the sacrifices necessary to underwrite the publication of a large textbook, and we appreciate them greatly, but recognize that it is just not enough. DEREK RAGHAVAN MARTIN L. BRECHER DAVID H. JOHNSON NEAL J. MEROPOL PAUL L. MOOTS PETER G. ROSE INGRID A. MAYER May 2006
Section 1 : Genitourinary Cancer
1
Uncommon Tumors of the Kidney Sujith R. Kalmadi, Ming Zhou, Andrew Novick and Ronald M. Bukowski
INTRODUCTION Tumors of the kidney account for about 3% of adult malignancies, with an incidence of approximately 36 000 new cases/year in the United States.1 Renal cell carcinomas (RCC) constitute the bulk of these malignancies and historically they were widely known as hypernephroma, a term that was coined by Grawitz in the 19th century reflecting his belief that they arose from the adrenal gland. RCCs are derived from the epithelial cells of renal tubules and account for more than 80% of primary renal malignant neoplasms. Transitional cell carcinomas, although arising in the renal pelvis, are frequently classified as renal tumors, and account for 7 to 8%. Other tumors, such as oncocytomas, collecting duct carcinomas (CDCs) of Bellini, and renal sarcomas, are uncommon but are becoming more frequently recognized pathologically. Nephroblastoma [Wilms’ tumor (WT)] is common in children and accounts for 5 to 6% of all primary renal tumors. This chapter focuses on the classification, pathology, genetics, clinical and radiographic manifestations, and surgical and systemic management of the less-common malignant and benign tumors of the kidney.
CLASSIFICATION OF RENAL NEOPLASMS The first classification of renal tumors was proposed by Konig2 in 1826. This classification was made on the basis of gross morphologic characteristics. Extensive study of renal neoplasms in the last couple of decades has led to a standardized nomenclature by the European and American authorities (see Table 1).3 The 2004 WHO classification, which includes nearly 50 distinct entities, is based on a combination of immunohistochemistry, histology, and clinical and genetic features that are widely accepted and relatively reproducible.4 Several large series have shown this classification to have prognostic significance5 and that it is relevant to diagnosis by fine needle aspiration techniques.6 It is anticipated this current classification system will be reassessed in 5 years.7
MOLECULAR DIAGNOSTIC TECHNIQUES IN RENAL NEOPLASMS The application of molecular and cytogenetic techniques has resulted in improved understanding of these tumors. Routine use of these techniques for diagnosis is cumbersome, although they serve as useful adjuncts in selected cases. In the forefront among these technologies are comparative genomic hybridization (CGH), fluorescent in situ hybridization (FISH), allelic loss analysis, classical cytogenetics, and karyotyping. The von Hippel-Lindau tumor suppressor gene resides in the short arm of chromosome 3 (3p25.3) and is commonly inactivated by gene mutation or promoter hypermethylation in sporadic clear cell RCC. It is also the causative gene for the von Hippel-Lindau syndrome.8,9 Papillary RCC, chromophobic RCC, carcinoma of the collecting ducts of Bellini, metanephric adenomas, and renal oncocytomas have also exhibited characteristic chromosomal anomalies.10 – 12 Recently, RCC associated with chromosomal translocation involving TFE3 gene on Xp11.2 has been described as a distinct clinicopathological entity.13 Differentiating the various histological subtypes of renal tumors is crucial, since they differ in their prognosis and therapeutic response to treatment. Definitive diagnosis based on H&E morphology is possible in the majority of cases. In the minority circumstances where distinction becomes difficult, immunohistochemistry and other molecular techniques are being increasingly relied upon to make the distinction.14 Renal cell carcinoma marker (RCC Ma) is a monoclonal antibody against the proximal tubular brush border antigen, which is relatively specific for renal neoplasms that originate from the proximal renal tubules including clear and papillary RCC, despite a rather low sensitivity.15 The antibody is positive in nearly 80% of clear cell and papillary RCC, is variably expressed in chromophobe RCC, and is absent in oncocytomas and CDCs. CD10 is another marker that helps in the differential diagnosis, by being expressed in clear cell and papillary RCC, and absent in chromophobe RCC and oncocytomas.16 Vimentin is variably expressed in clear cell and papillary RCC and is absent in chromophobe RCC and oncocytomas.17 Cytokeratins represent a widely
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
2
GENITOURINARY CANCER
Table 1 WHO histological classification of tumors of the kidney.3 (Adapted from reference 3, by permission of IARC Press).
Renal cell tumors Clear cell renal cell carcinoma Multilocular cystic renal cell carcinoma Papillary renal cell carcinoma Chromophobe renal cell carcinoma Carcinoma of the collecting ducts of Bellini Renal medullary carcinoma Xp11 translocation carcinoma Carcinoma associated with neuroblastoma Mucinous tubular and spindle cell carcinoma Renal cell carcinoma, unclassified Papillary adenoma of the kidney Oncocytoma Metanephric tumors Metanephric adenoma Metanephric adenofibroma Metanephric stromal tumor Nephroblastic tumors Occurring mainly in children Clear cell sarcoma Rhabdoid tumor Congenital mesoblastic nephroma Ossifying renal tumor of infants Occurring mainly in adults Leiomyosarcoma (including renal vein) Angiosarcoma Rhabdomyosarcoma Malignant fibrous histiocytoma Hemangiopericytoma Osteosarcoma Angiomyolipoma Epithelioid Angiomyolipoma Leiomyoma Hemangioma Lymphangioma Juxtaglomerular cell tumor Renomedullary interstitial cell tumor Schwannoma Solitary fibrous tumor Mixed epithelial and mesenchymal neoplasms Cystic nephroma Mixed epithelial and stromal tumor Synovial sarcoma Neuroendocrine tumors Carcinoid Neuroendocrine carcinoma Primitive neuroectodermal tumor Neuroblastoma Phaeochromocytoma Hematopoietic and lymphoid neoplasms Lymphoma Plasmacytoma Leukemia Germ cell tumors Teratoma Choriocarcinoma Metastatic tumors
used diagnostic immunohistochemical marker in differentiating renal tumors. Cytokeratin 7 (CK7) is strongly positive in most chromophobe RCC, absent in clear cell RCC, and variably expressed in oncocytomas. Routine metaphase cytogenetics, performed on cultured tumor cells, has been used to identify cytogenetic changes associated with each RCC
histological subtypes. Using specific probe sets, FISH can be used to identify those RCC with characteristic chromosomal alterations.
PAPILLARY ADENOMA Background Historically, adenomas were recognized as lesions smaller than 3 cm based on the work of Bell.18 This was then modified in 1970 by Murphy and Mostofi who felt that histological differentiation of adenomas from true adenocarcinomas was possible.19 Renal papillary adenomas are small, discrete, and arise from the renal tubular epithelium. In autopsy studies, they increase in frequency with age (7 to 40%).
Pathology By the most recent WHO classification, papillary adenomas are less than 5 mm in diameter with a low nuclear grade. They appear as pale yellow-gray, well-circumscribed nodules, generally below the renal capsule in the renal cortex. They are usually not encapsulated, however, some have thin pseudocapsules. On microscopic examination, they have tubular, papillary, or tubulopapillary architecture similar to papillary renal cell carcinoma. The cells have scanty cytoplasm with round to oval nuclei and do not have high nuclear grade (Fuhrman nuclear grade 3 or 4). Cytogenetic features include trisomy (chromosome 7 and 17) and loss of the Y chromosome.20,21 The resemblance of this tumor to renal papillary carcinoma has led to the view that it may represent a precursor lesion of RCC.
Clinical Presentation and Treatment Most of these lesions are discovered incidentally. They tend to appear more frequently in patients with underlying kidney disease related to atherosclerosis, scarring, acquired renal cystic disease secondary to hemodialysis, and other malignant conditions of the kidney.22 With small tumors being increasingly detected incidentally during radiological procedures, the current view is to regard all of them as probable early cancers until an unequivocal marker of benignity is discovered. Renal tumors with diameters 0.5–2 cm often behave in an indolent fashion, although the biological behavior is difficult to ascertain; therefore tumors less than 2 cm are sometimes termed “renal epithelial tumors of uncertain malignant potential”, and are observed for progression, while larger tumors may require surgical excision, depending on the clinical scenario. There is no defined role for radiotherapy or chemotherapy in the management of renal papillomas.
CARCINOMA OF THE COLLECTING DUCTS OF BELLINI Background CDC is a rare renal tumor derived from the cells of the collecting duct of Bellini, and comprises less than 1% of renal malignancies. It was reported by Mancilla-Jimenez et al. in their report on papillary tumors, that examination of kidney tissue distant from the tumor displayed, in some
UNCOMMON TUMORS OF THE KIDNEY
3
cases, atypical hyperplastic changes of collecting tubules. This raised the possibility that some papillary tumors arose from distal tubular epithelium.23 Fleming and Lewi later described the detailed pathological features of CDC as a distinct entity based on several case reports.24
a distinct pattern has not yet emerged. The immunohistochemistry profile is variable, being commonly positive for phytoagglutinins and high-molecular-weight cytokeratin, with coexpression of vimentin; and negative for CD10 and villin.25
Pathology
Clinical Presentation
This tumor is characterized by a medullary location, with a size ranging from 2–12 cm, firm whitish-gray appearance, and irregular infiltrative border. It grows radially from the renal hilum to invade into the renal cortex, the renal capsule, and the renal sinus. Histologically, it has an irregular tubulopapillary growth pattern embedded in a desmoplastic stroma (see Figure 1a). The tubules are lined with hobnail cells with a scant eosinophilic cytoplasm. The cells display high-grade nuclei with brisk mitotic activity, and prominent nucleoli (see Figure 1b). Occasionally, sarcomatoid changes or mucin can be seen. Molecular events and cytogenetic changes, which contribute, are poorly characterized; and
It is a highly aggressive tumor, which usually presents in an advanced stage, with gross hematuria, abdominal/back pain, and a flank mass. At diagnosis, it will frequently have distant metastases in the lung, liver, lymph nodes, bone, or adrenal gland. It is more common in males (ratio of about 2 : 1) with a wide range of age-groups (13–83 years with a mean of 55 years). On computerized tomography (CT) scan, this tumor appears as a centrally arising infiltrative mass with preservation of the renal contour and minimal enhancement with contrast. Patients can have generalized inflammatory symptoms secondary to cytokine release from the tumor and the inflammatory reaction associated with the tumor.24,25
Treatment and Prognosis
(a)
(b) Figure 1 Collecting duct renal cell carcinoma consists of high-grade tumor cells (b) forming complex and angulated tubules or tubulopapillary structures embedded in a remarkably desmoplastic stroma (a).
The diagnosis of CDC is generally made postoperatively, since radiological distinction from other RCCs is difficult and there is a low preoperative suspicion in view of the rarity of the disease. The prognosis is generally very poor with most patients rapidly developing systemic metastases and with a median survival of 22 months.26 The role of nephrectomy has been debated because of the frequent metastases at presentation. Radical nephrectomy in the setting of metastatic CDC appears to be useful only for palliation.27 On the basis of the pathologic, immunohistochemical, and cytogenetic similarity to urothelial (TCC) carcinomas as compared with conventional clear cell RCC, the preferred approach in the treatment of metastatic disease has been with chemotherapy rather than immunotherapy.28 In the largest reported series, Dimopoulos et al. reported retrospectively the M.D. Anderson cancer center’s experience involving 12 patients with CDC treated from 1980–90.26 Seven of eight patients with metastatic disease were treated with different chemotherapy combinations, with the methotrexate vinblastine doxorubicin and cisplatin (MVAC) regimen being the most common. Only one patient achieved a minor response lasting 5 months. Six patients were treated with a combination of interleukin-2 and IFN-α with a response seen in one patient. Peyromaure et al. reported two complete responses with cisplatin and gemcitabine combination chemotherapy, which lasted 9 and 27 months.29 Radiotherapy in these series has appeared to have minimal benefit for local recurrence. Chao et al. in a review noted that some patients having regional nodal disease without distant metastasis have had long-term disease-free survival with adjuvant therapy.30 While the overall benefit of chemotherapy or immunotherapy appears to be minimal, there appears to be a select group of patients who will benefit from these approaches. Cisplatin-gemcitabine has significant activity with a favorable toxicity profile in urothelial cancers and has evoked some significant response in patients with CDC. We thus view this regimen, to some extent, as the
4
GENITOURINARY CANCER
preferred first-line chemotherapy regimen, simply as it is less toxic than the MVAC regimen, and no other large series of chemotherapy of collecting duct tumors has yielded apparently better results.
RENAL MEDULLARY CARCINOMA Background Renal medullary carcinoma was first described by Davis et al.31 as a sickle cell nephropathy and termed so because of its predominantly medullary location. Prior to this report, many of these tumors were probably mistakenly classified as CDCs due to their histological resemblance to the latter. In a literature review of renal medullary carcinoma by Dimashkieh et al.,32 hemoglobinopathy was found in 53 of the 55 cases (50 patients had hemoglobin AS, two patients had hemoglobin SC, and one patient had hemoglobin SS disease).
Pathology Renal medullary carcinoma is a centrally located tumor with an infiltrative growth pattern similar to that of CDC. It is believed to arise from the epithelium of the distal portion of the collecting duct. The right kidney is involved three times more commonly than the left kidney, and the mean tumor size ranges from 4–12 cm (mean of 7 cm). Renal medullary carcinomas are widely infiltrative, and have variable areas of hemorrhage and necrosis. Histologically, a variety of growth patterns have been described with reticular growth pattern and compact adenoid cystic morphology being the common features. Most renal medullary carcinomas have areas of poorly differentiated cells with solid sheets of tumor cells. The tumor cells contain vesicular or clear nuclei with prominent nucleoli and amphophilic cytoplasm, which can have a squamoid or rhabdoid quality. The tumor cells are usually high grade and as with CDC, there is often marked desmoplasia and inflammation.33 – 36 The immunohistochemical profile is similar to CDC but can be helpful in distinguishing renal medullary carcinoma from other poorly differentiated kidney tumors. The clinical scenario is the key to diagnosing this rare neoplasm.
Clinical Presentation Renal medullary carcinoma is a highly aggressive tumor that occurs almost exclusively in young people (mean age 22 years), predominantly males (male to female ratio 2 : 1) with sickle cell disease or trait. The common presenting symptoms are gross hematuria, abdominal/flank pain, or weight loss. Metastatic disease in the lymph nodes or distant organs such as the brain can also be the initial evidence of the tumor. Of the patients with adequate staging information available from the two largest case series, 18% had stage III disease and 82% had stage IV disease on presentation.31,35,37
Treatment and Prognosis Renal medullary carcinomas are now widely regarded as a highly aggressive variant of RCC, with an almost uniformly
fatal outcome. The mean survival after surgery has been about 4 months. Strouse et al. have reported that only one of the over 80 reported patients is alive at 2 years.37 This patient had a small tumor (<2 cm) confined to the kidney at the time of resection. Chemotherapy has been shown to increase survival beyond 4 months in anecdotal reports, but with no reported long-term survivors. In this review of chemotherapy, of the 15 patients assessable for response, there was one complete response, two partial responses, one minor response, one stable disease and ten patients had progressive disease. The most common chemotherapy regimen used was MVAC. Radiation therapy in an adjuvant or palliative role was disappointing. Immunotherapy in a few patients also had disappointing results. In healthy patients with systemic disease, treatment plans similar to those for urothelial cancers and CDC, with combination chemotherapy (consisting of cisplatin-gemcitabine or the MVAC regimen) appears to be a reasonable choice, but is unsupported by specific data. We are unaware of any collaborative clinical trials addressing this issue.
XP11 TRANSLOCATION NEOPLASMS Background Xp11 translocation neoplasms are a subset of RCCs, characterized by various translocations involving chromosome Xp11.2, resulting in fusion of the TFE3 gene to a variety of recipient genes and overexpression of TFE3 protein. They affect children and young adults predominantly, although some older patients have been reported. Molecular analysis of several different Xp11.2-translocation carcinomas has shown that some bear a translocation that is identical to the breakpoints and ASPL-TFE3 gene fusion seen in alveolar soft part sarcoma (ASPS).13 These have been recognized as a distinctive subclass in the 2004 WHO renal tumor classification. Recently, an unusual renal epithelial neoplasm with a chromosomal translocation involving t(6 : 11) has been described, which resulted in the overexpression of another transcription factor gene in the same gene family as TFE3, TFEB.38,39
Pathology On gross examination they closely resemble conventional (clear cell) renal carcinomas. They are tan-yellow, and often necrotic and hemorrhagic. The most distinctive histopathologic appearance is that of papillary structures lined with clear cells; a finding that is uncommon in other renal carcinomas (see Figure 2). They often have a nested architecture, with cells containing clear and granular cytoplasm. The histology of Xp11-translocation carcinomas varies with specific chromosomal translocations. The ASPL –TFE3 renal carcinomas are notable for cells with voluminous, clear to eosinophilic cytoplasm; discrete cell borders; vesicular nuclear chromatin; and prominent nucleoli. They were previously labeled as the “voluminous cell variant” of pediatric RCC. Hyaline nodules containing psammoma bodies can be seen. In comparison, the PRCC –TFE3 renal carcinomas ordinarily have less abundant cytoplasm; fewer hyaline nodules with psammoma bodies; and a more nested, compact
UNCOMMON TUMORS OF THE KIDNEY
5
Pathology
Figure 2 Renal cell carcinoma associated with t(X;17)(p11.2;q25) consists of nested to pseudopapillary structures lined with tumor cells with abundant clear, sometimes eosinophilic, cytoplasm.
architecture. The immunohistochemistry is positive for TFE3 protein, RCC marker antigen, and CD 10. In contrast traditional RCC, only half of them express epithelial markers such as cytokeratin and epithelial membrane antigen.
Clinical Presentation and Prognosis Although RCC is uncommon in children (<5% of all renal tumors), approximately a third of RCC in children and young adults appears to belong to this translocation family of tumors. The clinical behavior of these tumors is still not well-characterized. Generally, these carcinomas present at an advanced stage, with lymph node metastases at diagnosis despite their small size in a majority of cases. Even with these advanced stage presentations, the clinical course tends to be variable. This clinical behavior resembles that of ASPS, which has a similar genetic translocation. It can recur 20 or 30 years after the initial diagnosis. Resection is the primary modality of treatment. Only anecdotal reports of the adjuvant use of chemotherapy and immunotherapy in pediatric patients are available.13
Grossly, they are well circumscribed, solid or cystic, large (mean size 6 cm), with a gray or light tan appearance, and a homogeneous, myxoid, and glistening cut surface. Microscopically, they are composed of compressed tubular structures and spindle cells separated by pale mucinous stroma. These can have a configuration simulating leiomyoma or sarcoma.46 Mitotic activity is minimal with a low Furhman nuclear grade of 1–2. The morphologic, immunohistochemical, and ultrastructural features of these lesions indicate origin from distal nephron segments. They have a complex and variable immunohistochemical profile; with positivity for a wide range of cytokeratins and epithelial membrane antigen. Proximal nephron markers such as CD 10 and villin are mostly absent. They also show positivity for Ulex europaeus, peanut, and soya bean agglutinins.42 On electron microscopy, they have usual epithelial features resembling the loop of Henle or a distal convoluted tubule. Cytogenetic changes involving chromosome losses of 1, 4, 6, 8, 13 and 14 and gains of chromosomes 7, 11, 16, and 17 have been described.
Clinical Presentation Although more common in middle aged and elderly women, these tumors can occur in a wide age range (17–82 years with a mean of 53 years) and predominantly in women (female to male ratio of 4 : 1). They usually present as asymptomatic masses on radiologic studies. Occasionally, they can present with flank pain or hematuria.
Treatment and Prognosis Previously these mucinous tubular and spindle cell carcinoma/tumors may have been misdiagnosed as leiomyoma, sarcoma, or CDC. It is crucial to make the diagnosis of these tumors as being distinct from the other tumors in view of the indolent nature of this disease to prevent undue and excessive treatment. Generally, these tumors are of low pathologic stage at the time of excision, and they behave in an indolent fashion; only 1 of 34 patients experienced metastasis during follow-up periods of up to 24 years. Fatal local recurrence has been reported in one patient.
MUCINOUS TUBULAR AND SPINDLE CELL CARCINOMA
ONCOCYTOMA
Background
Background
These tumors were first described by Farrow et al. from the Mayo Clinic, who collected a series of eight distinctive tumors that appeared to originate from the collecting tubules.40 Further studies of this collection of tumors resulted in the classification of these tumors as “low-grade collecting duct carcinoma”,41 a term that may not accurately reflect the biology of these tumors. Several other reports also postulated that they represented a low-grade CDC with less biologic aggressiveness.25 However, subsequent characterization and reports of these tumors showed that they had clinical and pathologic features quite distinct from CDC. Currently, they are recognized as a separate entity of renal cell tumor in the WHO classification.3,42 – 45
Oncocytomas first came to attention as benign kidney tumors after a case series reported by Klein and Valensi in 1976.47 They constitute about 3–5% of renal tumors in most large series.48 However, tumors similar to oncocytomas had been described earlier, by Zippel et al. The term oncocyte means “swollen cell” because of numerous cytoplasmic mitochondria. Similar tumors can occur in the salivary gland, thyroid, parathyroid, and adrenal sites.49
Pathology Oncocytomas are well-circumscribed, nonencapsulated neoplasms with a characteristic central stellate scar, seen in 33%, and most commonly in large tumors (see Figure 3a).
6
GENITOURINARY CANCER
(a)
(b) Figure 3 Renal oncocytoma forms a well-circumscribed, nonencapsulated mass with homogeneous cut surface and a central scar (a). The tumor cells (oncocytes) are uniform, round to polygonal with granular eosinophilic cytoplasm and regular round nuclei with evenly dispersed chromatin. They are nested in a loose hypocellular and hyalinized stroma (b).
The median size is about 5 cm; however, they can be as large as 20 cm. The color is classically mahogany-brown, but can be tan to pale yellow. Hemorrhage can be seen in 20%. They are composed of solid nests and sheets of oncocyte with abundant granular eosinophilic cytoplasm, residing in an edematous, mucopolysaccharide-rich extracellular stromal matrix (see Figure 3b). They are thought to arise from the intercalated cells of the collecting duct. The nuclei usually do not exhibit pleomorphism; with an evenly dispersed
chromatin, discrete central nucleoli and mitotic activities are rare to absent. Features such as perinephric fat and lymphovascular invasion can be seen and do not appear to confer a worse prognosis.50,51 However, “atypical features” such as gross involvement of the renal vein, extensive papillary architecture, foci of clear cells, sarcomatoid dedifferentiation, prominent necrosis, and frequent or atypical mitoses carry a different connotation and are inconsistent with a diagnosis of oncocytoma.52 The most common differential diagnoses for oncocytomas are chromophobe RCC and clear cell RCC with eosinophilic cells. Making a distinction between these tumors on a cytologic aspiration specimen can be difficult.53 Hale colloidal iron staining, parvalbumin, and vimentin are negative (although focal luminal), as opposed to the diffuse cytoplasmic, colloidal iron staining that can be observed in oncocytoma; and antimitochondrial antibody is positive in oncocytoma and can help in the differential diagnosis.50 On electron microscopy, they are characterized by numerous normal-appearing mitochondria accounting for the cytoplasmic granularity; leading to the use of the term “mitochondrioma” by some authors.49,52,54 Microvesicles that are seen in chromophobe RCC are absent in oncocytoma. Oncocytosis is a condition where the kidneys contain multifocal oncocytomatous nodules, with oncocytic changes in the renal tubules and cysts in the surrounding regions of the kidney. Multifocality and bilaterality can occur in 5–13% of resected oncocytomas. It behaves in a similar fashion to the solitary tumors. Sometimes, a lesion can contain both oncocytomatous and chromophobe RCC components; a condition referred to as “hybrid oncocytic tumor (HOT tumor)”. Therefore, it is crucial to thoroughly examine and adequately sample a lesion that grossly resembles an oncocytoma.55 The coexistence of oncocytoma with chromophobe RCC and their morphologic similarities have sparked a debate as to whether these two entities represent the two extreme ends of the spectrum with a common origin from the intercalated cells of the collecting duct. Recent molecular evidence suggests that renal oncocytoma and chromophobe RCC share not only some morphological similarities, but early cytogenetic alterations also, including loss of chromosomes Y, 1, and 14. Birt–Hogg–Dube syndrome is a familial, autosomaldominant syndrome where the gene, folliculin, has been localized to the short arm of chromosome 17. This is characterized by dome-shaped skin papules in the facial area, renal tumors (27% of patients), pulmonary cysts, and spontaneous pneumothorax.56 The most common renal tumor is a hybrid of chromophobe RCC and oncocytoma with multiple tumors in a majority. The possibility of this familial syndrome should be entertained in a diagnosis of renal oncocytosis.57
Clinical Presentation Oncocytoma can present anywhere between the ages of 14–90 years with no gender predilection. It is detected mostly as an incidental finding on routine imaging studies. Occasionally, patients may have hematuria, flank pain, or a palpable mass. On CT scan, the lesion is usually hypodense,
UNCOMMON TUMORS OF THE KIDNEY
well circumscribed, and peripherally located with a central scar.58 It has a “spoke-wheel pattern” on angiography.
Treatment and Prognosis Current literature supports the benign nature of the disease, with surgery being curative.59 Metachronous lesions have been reported as late as 9 years after initial diagnosis.60 If the clinical data or preoperative information confirms oncocytoma, it can be treated safely by partial nephrectomy. Therefore, preoperative diagnosis can avoid excessive treatment. However, most patients do undergo nephrectomy because of the inability of current diagnostic methods to reliably distinguish this from renal cell carcinoma; and the occasional coexistence of chromophobe RCC and clear cell RCC with oncocytoma. [Editorial Note: Metastatic oncocytoma of the kidney is an extremely rare entity, if it exists at all. Prior to initiating any form of treatment for this purported entity, histo-pathological review at a subspecialty center of excellence is crucial. Although there are no defined approaches to management, radiotherapy to isolated lesions may cause local tumor response and reduction of symptoms. Cisplatinbased chemotherapy is a reasonable choice if some form of chemotherapy is required, but unsupported by any extensive literature.]
7
The higher prevalence of angiomyolipoma in females has raised the speculation that there may be a sex hormonal potentiation of angiomyolipoma growth. Progesterone and estrogen receptors have been detected in these tumors along with reports of rapid growth during pregnancy. The immunohistochemical profile is characterized by a coexpression of melanocytic (e.g. HMB 45) (see Figure 4b) and smooth muscle (e.g. smooth muscle actin) markers. They can also be positive for CD68, neuron specific enolase, S100, desmin, and hormonal receptors; whereas the epithelial markers are consistently negative.3,66 – 68 Two genes are known to cause tuberous sclerosis, TSC gene 1 (TSC1) on chromosome 9q34, which encodes hamartin, and TSC2 on chromosome 16p13, which produces tuberin, a GTPase activating protein. The biology of these genes has been elucidated extensively in recent years. Angiomyolipoma frequently has loss of heterozygosity at one of the two TSC loci in both the sporadic and tuberous sclerosis-associated tumors.62
ANGIOMYOLIPOMA Background Renal angiomyolipoma is a benign mesenchymal lesion initially described by Grawitz et al. in 1900.61 They account for approximately 1% of surgically removed renal tumors. It was also described by Bourneville and Brissaud as part of the tuberous sclerosis complex (TSC) around the same time.62,63 Hartwick et al. reported in 1989 an epithelioid angiomyolipoma with uncommon histologic patterns mimicking malignancy, which after further investigation had revealed malignant characteristics in some cases. It has recently been classified as a variant of angiomyolipoma with malignant potential.46
(a)
Pathology Grossly, renal angiomyolipoma are large, yellow-gray mass lesions, well demarcated from the kidney, but not truly encapsulated. Generally they are solitary, but if present in multiple numbers, the picture is that of a dominant tumor with associated smaller lesions. When they grow larger, they usually cause a mass effect rather than infiltrate into surrounding tissue. They are composed of a varying proportion of thick-walled, poorly organized blood vessels, smooth muscle bundles, and mature adipose tissue (see Figure 4a), and the varying proportion of each component accounts for the gross appearance of the lesion. Smooth muscle cells are frequently spindle shaped and occasionally rounded epithelioid. Regional lymph node involvement is thought to represent multicentric involvement rather than metastases.64,65 Infrequently, direct tumor extension into the inferior vena cava and the renal venous system, in the absence of distant metastases, has been described.
(b)
Figure 4 (a) Angiomyolipoma characteristically consists of 3 elements : blood vessels (lower right), adipocytes (lower left) and smooth muscle cells (upper). Melanocytic markers, such as HMB-45, are positive in some tumor cells (b).
8
GENITOURINARY CANCER
Epithelioid angiomyolipoma is a potentially malignant mesenchymal neoplasm closely related to classic angiomyolipoma. There is a higher association (>50%) with tuberous sclerosis. Patients are frequently symptomatic, both sexes are equally affected, and the mean age of presentation (38 years) is generally younger as compared with classic angiomyolipoma. Scarcity of adipose tissue makes radiologic diagnosis more difficult. About a third of reported epithelioid angiomyolipoma cases have metastasis to lymph nodes, liver, lungs, or spine.69 – 72 Microscopically there are sheets of epithelioid cells with abundant granular eosinophilic cytoplasm, enlarged vesicular nuclei, and prominent nucleoli. Areas of classic angiomyolipoma can be found interspersed focally. Tumors with necrosis, increased mitotic activity, nuclear atypia, and infiltration into surrounding tissue should be regarded as potentially malignant. The immunohistochemical profile is positive for melanocytic markers such as HMB-45, but not as consistently positive for smooth muscle markers such as actin.3,46 Although rare, epithelioid angiomyolipoma should be kept in perspective when a diagnosis of angiomyolipoma is made.
Clinical Presentation The prevalence of this tumor in the general population is low. A series of 8501 autopsies in patients without the TSC showed angiomyolipoma in 2 males and 25 females.73 A Japanese study of 12 970 males and 4971 females using population-based ultrasound screening identified 13 males (0.1%) and 11 females (0.22%) with angiomyolipoma, respectively, as confirmed by CT scan or tissue diagnosis.74 Angiomyolipoma can present either sporadically (80%) or in association with the TSC (20%). There is a female preponderance (4 : 1) in the sporadic form, but not in the tuberous sclerosis-associated tumors. Tuberous sclerosis is an autosomal-dominant disease with incomplete penetrance. It was initially described by Bourneville in 1880, in a girl with mental retardation, epilepsy, and characteristic sclerotic brain lesions (tubers).63 Bourneville and Brissaud later in 1900, noted the association of the syndrome with renal tumors. Vogt described the classic triad of seizures, mental retardation, and adenoma sebaceum.75 This triad has been substituted by a constellation of findings that establish the diagnosis. When patients with tuberous sclerosis are followed up, more than half of them show development of angiomyolipoma. In a pooled analysis by Nelson et al., they suggest that tuberous sclerosis-associated angiomyolipoma, in contrast with sporadic cases, is likely to present at an earlier age (mean age 30 vs 52 years), with larger tumors (8.9 vs 5.4 cm), frequent multicentricity (97 vs 13%) and hemorrhage (44 vs 14%).76 Steiner et al. found that tuberous sclerosis-associated tumors were more likely to grow (67 vs 21%) and require surgical intervention (50 vs 28%) during the 4 years of follow-up.77 De Luca et al. reported on 51 patients with sporadic angiomyolipoma who had either immediate surgery or were on observation. Ninety-two percent of the observed patients (mean tumor size 1.5 cm) showed no radiographic growth of angiomyolipoma during the 5-year follow-up. The larger tumors (>4 cm) were more likely to grow (46 vs
27%) or require surgical intervention (54 vs 7%) than smaller tumors.78 Angiomyolipoma was once considered a rare benign hamartoma. However with increased use of imaging studies, it is now recognized as a relatively common lesion. It can present with symptoms such as flank pain, a palpable tender mass, and gross hematuria or as an incidental finding on radiologic studies. Morbidities secondary to angiomyolipoma are related to retroperitoneal hemorrhage (Wunderlich syndrome)79 and renal failure secondary to encroachment of normal renal tissue.80 The radiologic diagnosis has been refined in recent years with the advent of helical CT scan and thin cut sectioning. Identification of fat in a renal lesion is the key to radiologic diagnosis. Some of the lesions contain a minimal amount of fat that may not be detected and may lead to a nephrectomy. There have been reports of fat in RCC tumors, secondary to invasion and entrapment of perirenal fat. MRI can help in differentiating angiomyolipoma from RCC in select cases and in evaluation during pregnancy. Diagnosis of incidental angiomyolipoma should also prompt a workup for tuberous sclerosis. Historically there was an association between renal cell carcinoma, angiomyolipoma, and the tuberous sclerosis syndrome. Clear cell renal cell carcinoma appears to develop at a higher rate in patients with tuberous sclerosis syndrome than in general. Eble et al. also suggested that some tumors historically diagnosed as renal cell carcinoma in this set of patients may actually have been epithelioid angiomyolipoma, overestimating the association.81
Treatment and Prognosis Renal angiomyolipomas are typically slow-growing tumors, and their morbidity is secondary to their growth. With the lack of randomized trials, there is considerable controversy as to the exact indication for treating asymptomatic angiomyolipoma, and organ preservation with partial nephrectomy is a valid approach in this setting. Asymptomatic, small, benign-appearing lesions can be observed. Minimal hematuria will usually resolve with hydration and bed rest. In this approach, patients should be cautioned to avoid contact sports and should be followed up closely to evaluate the growth pattern of the lesion. The primary reasons to intervene should be suspicion of malignancy in a lesion with low fat content, alleviation of symptoms secondary to spontaneous hemorrhage, and risk of rupture or other complications. European surveys indicate that 1% of the tuberous sclerosis population is dependent on dialysis, emphasizing the importance of preserving renal function.82 Surgical removal or core biopsy with immunohistochemical staining should be considered when there is no diagnostic certainty. Symptomatic angiomyolipoma can be managed by angio-embolization or surgical removal. When clinical risk factors such as childbearing age, large tumor, suspected TSC or anticipated difficulty with periodic reimaging are present, asymptomatic angiomyolipoma can be managed by observation or intervention, according to individual preference. Retrospective data demonstrate that patients presenting with symptoms or hemorrhage are more likely to have
UNCOMMON TUMORS OF THE KIDNEY
larger tumors. Oesterling et al. proposed a 4-cm threshold for the risk of symptoms and intervention in asymptomatic patients.83 Prospective data from De Luca et al.78 and others suggest that larger lesions may become symptomatic with time, especially in patients with the TSC. These studies have also shown that it is not necessary to treat all large asymptomatic lesions, since they may grow slowly without morbidity. On the basis of available evidence, we suggest an intervention in an asymptomatic patient should be based on the comprehensive evaluation of the clinical scenario including tumor size, tumor growth pattern, presence of tuberous sclerosis, patient comorbidities, renal function, pregnancy plans, and compliance. Angiographic embolization has become a common modality for nonsurgical management of these tumors. In a review of embolization by Nelson et al., complications were reported in 10% of cases, including abscess (5%) and pleural effusion (3%). One death from respiratory failure in a patient with preexisting lung disease was also reported. Recurrent bleeding or persistent symptoms necessitated repeat embolization and surgical intervention in 14% and 16% of the patients, respectively. The advantages of selective embolization includes preservation of kidney function, short recovery time of 2–5 days, and the ability to embolize bleeding vessels selectively without major surgery. The effect of embolization in reducing tumor size appears to be durable at a median follow-up of 23 months. Embolization appears to be appropriate for acute hemorrhage caused by tumor rupture, resulting in stabilization and elimination of the need for surgical treatment. Elective tumor embolization may also be appropriate for symptomatic patients with multiple renal tumors in whom renal function preservation is crucial for preventing dialysis dependence, and also for patients with limited renal reserve or poor performance status.76 Surgical intervention for angiomyolipomas is usually reserved for patients with symptoms not responsive to conservative measures, with lesions having renal vein or soft tissue invasion, or with suspicion of malignancy on imaging. Nelson et al. reported that 19% of patients underwent enucleation or partial nephrectomy, 35% underwent complete nephrectomy, 6% underwent embolization, and 40% observation.76 Nephron-sparing surgery is the preferred modality because of the benign nature of the tumor. FazeliMatin and Novick reported 27 patients with angiomyolipoma who underwent partial nephrectomy, of whom 21 had a solitary or impaired contralateral kidney.84 All operated kidneys were functional after surgery, including seven with tumors larger than 12 cm. None of the patients required dialysis postoperatively and none had recurrent angiomyolipoma symptoms at a median follow-up of 39 months. Total nephrectomy should be reserved for patients with a nonfunctioning kidney secondary to replacement by angiomyolipoma, with a strong evidence of malignancy on radiologic or pathologic examination, and in situations where other conservative measures have been unsuccessful. Angiomyolipoma with thrombus involving the renal vein, inferior vena cava and the right atrium has been reported. This can be asymptomatic or can present with life-threatening
9
thromboembolic consequences. Because of the risks of embolism, the traditional intervention has included total nephrectomy with tumor thrombectomy, although there are several reports that angiomyolipoma has been treated with embolization followed by partial nephrectomy to preserve kidney function. Pregnancy can complicate the management of young women with angiomyolipoma. While the incidence of hemorrhage during pregnancy is low, the consequences can be catastrophic with potential harm to the mother and the fetus. The hormonal links also suggest that these tumors may grow faster secondary to the altered milieu, leading to tumor rupture. Optimal diagnostic methods may be limited by pregnancy, and it may be difficult to distinguish tumor rupture from uterine or placental rupture. This lends greater credence to the belief that women with known angiomyolipoma greater than 4 cm, who intend to conceive should be treated prophylactically to avoid the risk of rupture.
CARCINOIDS Background Primary renal carcinoid is an extremely rare, welldifferentiated neuroendocrine tumor with unclear etiology. The origin of this tumor is still unclear, since neuroendocrine cells are not normally present in the renal parenchyma. There have been about 50 cases reported in the literature, with a strong association with horseshoe kidney in about 20% of the cases.85 The relative risk for a person with horseshoe kidney to develop this tumor is estimated to be 82-fold higher than the general population.86 It was first reported by Resnick in 1966 in a patient with carcinoid syndrome;87 however, the majority of the organ-confined tumors do not produce the carcinoid syndrome, which is akin to carcinoids in other organ systems.
Pathology Grossly, it presents as a solitary well-circumscribed, moderately firm tumor with a bulging appearance. The color is variable, appearance is homogeneous with variable focal hemorrhage and calcification.3 The histopathologic features appear to be similar to carcinoid tumors in other organ systems. The cells are uniform in size and arranged in a trabecular pattern. They have small nuclei with “salt and pepper” chromatin and eosinophilic cytoplasm. Immunohistochemical staining is positive for chromogranin, neuron specific enolase, and keratin. Variable positivity for serotonin, pancreatic polypeptide, prostatic acid phosphatase, and vasoactive intestinal polypeptide has been reported.88
Clinical Presentation Most patients present with an asymptomatic mass; however, they can also present with abdominal pain, mass, or hematuria. Workup for alternate origin is indicated when a lesion is discovered in the kidney, because of the rarity of primary carcinoid of kidney.85 Carcinoid symptoms are present in less than 10% of patients on presentation.3,88,89 The median age at diagnosis is 50 years, with no sex predilection. On CT scan it
10
GENITOURINARY CANCER
appears as a circumscribed solid mass with occasional calcification or cystic changes.89 Some authorities have suggested that an octreotide scan can contribute to accurate staging and diagnosis.90,91
Treatment and Prognosis Because of the rarity of the disease, the clinical outcome is difficult to predict. This is also complicated by the fact that a significant proportion of patients with metastatic disease have a prolonged survival. Complete excision of localized disease in the kidney appears to have good long-term results, with the limited reports available. Carcinoid tumors that arise in horseshoe kidneys tend to have a more indolent course.85 Carcinoid crisis secondary to release of vasoactive substances can occur with biopsy or surgical resection of the tumor and can be managed with somatostatin. In patients with a solitary metastasis, especially in the liver, it would be appropriate to consider resection of the primary tumor with the metastatic lesions, because of the lack of effective systemic treatment options. Radiation is effective for short-term palliation. Limited information on the therapy for metastatic renal carcinoid is available. In patients with the carcinoid syndrome, symptom control with somatostatin or its analogues such as octreotide is possible.92 For individuals with metastatic disease, systemic treatment options are based on clinical trials in patients with gastrointestinal carcinoids. Interferon produces tumor regression (15%) and biochemical responses in patients with metastatic disease.93 Combination chemotherapy is of limited value. In a Eastern cooperative oncology group (ECOG) trial, 118 patients with metastatic carcinoid tumor were randomized for treatment with streptozotocin combined with cyclophosphamide or with 5-fluorouracil (5-FU). Objective response rates among the evaluable patients treated with the 5-FU combination was 33% and with the cyclophosphamide combination 26%, with substantial toxicity in both regimens.94
LYMPHOMA Background Lymphomatous involvement of the kidney occurs in three distinct clinical scenarios. Most commonly, lymphoma of advanced stage involves the kidney secondarily. Posttransplantation lymphoproliferative disease (PTLD) can also involve the kidney secondary to iatrogenic immunosuppression. Primary renal lymphoma (PRL) is the least common. The incidence of PTLD arising in transplant kidneys has been rising in the last couple of decades because of the increasing frequency of transplantations. Secondary renal involvement tends to be bilateral and is seen with a high incidence (37–47%) in advanced disease. Whether PRL truly exists is still a controversial issue because of its rarity, about 60 cases have been reported, and only about 30 cases truly fulfill the diagnostic criteria as PRL.95,96 This chapter will focus on PRL.
Pathology Nephrectomy specimens in PRL can have a homogeneous, firm, pale appearance with occasional tumor thrombus or
renal vein involvement. The most common pattern of involvement is a diffuse involvement with lymphoma cells permeating between the nephrons, a so-called “interstitial pattern”. However, nodular involvement with discrete masses and intravascular lymphoma has also been described. Diffuse large B cell is the most common histologic type, although Burkitt’s, lymphoblastic lymphoma, and other histologies have been described. The origin of these tumors is still controversial, since renal parenchyma does not contain any lymphoid tissue. PTLD in transplant kidneys are related to the degree of immunosuppression and EBV (Epstein Barr Virus) infection and can present as monoclonal or polyclonal process.3
Clinical Presentation Patients can present with flank/abdominal pain, fever, cachexia, renal insufficiency, or hematuria. Acute renal failure is a common complication. CT scan is the most sensitive and efficient diagnostic modality for renal lymphoma. In a review by Urban et al.,97 the typical patterns of involvement are single and multiple masses, renal invasion from retroperitoneal disease, perirenal disease, and diffuse renal infiltration. Patients with PRL are usually treated by nephrectomy, because PRL is regarded clinically as a renal epithelial tumor. Once the renal involvement of lymphoma is confirmed, a thorough search for extrarenal disease and staging studies, such as CT scans and bone marrow biopsy, are warranted to rule out a secondary lymphoma, as the latter is much more common (30 times more common). Stallone et al.95 have proposed that the diagnosis of PRL be made only when the following criteria are fulfilled: 1) lymphomatous renal infiltration, 2) nonobstructive kidney enlargement and 3) no extrarenal localization at the time of diagnosis.
Treatment and Prognosis Although there have been reports of modest disease-free survival after nephrectomy of PRL, the prognosis is generally poor due to dissemination to secondary sites. Early detection and systemic combination chemotherapy may reverse the renal failure and improve survival by preventing dissemination.95,98 – 100 Secondary renal lymphomas are usually seen in the setting of advanced lymphoma and have a dismal prognosis. PTLD is treated by reducing the immunosuppression if possible, although recent reports with the anti-CD20 monoclonal antibody rituximab have been encouraging and is commonly used as frontline therapy in the appropriate setting.
RENAL SARCOMAS Background Primary renal sarcoma in adults are rare, representing approximately 1% of all primary tumors of the kidney.101 – 103 Sarcomatoid components can be seen in approximately 5% of RCC, including clear cell, papillary, chromophobe, and CDCs. It should not be confused with primary renal sarcomas, as the two entities have entirely different biology, pathology, and clinical features.3,104
UNCOMMON TUMORS OF THE KIDNEY
Pathology Any sarcomas that arise in other parts of the body can occur in the kidney, including leiomyosarcoma, osteosarcoma, malignant fibrous histiocytoma (MFH), angiosarcoma, rhabdomyosarcoma, and synovial sarcoma. As in other parts of the body, the diagnosis, and classification are traditionally based on H&E histology and immunohistochemistry.105 However, molecular studies have been increasingly used in the classification of sarcomas. For example, synovial sarcoma of the kidney, has a characteristic chromosomal translocation t (X;18) between the SYT gene on chromosome 18 and a member of the SSX family gene on chromosome X.106,107 Leiomyosarcoma constitutes the majority of primary renal sarcomas, and fewer than 50 cases of renal MFH have been described. Tumor grade, which is recognized as an important prognostic factor in soft tissue sarcomas, is also believed to be prognostic in primary renal sarcomas.
Clinical Presentation
11
which is a pediatric tumor identical to the stromal component of metanephric adenofibroma.111 Most of the clinical and pathologic characteristics of metanephric adenoma were outlined around the mid-1990s in two large series.112,113 Metanephric adenofibroma was then identified as a biphasic tumor with both epithelial and stromal elements and presented mostly in children and young adults. Metanephric tumors as a group are highly cellular benign epithelial tumors, with close relationship to WT and are conceptualized by some as being the benign, well-differentiated end of a spectrum of tumors that also includes WT as its malignant counterpart.111,114
Pathology These tumors are variable in diameter from 3–15 cm. They are usually unicentric, sharply circumscribed without a capsule (see Figure 5a). The cut surface is gray to yellow, with foci of hemorrhage, cystic changes, and necrosis being uncommonly present. Histologically they are
Pain and a palpable mass are the most common presenting complaints. They can spread along tissue spaces and attain a large size before they are symptomatic, similar to retroperitoneal sarcomas. Gross hematuria may be present. Systemic symptoms are less commonly reported. The prevalence of primary renal sarcomas increases with age. Metastasis is generally hematogenous to involve the lungs, liver, and bone. CT and MRI scans are used in the evaluation of these tumors to define the local extent, vascular relations, and involvement of adjacent organs. Preoperative radiographic imaging of the chest should also be performed since this is one of the most common sites of metastatic disease.
Treatment and Prognosis Complete surgical excision is the mainstay of treatment of soft tissue sarcomas at any location. Adjuvant radiation therapy, although used in locally extensive disease, has not been proven to prevent local recurrence or increase survival. The use of adjuvant chemotherapy in sarcomas other than extremity sarcomas is still experimental. Because of the rarity of this disease in the kidney, the role of either chemotherapy or radiation should be considered investigational. On the basis of case reports, complete surgical extirpation of the organ-confined tumor appears to offer patients the only reasonable chance for prolonged survival.101 – 103,108 The best outcome is seen with small tumors (<5 cm) of low histological grade that are confined to the kidneys. Surgical resection of locally recurrent or oligo-metastatic disease may be beneficial in select patients.109
(a)
METANEPHRIC NEOPLASMS Background These tumors were first described in 1980 by Pag`es and Granier in the French literature.110 Several reports of similar lesions described under a variety of names then emerged and confirmed them as a distinct entity. They include metanephric adenoma (predominantly epithelial), metanephric adenofibroma, and metanephric stromal tumor (stromal tumor)
(b)
Figure 5 Metanephric adenoma is sharply demarcated from the adjacent renal parenchyma (a). The tumor cells are closely packed to form tubules with inconspicuous lumens. They have striking uniform appearance with scant cytoplasm and smooth chromatin (b).
12
GENITOURINARY CANCER
composed of tightly-packed small, round acini. Half the tumors contain papillary structures, which resemble primitive glomeruli. Psammoma bodies are frequently present. No blastemal elements are present. The stroma can be inconspicuous or edematous. The cells are generally cuboidal with monotonous appearance, scanty cytoplasm, and small, uniform nuclei with inconspicuous nucleoli3,111 – 113 (see Figure 5b). Metanephric adenofibroma is composed of nests of epithelial elements similar to metanephric adenoma embedded in bands and sheets of fibroblast-like spindle cells. The proportion of spindle cells and epithelial components in these tumors varies.115,116 Metanephric stromal tumor, as the name implies, is very similar to the stromal component of the metanephric adenofibroma. The immunohistochemical profile includes positive WT1 and CD56; and negative epithelial membrane antigen and CK7. The differential diagnosis usually includes papillary renal cell carcinoma (PRCC) and epithelial predominant WT. PRCC is more common in males, tends to be multifocal, with a pseudocapsule, and a different immunohistochemical profile (WT-1 negative, EMA, and CK7 positive). Epithelial predominant WT is usually seen in younger patients, with a pseudocapsule, and cells are more atypical with abundant mitotic activity. Immunohistochemically, metanephric tumors and WT share some similarity – both being positive for WT-1 antigen and negative for EMA and CK7. However, CD56 is positive in metanephric tumors, and negative in WT.
Clinical Presentation Metanephric adenoma can occur in children and adults; however, it is predominantly seen in the fifth and sixth decades of life, with a distinctive female preponderance (female to male ratio 2 : 1). Metanephric adenofibroma is generally seen in children and young adults from 5 months to 36 years, with a male preponderance. Metanephric stromal tumors are seen mostly in children, with rare adult case reported. These tumors as a group comprise less than 1% of renal cell neoplasms. Most of these cases are incidentally discovered during radiologic studies and workup for incidental hematuria. Radiologically, metanephric adenoma presents as a hypovascular tumor protruding extrarenally. When symptomatic, they can cause abdominal pain and hematuria. Erythrocytosis has been reported in patients at presentation.112
Treatment and Prognosis These tumors are benign, with the exception of a few controversial case reports. If metanephric adenoma is suspected from the clinical findings, it is important to obtain an intraoperative diagnosis in order to avoid excessive resection. Erythrocytosis associated with these tumors resolve following complete resection. No local recurrence or distant metastasis has been reported for metanephric stromal tumors. WT has been reported to have arisen in metanephric adenofibroma and metanephric stromal tumor, pointing to the possible common origin of these entities.116 Renal angiodysplasia associated with these lesions can cause morbidity secondary to vascular complications. Resection without adjuvant chemotherapy is the preferred modality of treatment.
ADULT WILMS TUMOR(SYNONYM: NEPHROBLASTOMA) Background WT is the most common malignant renal tumor in children. It affects approximately 1 in every 8000 children without a significant sex predilection, and about 450 new cases are reported yearly in the United States. Ninety-eight percent of all cases occur in children below the age of 10 years; and less than 300 cases of adult WT have been reported in the literature. It tends to occur with almost the same incidence across the globe, suggesting the absence of an environmental factor. However, the incidence in the United States is highest in African-Americans and lowest in Asians, indicating a possible genetic predisposition.3,117 Its true incidence in adults is difficult to ascertain because it is included with renal cell carcinoma in epidemiological reports, and varying diagnostic criteria are utilized in case reports. Currently, most experts use the following criteria to define adult WT: (i) primary renal neoplasm, (ii) primitive blastematous spindle or round cell component, (iii) formation of abortive or embryonal tubular or glomeruloid structure, (iv) no area of tumor diagnostic of RCC, (v) pictorial confirmation of histology, and (vi) age >15 years.118,119 Approximately 1–2% of pediatric WT have a familial origin; however, this has not been reported in adults.
Pathology In contrast with childhood WTs, which are often multicentric and bilateral, most of the adult WT cases are unincentric, with multicentric and bilateral disease reported in 7 and 5% of patients, respectively. Horseshoe kidneys are associated with a two-fold higher incidence of WT. The gross and microscopic appearance of adult WT otherwise tends to resemble pediatric WT. Gross appearance of WT varies and reflects the proportion of stromal and nonstromal components. Generally WT is pale gray or tan and has a soft consistency; however, tumors with predominant stroma may be white and firm. Cyst formation may be prominent in certain cases. WT contains varying proportions of undifferentiated blastemal cells and differentiated cells of epithelial and stromal lineage. Blastemal cells are undifferentiated, small, mitotically active, rounded or oval, and densely packed with scant cytoplasm. They can occur in several distinctive growth patterns within individual tumors, including diffuse, nodular, serpentine, and basaloid. The epithelial component of WT can manifest as primitive tubules with rosette-like forms and occasionally as glomeruloid structures. Heterologous epithelial differentiation with squamous, mucinous, or ciliated epithelial components can be detected. The stromal component can exhibit significant diversity; but is usually composed of spindle cells with a myxoid background. Heterologous elements including skeletal muscle, cartilage, bone, adipose tissue, and neural tissue can be present. The histological diversity of WT is a hallmark. Characteristically, it has a so-called triphasic pattern with blastemal, epithelial, and stromal components, although monophasic pattern with only one
UNCOMMON TUMORS OF THE KIDNEY
component, and biphasic pattern with two components, are also seen often. Chemotherapy can alter the morphology by inducing maturation of blastemal, epithelial, and stromal elements leading to a disproportionate reduction of actively proliferating cells as compared to the prechemotherapy specimen. Metastatic WT may comprise a single element or a combination of what is present in the primary tumor. WT-1 antigen is usually identified in the blastemal and epithelial elements, but not in differentiated epithelial or stromal components.120,121 Nuclear anaplasia associated with an adverse outcome has been recognized in 5% of pediatric cases and increases in prevalence with age and in certain populations (e.g. AfricanAmericans). Anaplasia requires the presence of multipolar mitotic figures, marked nuclear enlargement (3 times that of nonanaplastic nuclei), and nuclear hyperchromasia. The prognostic significance is more profound in diffuse anaplasia as compared to focal anaplasia. Extensive blastemal cells have also been identified as an adverse prognostic factor. Nephrogenic rests are foci of abnormally persistent embryonal renal tissue that are capable of developing into WT. The presence of diffuse or multifocal nephrogenic rests is defined as nephroblastomatosis. There are two variants of nephrogenic rests called perilobar nephrogenic rests (PLNR) and intralobar nephrogenic rests (ILNR). They can be seen in 25–45% of pediatric WT and have also been seen in adult WT patients, as well as in several ectopic sites outside the kidney. Pediatric WT has been associated with a number of wellknown syndromes and genetic mutations. WAGR (Wilms’ tumor, aniridia, genitourinary anomalies, mental retardation) and Denys-Drash (gonadal dysgenesis, early-onset nephropathy) syndromes are associated with deletion or mutations of WT1 gene (11p13), a gene critical for renal and gonadal development. Beckwith-Wiedemann syndrome (Hemihypertrophy, macroglossia, omphalocele, and visceromegaly) is associated with loss of imprinting on WT2 (11p15). Because of the lack of sufficient cases, the genetics and syndromic associations have not been well elucidated in adults.
Clinical Presentation The most common clinical presentation of an adult WT is flank pain, hematuria, abdominal mass, or constitutional symptoms. Hypertension, which is commonly present in pediatric WT, has not been commonly reported in adult WT. While Kilton et al.122 had described 42% of their patients being symptomatic for more than a year prior to diagnosis; this has not been seen in other adult case series.118,120,123 – 126 The tumors are usually fairly large on initial presentation. A varicocele may signal obstruction of the spermatic vein secondary to tumor thrombus in the renal vein or inferior vena cava. Acquired von Willebrand’s disease has been associated with pediatric WT and testing for this is warranted in adult patients with clinical bleeding tendency. CT scan of the chest and abdomen should be done preoperatively to evaluate for metastases and extrarenal WT. Intravascular extension involving the renal vein and IVC can be seen. The most common sites for metastasis of WT are the lung, lymph nodes, and the liver. Metastasis to the bone is unusual and a
13
bone scan or skeletal survey is warranted only in the presence of symptoms. The staging system used by the Children’s Oncology Group (COG), Societe International d’Oncology Paeditrique/ International society of pediatric oncology (SIOP) and National Wilms tumor study group (NWTSG) has been accepted by most adult WT authorities in the staging of WT (see Table 2). The staging is based on both radiologic and surgical pathology data. The prognosis of WT in adults is worse than children, possibly because of a constellation of factors including frequent advanced stage disease at presentation, higher incidence of nuclear anaplasia, higher incidence of recurrence, poorer tolerance for aggressive treatment, and poorer response to treatment.
Treatment and Prognosis The prognosis of adults with WT is poor as compared to children who have an 85% chance of being cured. This success in pediatric WT represents a paradigm change to multimodality treatment.127 Historically in the 1980s, most adult case series had reported a long-term survival of about 25%. Prior to the report by Arrigo et al. of 27 patients reported to NWTSG from 1979–87 with a 3-year overall survival of 67%, it was believed that this high rate of cure could not be achieved in adults.124 This series included six stage I, five stage II, four stage III, eleven stage IV, one stage V patients and four patients with anaplastic histology. In this series, 26 patients underwent nephrectomy, 25 received chemotherapy, and 20 received radiation treatment. This led to their recommendations that patients with stage I disease and favorable histology should be treated with surgery followed by 6 months of postoperative chemotherapy using actinomycin-D and vincristine without postoperative radiation therapy; and for stage II, III, and IV/FH, vincristine, actinomycin-D and doxorubicin for 15 months along with radiation of 2000 cGy to the tumor bed, 1200 to 1500 cGy to the lungs, 2000 cGy to the liver, and 3000 cGy to other sites as appropriate in patients with metastases at diagnosis. Kattan et al. reported the French experience in 22 adult patients from 1973–92.125 Their series included four stage I, eight stage II, three stage III and seven stage IV patients. All patients underwent nephrectomy followed by single modality adjuvant treatment in seven patients Table 2 WT staging system.
Stage
Definition
I
Tumor limited to the kidney and is resected completely Tumor extends beyond kidney, but is resected completely Gross or microscopic residual tumor present and confined to the abdomen Hematogenous metastasis or lymph node metastasis outside the abdomen and pelvis Bilateral renal involvement at diagnosis; tumor in each kidney should be separately substaged
II III IV V
14
GENITOURINARY CANCER
(radiotherapy in one and chemotherapy in six) and combined modality in 15 patients. The chemotherapeutic agents used most often were actinomycin-D, vincristine and doxorubicin. Two of seven (29%) and 7/15 (47%) patients were disease-free after first-line treatment. Salvage chemotherapy had to be given in 13 patients. After a mean follow-up of 100 months, 12/22 patients (55%) were alive, including ten who were disease-free (45%). They recommended aggressive treatment, including the three-drug regimen (actinomycinD+vincristine+doxorubicin), regardless of stage, and irradiation starting from stage II. Terenziani et al. reviewed the Italian experience with 17 adult patients between 1983–2001 who were treated with an Italian protocol and were followed for a median of 131 months.127 This included eight patients with stage II, four patients with stage III and five patients with stage IV and included one patient with anaplasia. Sixteen patients underwent nephrectomy, fifteen patients received chemotherapy (ten with two drugs and five with three drugs) and seven patients received radiation. The 5-year disease-free survival was 45% with an overall survival of 62%. Reinhard et al. reviewed the German experience which included 30 adult patients on the SIOP 93–01 study.126 Ten patients (33%) had metastatic disease at presentation. There was a predominance of higher stage (stage I, 8; stage II, 7; stage III, 15 patients), with histology revealing intermediate risk in 23 patients and high risk in 2 patients. Twentysix patients underwent primary radical nephrectomy and the other four patients received neoadjuvant chemotherapy prior to surgery. Nineteen patients received the intermediate risk chemotherapy and 11 patients received the high-risk chemotherapy as per protocol. Intermediate risk chemotherapy included vincristine, actinomycin-D ± doxorubicin for 18–27 weeks; and the high-risk regimen was etoposide, carboplatin, ifosfamide, and doxorubicin for 34 weeks. Fourteen patients received local radiation from 15–35 Gy and three patients received radiation to metastatic sites. Complete remission was obtained in 24 patients (80%) with an event-free survival of 57% and an overall survival of 83% with a median observation time of 4 years. These four contemporary modern series have lain to rest the skepticism regarding multimodality treatment of WT in adults. Risk adapted multimodality treatment approach similar to pediatric WT protocols is the current standard of care. In contrast with the pediatric population, where the opinion differs as to whether nephrectomy should be done primarily or after neoadjuvant treatment, there is a consensus opinion that primary surgery is advisable for adult WT because of the difficulty in establishing this rare diagnosis preoperatively. Only in cases of primarily inoperable patients should diagnosis be established by biopsy and neoadjuvant treatment initiated to attempt regression of the tumor and enhance operability. In the absence of bilateral disease, which is rare in adults, the primary surgery should include a radical nephrectomy with lymph node sampling. Although a complete resection of all viable tumor is desirable, surgical effort that may endanger vital organs is not advisable, because local control can be achieved by adjuvant treatment. Because of the rarity of the disease, there are no established treatment guidelines in adult WT. Treatment should
preferably be done in a tertiary center with experience in this disease. On the basis of the recommendations available in the literature, including the four-case series cited above, current, and previous NWTSG trial experience in the pediatric population, we suggest the following:128 1. On the basis of the NWTSG trial data and previous multinational experience, radiation treatment can be avoided in stage I patients and in those stage II patients with a favorable histology, when they are treated with a combination chemotherapy regimen such as vincristine and actinomycin-D. Radiation should be used in the adjuvant setting in stage III and IV postoperatively. Radiation should probably also be given to metastatic sites. 2. Chemotherapy should be used in the adjuvant setting in all stages in patients with adequate organ function and performance status. Stage I can be treated with a two-drug regimen for 18 weeks. Chemotherapy for stage II needs to be risk-adapted based on the histology and can vary from two-to-four drug regimens from 18–24 weeks. Chemotherapy for stage III and stage IV also needs to be risk-adapted based on the histology and requires three- or four-drug regimens from 18–24 weeks. The chemotherapy regimens utilized should be based on the existing pediatric experience. In the recently-closed NTWS-5 study, the two-drug regimen was vincristine and actinomycin-D for 18 weeks; the three-drug regimen was vincristine, actinomycin-D, and doxorubicin for 24 weeks; and the four-drug regimen was vincristine, doxorubicin, cyclophosphamide, and etoposide for 24 weeks. 3. Patients with bilateral tumors (stage V disease) should be given primary chemotherapy for about 6–8 weeks followed by nephron-sparing bilateral partial nephrectomy in an attempt to preserve normal renal tissue. Additional chemotherapy and radiation treatment may be needed after the surgery. In an earlier series by Byrd et al., it was noted that adults are at risk for relapse for a greater period of time as compared with children.123 This has not been supported by more recent series.124,126 Recurrent disease in children has been treated successfully with radiation, multiagent salvage chemotherapy regimens (etoposide, carboplatin, and ifosfamide),129 or highdose chemotherapy with stem cell support130 leading to long-term remissions (30–60%).
REFERENCES 1. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55(1): 10 – 30. 2. Konig G. Practical Treatment of Diseases of the Kidney as Explained by Case Histories. Leipzig, Germany: C. Cnobloch, 1826. 3. Eble JN, et al. Pathology and Genetics of Tumors of the Urinary System and Male Genital Organs. Lyon, France: IARC Press, 2004. 4. Renshaw AA. Subclassification of renal cell neoplasms: an update for the practising pathologist. Histopathology 2002; 41(4): 283 – 300. 5. Amin MB, et al. Prognostic impact of histologic subtyping of adult renal epithelial neoplasms: an experience of 405 cases. Am J Surg Pathol 2002; 26(3): 281 – 91. 6. Renshaw AA, Granter SR, Cibas ES. Fine-needle aspiration of the adult kidney. Cancer 1997; 81(2): 71 – 88.
UNCOMMON TUMORS OF THE KIDNEY 7. Delahunt B, Eble JN. History of the development of the classification of renal cell neoplasia. Clin Lab Med 2005; 25(2): 231 – 46. 8. Latif F, et al. Identification of the von Hippel-Lindau disease tumor suppressor gene. Science 1993; 260(5112): 1317 – 20. 9. Cohen D, Zhou M. Molecular genetics of familial renal cell carcinoma syndromes. Clin Lab Med 2005; 25(2): 259 – 77. 10. Jones TD, Eble JN, Cheng L. Application of molecular diagnostic techniques to renal epithelial neoplasms. Clin Lab Med 2005; 25(2): 279 – 303. 11. Kovacs G. The value of molecular genetic analysis in the diagnosis and prognosis of renal cell tumours. World J Urol 1994; 12(2): 64 – 8. 12. Zambrano NR, et al. Histopathology and molecular genetics of renal tumors toward unification of a classification system. J Urol 1999; 162(4): 1246 – 58. 13. Argani P, et al. Primary renal neoplasms with the ASPL-TFE3 gene fusion of alveolar soft part sarcoma: a distinctive tumor entity previously included among renal cell carcinomas of children and adolescents. Am J Pathol 2001; 159(1): 179 – 92. 14. Zhou M, Roma A, Magi-Galluzzi C. The usefulness of immunohistochemical markers in the differential diagnosis of renal neoplasms. Clin Lab Med 2005; 25(2): 247 – 57. 15. McGregor DK, et al. Diagnosing primary and metastatic renal cell carcinoma: the use of the monoclonal antibody ‘Renal Cell Carcinoma Marker’. Am J Surg Pathol 2001; 25(12): 1485 – 92. 16. Avery AK, et al. Use of antibodies to RCC and CD10 in the differential diagnosis of renal neoplasms. Am J Surg Pathol 2000; 24(2): 203 – 10. 17. Kim MK, Kim S. Immunohistochemical profile of common epithelial neoplasms arising in the kidney. Appl Immunohistochem Mol Morphol 2002; 10(4): 332 – 8. 18. Bell ET. Renal Diseases. Philadelphia, Pennsylvania: Lee & Febiger, 1950. 19. Murphy GP, Mostofi FK. Histologic assessment and clinical prognosis of renal adenoma. J Urol 1970; 103(1): 31 – 6. 20. Eble JN. Tumors of the kidney. Semin Diagn Pathol 1998; 15: 1 – 81. 21. Kovacs G, et al. The Heidelberg classification of renal cell tumours. J Pathol 1997; 183(2): 131 – 3. 22. Kipell J. The incidence of benign renal nodules (a clinicopathologic study). J Urol 1971; 106: 503. 23. Mancilla-Jimenez R, Stanley RJ, Blath RA. Papillary renal cell carcinoma: a clinical, radiologic, and pathologic study of 34 cases. Cancer 1976; 38(6): 2469 – 80. 24. Fleming S, Lewi HJ. Collecting duct carcinoma of the kidney. Histopathology 1986; 10(11): 1131 – 41. 25. Srigley JR, Eble JN. Collecting duct carcinoma of kidney. Semin Diagn Pathol 1998; 15(1): 54 – 67. 26. Dimopoulos MA, et al. Collecting duct carcinoma of the kidney. Br J Urol 1993; 71(4): 388 – 91. 27. Mejean A, et al. Is there a place for radical nephrectomy in the presence of metastatic collecting duct (Bellini) carcinoma? J Urol 2003; 169(4): 1287 – 90. 28. Milowsky MI, et al. Active chemotherapy for collecting duct carcinoma of the kidney: a case report and review of the literature. Cancer 2002; 94(1): 111 – 6. 29. Peyromaure M, et al. Collecting duct carcinoma of the kidney: a clinicopathological study of 9 cases. J Urol 2003; 170(4): Pt 1 1138 – 40. 30. Chao D, et al. Collecting duct renal cell carcinoma: clinical study of a rare tumor. J Urol 2002; 167(1): 71 – 4. 31. Davis CJ Jr, Mostofi FK, Sesterhenn IA. Renal medullary carcinoma. The seventh sickle cell nephropathy. Am J Surg Pathol 1995; 19(1): 1 – 11. 32. Dimashkieh H, Choe J, Mutema G. Renal medullary carcinoma: a report of 2 cases and review of the literature. Arch Pathol Lab Med 2003; 127(3): e135 – 8. 33. Selby DM, et al. Renal medullary carcinoma: can early diagnosis lead to long-term survival? J Urol 2000; 163(4): 1238. 34. Avery RA, et al. Renal medullary carcinoma: clinical and therapeutic aspects of a newly described tumor. Cancer 1996; 78(1): 128 – 32. 35. Swartz MA, et al. Renal medullary carcinoma: clinical, pathologic, immunohistochemical, and genetic analysis with pathogenetic implications. Urology 2002; 60(6): 1083 – 9.
15
36. Assad L, et al. Cytologic features of renal medullary carcinoma. Cancer 2005; 105(1): 28 – 34. 37. Strouse JJ, et al. Significant responses to platinum-based chemotherapy in renal medullary carcinoma. Pediatr Blood Cancer 2005; 44(4): 407 – 11. 38. Argani P, Ladanyi M. Translocation carcinomas of the kidney. Clin Lab Med 2005; 25(2): 363 – 78. 39. Sidhar SK, et al. The t(X;1)(p11.2;q21.2) translocation in papillary renal cell carcinoma fuses a novel gene PRCC to the TFE3 transcription factor gene. Hum Mol Genet 1996; 5(9): 1333 – 8. 40. Hennigar R, Epstein J, Farrow GM. Tubular renal cell carcinomas of collecting duct origin. Mod Pathol 1994; 7: 76A. 41. MacLennan GT, Farrow GM, Bostwick DG. Low-grade collecting duct carcinoma of the kidney: report of 13 cases of low-grade mucinous tubulocystic renal carcinoma of possible collecting duct origin. Urology 1997; 50(5): 679 – 84. 42. Srigley JR. Phenotypic, molecular and ultrastructural studies of a novel low grade renal epithelial neoplasm possibly related to the loop of Henle. Mod Pathol 2002; 15(182A). 43. Amin M. Tubulocystic carcinoma of the kidney. Mod Pathol 2004; 17(137A). 44. Razoky C. Low-grade tubular-mucinous renal neoplasms. Mod Pathol 2002; 15(11): 1162 – 71. 45. Parwani AV, et al. Low-grade myxoid renal epithelial neoplasms with distal nephron differentiation. Hum Pathol 2001; 32(5): 506 – 12. 46. MacLennan GT, Bostwick DG. Tubulocystic carcinoma, mucinous tubular and spindle cell carcinoma, and other recently described rare renal tumors. Clin Lab Med 2005; 25(2): 393 – 416. 47. Klein MJ, Valensi QJ. Proximal tubular adenomas of kidney with socalled oncocytic features. A clinicopathologic study of 13 cases of a rarely reported neoplasm. Cancer 1976; 38(2): 906 – 14. 48. Lieber MM. Renal oncocytoma: prognosis and treatment. Eur Urol 1990; 18(Suppl 2): 17 – 21. 49. Huben RPaV. R.C. Textbook of Uncommon Cancer, 2nd ed: John Wiley & Sons, 1999. 50. Abrahams NA, Tamboli P. Oncocytic renal neoplasms: diagnostic considerations. Clin Lab Med 2005; 25(2): 317 – 39. 51. Fleming S, O’Donnell M. Surgical pathology of renal epithelial neoplasms: recent advances and current status. Histopathology 2000; 36(3): 195 – 202. 52. Amin MB, et al. Renal oncocytoma: a reappraisal of morphologic features with clinicopathologic findings in 80 cases. Am J Surg Pathol 1997; 21(1): 1 – 12. 53. Liu J, Fanning CV. Can renal oncocytomas be distinguished from renal cell carcinoma on fine-needle aspiration specimens? A study of conventional smears in conjunction with ancillary studies. Cancer 2001; 93(6): 390 – 7. 54. Tickoo SK, et al. Ultrastructural observations on mitochondria and microvesicles in renal oncocytoma, chromophobe renal cell carcinoma, and eosinophilic variant of conventional (clear cell) renal cell carcinoma. Am J Surg Pathol 2000; 24(9): 1247 – 56. 55. Tickoo SK, et al. Renal oncocytosis: a morphologic study of fourteen cases. Am J Surg Pathol 1999; 23(9): 1094 – 101. 56. Zbar B, et al. Risk of renal and colonic neoplasms and spontaneous pneumothorax in the Birt-Hogg-Dube syndrome. Cancer Epidemiol Biomarkers Prev 2002; 11(4): 393 – 400. 57. Pavlovich CP, et al. Evaluation and management of renal tumors in the Birt-Hogg-Dube syndrome. J Urol 2005; 173(5): 1482 – 6. 58. Hilton S. Imaging of renal cell carcinoma. Semin Oncol 2000; 27(2): 150 – 9. 59. Perez-Ordonez B, et al. Renal oncocytoma: a clinicopathologic study of 70 cases. Am J Surg Pathol 1997; 21(8): 871 – 83. 60. Dechet CB, et al. Renal oncocytoma: multifocality, bilateralism, metachronous tumor development and coexistent renal cell carcinoma. J Urol 1999; 162(1): 40 – 2. 61. Grawitz P. Demonstration eines grossen Angio-Myo-Lipoms der Niere. Dtsch Med Wochenschr 1900; 26: 290. 62. Bissler JJ, Kingswood JC. Renal angiomyolipomata. Kidney Int 2004; 66(3): 924 – 34. 63. Bourneville D-MB. A l’idiotie et epilepsie symptomatique de sclerose tubereuse ou hypertrophique. Arch Neurol 1900; 10: 29 – 39.
16
GENITOURINARY CANCER
64. Tallarigo C, et al. Diagnostic and therapeutic problems in multicentric renal angiomyolipoma. J Urol 1992; 148(6): 1880 – 4. 65. Abdulla M, et al. Renal angiomyolipoma. DNA content and immunohistochemical study of classic and multicentric variants. Arch Pathol Lab Med 1994; 118(7): 735 – 9. 66. Ashfaq R, Weinberg AG, Albores-Saavedra J. Renal angiomyolipomas and HMB-45 reactivity. Cancer 1993; 71(10): 3091 – 7. 67. L’Hostis H, et al. Renal angiomyolipoma: a clinicopathologic, immunohistochemical, and follow-up study of 46 cases. Am J Surg Pathol 1999; 23(9): 1011 – 20. 68. Pea M, et al. Melanocyte-marker-HMB-45 is regularly expressed in angiomyolipoma of the kidney. Pathology 1991; 23(3): 185 – 8. 69. Lowe BA, et al. Malignant transformation of angiomyolipoma. J Urol 1992; 147(5): 1356 – 8. 70. Martignoni G, et al. Renal angiomyolipoma with epithelioid sarcomatous transformation and metastases: demonstration of the same genetic defects in the primary and metastatic lesions. Am J Surg Pathol 2000; 24(6): 889 – 94. 71. Mai KT, Perkins DG, Collins JP. Epithelioid cell variant of renal angiomyolipoma. Histopathology 1996; 28(3): 277 – 80. 72. Eble JN, Amin MB, Young RH. Epithelioid angiomyolipoma of the kidney: a report of five cases with a prominent and diagnostically confusing epithelioid smooth muscle component. Am J Surg Pathol 1997; 21(10): 1123 – 30. 73. Hajdu SI, Foote FW Jr. Angiomyolipoma of the kidney: report of 27 cases and review of the literature. J Urol 1969; 102(4): 396 – 401. 74. Fujii Y, et al. Benign renal tumors detected among healthy adults by abdominal ultrasonography. Eur Urol 1995; 27(2): 124 – 7. 75. Vogt H. Zur diagnostik der tuberosen sklerose. Z Erforsch Behandl Jugendl Schwachsinns 1908; 2: 1 – 12. 76. Nelson CP, Sanda MG. Contemporary diagnosis and management of renal angiomyolipoma. J Urol 2002; 168(4Pt 1): 1315 – 25. 77. Steiner MS, et al. The natural history of renal angiomyolipoma. J Urol 1993; 150(6): 1782 – 6. 78. De Luca S, Terrone C, Rossetti SR. Management of renal angiomyolipoma: a report of 53 cases. BJU Int 1999; 83(3): 215 – 8. 79. Chesa Ponce N, et al. Wunderlich’s syndrome as the first manifestation of a renal angiomyolipoma. Arch Esp Urol 1995; 48(3): 305 – 8. 80. Dickinson M, et al. Renal angiomyolipoma: optimal treatment based on size and symptoms. Clin Nephrol 1998; 49(5): 281 – 6. 81. Eble JN. Angiomyolipoma of kidney. Semin Diagn Pathol 1998; 15(1): 21 – 40. 82. Clarke A, et al. End-stage renal failure in adults with the tuberous sclerosis complex. Nephrol Dial Transplant 1999; 14(4): 988 – 91. 83. Oesterling JE, et al. The management of renal angiomyolipoma. J Urol 1986; 135(6): 1121 – 4. 84. Fazeli-Matin S, Novick AC. Nephron-sparing surgery for renal angiomyolipoma. Urology 1998; 52(4): 577 – 83. 85. Krishnan BT, et al. Horseshoe kidney is associated with an increased relative risk of primary renal carcinoid tumor. Clin Urol 1997; 157: 2059 – 66. 86. Begin LR, et al. Renal carcinoid and horseshoe kidney: a frequent association of two rare entities – a case report and review of the literature. J Surg Oncol 1998; 68(2): 113 – 9. 87. Resnick ME, Unterberger H, McLoughlin PT. Renal carcinoid producing the carcinoid syndrome. Med Times 1966; 94(8): 895 – 6. 88. Raslan WF, et al. Primary carcinoid of the kidney. Immunohistochemical and ultrastructural studies of five patients. Cancer 1993; 72(9): 2660 – 6. 89. Moulopoulos A, et al. Primary renal carcinoid: computed tomography, ultrasound, and angiographic findings. J Comput Assist Tomogr 1991; 15(2): 323 – 5. 90. McCaffrey JA, et al. Carcinoid tumor of the kidney. The use of somatostatin receptor scintigraphy in diagnosis and management. Urol Oncol 2000; 5(3): 108 – 11. 91. Lamberts SW, et al. Somatostatin-receptor imaging in the localization of endocrine tumors. N Engl J Med 1990; 323(18): 1246 – 9. 92. Kulke MH, Mayer RJ. Carcinoid tumors. N Engl J Med 1999; 340(11): 858 – 68. 93. Oberg K, Eriksson B. The role of interferons in the management of carcinoid tumors. Acta Oncol 1991; 30(4): 519 – 22.
94. Moertel CG, Hanley JA. Combination chemotherapy trials in metastatic carcinoid tumor and the malignant carcinoid syndrome. Cancer Clin Trials 1979; 2(4): 327 – 34. 95. Stallone G, et al. Primary renal lymphoma does exist: case report and review of the literature. J Nephrol 2000; 13(5): 367 – 72. 96. Da’as N, et al. Kidney involvement and renal manifestations in nonHodgkin’s lymphoma and lymphocytic leukemia: a retrospective study in 700 patients. Eur J Haematol 2001; 67(3): 158 – 64. 97. Urban BA, Fishman EK. Renal lymphoma: CT patterns with emphasis on helical CT. Radiographics 2000; 20(1): 197 – 212. 98. Yasunaga Y, et al. Malignant lymphoma of the kidney. J Surg Oncol 1997; 64(3): 207 – 11. 99. Dimopoulos MA, et al. Primary renal lymphoma: a clinical and radiological study. J Urol 1996; 155(6): 1865 – 7. 100. Okuno SH, et al. Primary renal non-Hodgkin’s lymphoma. An unusual extranodal site. Cancer 1995; 75(9): 2258 – 61. 101. Farrow GM, et al. Sarcomas and sarcomatoid and mixed malignant tumors of the kidney in adults. I. Cancer 1968; 22(3): 545 – 50. 102. Srinivas V, et al. Sarcomas of the kidney. J Urol 1984; 132(1): 13 – 6. 103. Vogelzang NJ, et al. Primary renal sarcoma in adults. A natural history and management study by the American Cancer Society, Illinois Division. Cancer 1993; 71(3): 804 – 10. 104. Tomera KM, Farrow GM, Lieber MM. Sarcomatoid renal carcinoma. J Urol 1983; 130(4): 657 – 9. 105. Grignon DJ, et al. Primary sarcomas of the kidney. A clinicopathologic and DNA flow cytometric study of 17 cases. Cancer 1990; 65(7): 1611 – 8. 106. Argani P, et al. Primary renal synovial sarcoma: molecular and morphologic delineation of an entity previously included among embryonal sarcomas of the kidney. Am J Surg Pathol 2000; 24(8): 1087 – 96. 107. Ladanyi M, et al. Impact of SYT-SSX fusion type on the clinical behavior of synovial sarcoma: a multi-institutional retrospective study of 243 patients. Cancer Res 2002; 62(1): 135 – 40. 108. Spellman JE Jr, Driscoll DL, Huben RP. Primary renal sarcoma. Am Surg 1995; 61(5): 456 – 9. 109. Karakousis CP, et al. Surgery for disseminated abdominal sarcoma. Am J Surg 1992; 163(6): 560 – 4. 110. Pages A, Granier M. Nephronogenic nephroma (author’s transl). Arch Anat Cytol Pathol 1980; 28(2): 99 – 103. 111. Argani P. Metanephric neoplasms: the hyperdifferentiated, benign end of the Wilms tumor spectrum? Clin Lab Med 2005; 25(2): 379 – 92. 112. Davis CJ Jr, et al. Metanephric adenoma. Clinicopathological study of fifty patients. Am J Surg Pathol 1995; 19(10): 1101 – 14. 113. Jones EC, et al. Metanephric adenoma of the kidney. A clinicopathological, immunohistochemical, flow cytometric, cytogenetic, and electron microscopic study of seven cases. Am J Surg Pathol 1995; 19(6): 615 – 26. 114. Grignon DJ, Eble JN. Papillary and metanephric adenomas of the kidney. Semin Diagn Pathol 1998; 15(1): 41 – 53. 115. Hennigar RA, Beckwith JB. Nephrogenic adenofibroma. A novel kidney tumor of young people. Am J Surg Pathol 1992; 16(4): 325 – 34. 116. Arroyo MR, et al. The spectrum of metanephric adenofibroma and related lesions: clinicopathologic study of 25 cases from the National Wilms Tumor Study Group Pathology Center. Am J Surg Pathol 2001; 25(4): 433 – 44. 117. Breslow N, et al. Age distribution of Wilms’ tumor: report from the National Wilms’ Tumor Study. Cancer Res 1988; 48(6): 1653 – 7. 118. Petruzzi MaG D. Adult Wilms tumor. In Raghavan D (ed) Textbook of Uncommon Cancer, 2nd ed: John Wiley & Sons, 1999. 119. Bozeman G, et al. Adult Wilms’ tumor: prognostic and management considerations. Urology 1995; 45(6): 1055 – 8. 120. Huser J, et al. Adult Wilms’ tumor: a clinicopathologic study of 11 cases. Mod Pathol 1990; 3(3): 321 – 6. 121. Khoury JD. Nephroblastic neoplasms. Clin Lab Med 2005; 25(2): 341 – 61. 122. Kilton L, Matthews MJ, Cohen MH. Adult Wilms tumor: a report of prolonged survival and review of literature. J Urol 1980; 124(1): 1 – 5. 123. Byrd RL, Evans AE, D’Angio GJ. Adult Wilms tumor: effect of combined therapy on survival. J Urol 1982; 127(4): 648 – 51.
UNCOMMON TUMORS OF THE KIDNEY 124. Arrigo S, et al. Better survival after combined modality care for adults with Wilms’ tumor. A report from the National Wilms’ Tumor Study. Cancer 1990; 66(5): 827 – 30. 125. Kattan J, et al. Adult Wilms’ tumour: review of 22 cases. Eur J Cancer 1994; 30A(12): 1778 – 82. 126. Reinhard H, et al. Wilms’ tumor in adults: results of the Society of Pediatric Oncology (SIOP) 93-01/Society for Pediatric Oncology and Hematology (GPOH) Study. J Clin Oncol 2004; 22(22): 4500 – 6. 127. Terenziani M, et al. Adult Wilms’ tumor: a monoinstitutional experience and a review of the literature. Cancer 2004; 101(2): 289 – 93.
17
128. Kalapurakal JA, et al. Management of Wilms’ tumour: current practice and future goals. Lancet Oncol 2004; 5(1): 37 – 46. 129. Abu-Ghosh AM, et al. Ifosfamide, carboplatin and etoposide in children with poor-risk relapsed Wilms’ tumor: a Children’s Cancer Group report. Ann Oncol 2002; 13(3): 460 – 9. 130. Pein F, et al. High-dose melphalan, etoposide, and carboplatin followed by autologous stem-cell rescue in pediatric high-risk recurrent Wilms’ tumor: a French Society of Pediatric Oncology study. J Clin Oncol 1998; 16(10): 3295 – 301.
Section 1 : Genitourinary Cancer
2
Uncommon Cancers of the Bladder Arlene O. Siefker-Radtke, Bogdan A. Czerniak, Colin P. Dinney and Randall E. Millikan
INTRODUCTION The uncommonly encountered tumors arising in the bladder fall naturally into two groups: unusual histologies of urothelial origin and nonurothelial malignancies. Our discussion follows these lines, and we take up the unusual urothelial cancers first, as these are by far the most common, and the most clinically important. Most of what is called urothelial carcinoma, about 80% in fact, is of low histologic grade and reflects a primarily hyperplastic process. This process typically results in grossly papillary architecture from the mass of proliferating but noninvasive cells. These lesions, designated “Ta”, tend to recur, and can progress to dysplasia and invasion in 15 to 20% of patients. However, for the most part, these neoplasms are more like polyps than true carcinomas. An overlapping, but distinct, pattern can be recognized in the remaining 20% of cases. In this group, dysplasia, high histologic grade, and invasion are the hallmarks.1 These “nonpapillary” cancers are the cause of most of the mortality, and they demonstrate a significant spectrum of histomorphology. At the extreme, they are so unlike transitional cell carcinoma (TCC) that alternative taxonomy is called for. Although quite rare as “pure” variants, it is nevertheless true that about one-third of nonpapillary bladder cancers will exhibit at least focal areas of one of the unusual histologies discussed below. Thus, as a practical matter, there is significant uncertainty about which of the cases showing minor degrees of variant histology should be considered to be truly outside the usual spectrum of TCC, and which should be classified as distinct entities.
SARCOMATOID CARCINOMA The clinical course of advanced urothelial cancer is generally characterized by more aggressive biological behavior with time. This is typically accompanied by phenotypic evolution to patterns that are readily recognized to reflect more aggressive biological behavior. In this context, the recognition of areas with spindled histomorphology is fairly common.
When this pattern becomes dominant, the cancers are typically described as sarcomatoid carcinomas. Such cases have been long recognized, and many case series and reviews are available.2 – 4 There continue to be several sporadic case reports each year. The typical histologic appearance is shown in Figure 1. Almost always, nonspindled areas recognizable as highgrade TCC are also present, suggesting that this pattern results from evolution from a common progenitor. Indeed, by immunohistochemistry, the spindled areas are generally positive for keratin, epithelial membrane antigen, and vimentin. (When epithelial markers are lost in a significant fraction of tumor cells, the term carcinosarcoma is appropriate, but the literature makes no systematic distinction between these two terms.) Analysis of clonality based on loss of heterozygosity (LOH) of microsatellite markers5,6 provides strong evidence that although the various components can and do evolve independently once they diverge, they do, in fact, arise from a common precursor. While it is clear that we can define a subset with a sarcomatoid appearance, it is more important to know what biologic significance such morphology portends. Although there do not seem to be particular risk factors, or clinically distinctive features at initial presentation, a recurring theme in the clinical experience with sarcomatoid urothelial cancer is that it has an aggressive natural history and poor outcome relative to that of classical TCC. This is borne out in the MD Anderson registry, both in the locally advanced and metastatic settings. In view of this, we consider the presence of a sarcomatoid component in an otherwise minimally invasive bladder cancer to be a strong indication for early cystectomy. We know of no data to recommend any particular therapeutic approach to systemic therapy. Even though we have investigated more intensive chemotherapy in this subset, we do not have any sense that this is justified by improved outcome, and do not endorse this approach in the absence of a clinical trial. It is very important to realize that not everything that appears spindled is dangerous.7 In particular, the postresection sarcomatous nodule and inflammatory pseudotumor must not be confused for the aggressive cancers to which
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
UNCOMMON CANCERS OF THE BLADDER
(a)
(b)
(c)
(d)
19
Figure 1 Microscopic features of high-grade urothelial carcinomas with sarcomatoid and small cell phenotype. (a and b) Low and high-power views of sarcomatoid urothelial carcinoma composed of atypical spindle cells. (c and d) Low and high-power views of small cell variant of urothelial carcinoma. Note poorly differentiated small cells with inconspicuous cytoplasm and densely packed, hyperchromatic nuclei.
they bear some resemblance. Besides the clinical context, it is reported that the presence of necrosis at the muscle interface and nuclear atypia are the most useful features that distinguish true sarcomas or sarcomatoid carcinomas from these benign conditions. Although they are extremely uncommon, true sarcomas without an epithelial component do arise in the bladder, and they are taken up in the section “Sarcoma”.
SMALL CELL CARCINOMA Like spindled morphology, histologic features reminiscent of neuroendocrine carcinomas are commonly encountered in patients with high-grade TCC, especially as it evolves
over time. At present, there is no consensus about the exact diagnostic criteria for declaring a “small cell” or “neuroendocrine” subset. Some authorities place more emphasis on the histomorphology and some on the expression of neuroendocrine markers. At MD Anderson Cancer Center, it has been our understanding that the morphology is more predictive of clinical course than any particular pattern of immunohistochemical markers, and thus we continue to prefer the term “small cell carcinoma” and do not require immunohistochemical “confirmation” to apply the term. The typical appearance is shown in Figure 1. As might be inferred from clinical observation, LOH studies8 of coexisting small cell and TCC suggest descent
GENITOURINARY CANCER
from a common precursor, just as has been shown for sarcomatoid carcinoma. In part because of the lack of well-defined diagnostic criteria, the incidence of small cell carcinoma is very difficult to estimate. In the MD Anderson registry (which obviously reflects referral bias) we find small cell carcinoma has an incidence of 3% (115 of 3833). Other investigators have suggested 0.5 to 0.7%. One occasionally encounters patients with small cell carcinoma found in the prostate and it is not clear if this should be interpreted as being urothelial or prostatic in origin. Generally, there will either be urothelial dysplasia in the prostatic urethra (favoring a urothelial origin), or signs of dysplasia in the prostatic acini (favoring a prostatic origin). In truth, the distinction matters little, as the foundation of therapy for either primary site is early exposure to systemic chemotherapy (e.g., etoposide and cisplatin), followed by local consolidation (which would be radical surgery at our center). As is well established for other sites, small cell carcinoma of the urothelium is an aggressive cancer, characterized by the early development of micrometastases, a very high incidence of liver involvement, and significantly, for the brain to be a “sanctuary” site of metastasis. Rapid growth within the bladder is also typical. Indeed, it is not uncommon to see patients with very large tumors that had been “completely resected” just 2 to 3 weeks previously. This may underlie the relatively poor results (detailed below) with surgical management of this variant. Once a small cell component has been identified, it is important to liberalize staging studies just a bit. Even for patients with minimal invasion within the bladder, the possibility of metastases is significant, and thus all patients should have a computed tomography (CT) of the abdomen and pelvis to look for nodal and hepatic involvement. In addition, it is quite important to remain vigilant for development of brain metastases at any point in the clinical evolution of the disease. As more effective systemic therapy becomes available, the issue of prophylactic cranial irradiation (as is done for some patients with small cell carcinoma of the lung) will likely need to be studied. We have recently seen two patients treated with neoadjuvant chemotherapy and surgery who subsequently succumbed to brain metastases without any evidence of other sites of involvement. It has long been recognized that cystectomy in this context is associated with a much lower cure rate than is obtained with conventional TCC. Clinical understaging has been the rule, with up to 76% of patients with small cell tumors having metastases at cystectomy.9 In a recent review of the MD Anderson Cancer Center experience10 we found that the pathologic stage was higher than expected in 56% of patients treated with initial cystectomy. Furthermore, 20% of patients for whom initial cystectomy was planned were found to be surgically unresectable in the operating room, despite a median time of 24 days from diagnosis to surgery. In view of this experience, many institutions have reported multimodality approaches incorporating systemic therapy with radiation and/or surgery. In a literature review from 1995, Abbas reported that the best disease-free survival
was observed when cystectomy was followed by adjuvant chemotherapy, with a median survival of 21.1 months.11 Of the 19 patients reported by Grignon,12 four of the five survivors were treated with adjuvant chemotherapy following cystectomy. Likewise, the University of Southern California-Norris Cancer Center reported an improved overall and recurrence-free survival among patients receiving multimodality therapy, the majority of whom received adjuvant chemotherapy.9 However, the median survival in these patients was only 13 months, with a 5-year survival of 10%. Initial chemotherapy (i.e. neoadjuvant systemic therapy) has also been investigated, and with somewhat more promising results. Walther13 reported that five of seven patients were alive and free of disease at more than 36 months after combined modality treatment, and, in fact, the five who were alive were treated with preoperative chemotherapy. A retrospective analysis of our own experience10 in 46 patients, who were both candidates for surgery and had surgically resectable disease, found a statistically improved survival for those treated with preoperative chemotherapy. Patients treated with initial cystectomy had median cancer-specific survival (CSS) of 23 months, and a 36% CSS at 5 years (see Figure 2). By contrast, for those treated with preoperative chemotherapy, the median CSS has not been reached, and the 5-year CSS was 78%. There were only four cancer-related deaths among the 21 patients treated with initial chemotherapy, and none in the subset “downstaged” to pT2N0M0 or less. It is interesting to note that 7 of 25 patients treated with initial cystectomy received adjuvant chemotherapy; however, their survival was no better than those treated with cystectomy alone. Given these results, we take the view that all patients with a small cell component should be treated primarily (and without delay) with three or four cycles of chemotherapy incorporating etoposide and cisplatin. Following chemotherapy, we favor surgical consolidation. This is based on the observations that most of these patients have widespread dysplasia and often show glandular elements after chemotherapy. Thus, we infer that a significant fraction of patients will not have
1 Cause-specific survival
20
0.8 0.6 0.4 0.2
Bottom curve Adjuvant or no chemotherapy Top curve Neoadjuvant chemotherapy
0 0
2
4
6
8
10
12
14
16
Years from diagnosis of small cell carcinoma Figure 2 Cancer-specific survival of patients with resectable small cell carcinoma. Patients treated with neoadjuvant chemotherapy followed by cystectomy (top curve) had a better outcome (p = 0.03) than those treated with initial surgery (bottom curve).
UNCOMMON CANCERS OF THE BLADDER
a good long-term control in the bladder by means of radiotherapy. Limited data from radiotherapy case series suggest this is, in fact, the case.14,15 Unfortunately, given the rapid progression and early metastatic potential of small cell bladder cancer, many patients have metastatic or surgically unresectable disease at presentation. The most frequently reported sites are the lymph nodes, liver, bone, lung, and the brain.10,16 With chemotherapy, median survival ranges from 7.5 months to 15 months. There are few long-term survivors, but approximately 20% will be alive at 3 years, which is more than 2 years longer than the untreated natural history of the disease. Despite the disappointing long-term outcome, small cell urothelial tumors are highly responsive to systemic chemotherapy, and most patients experience marked objective response that is associated with gratifying (if temporary) palliation of symptoms.
MICROPAPILLARY BLADDER CANCER The histomorphologic spectrum of many epithelial cancers is now recognized to include a subset characterized by clusters of high-grade cells nesting in lacunar spaces that have a “micropapillary” architecture. This pattern was first reported in 1982 by Henderickson et al.17 who described an aggressive variant of endometrial adenocarcinoma. In that context, an infiltrative, biologically aggressive pattern of spread, strikingly reminiscent of the behavior of papillary serous carcinoma of the ovary was apparent. Subsequently, a micropapillary variant has been recognized in cancers arising from the breast, bladder, thyroid, lung, and pancreas, and has thus come to be seen as a general feature of epithelial cancer. It seems likely that this phenotype derives from some fundamental aspect of epithelial carcinogenesis, and the finding of a recognizable “gene signature” that cuts across these different sites would not be surprising. Investigators from MD Anderson Cancer Center were the first to report a subset of bladder cancers showing such a micropapillary histomorphology, publishing the initial series of 18 patients in 1994.18 As discussed in the context of sarcomatoid and small cell morphology, it is typical to appreciate a micropapillary component in a context that includes more typical TCC. From this initial series of patients, the micropapillary variant was noted to have a more aggressive clinical course and a particular tendency for prominent lymphovascular invasion. Subsequently, case series were published from Sweden,19 Harvard,20 Australia,21 Wayne State University,22 and Mexico City,23 along with many individual case reports. These reports established that micropapillary bladder cancer does have a characteristic morphology (see Figure 3) that can be reliably identified. Furthermore, the initially reported propensity for clinical understaging, aggressive behavior, and relatively poor response to standard systemic chemotherapy has been fully confirmed. Ultrastructural studies in the setting of micropapillary breast cancer suggested secretory granules along the basal membrane, i.e. loss of normal cell polarity in which secretory activity is found at the basal surface, not just at the apical surface.24 This concept of “inside out” morphology has
21
been reinforced by the demonstration that the mucinous glycoprotein product of the MUC1 gene is also abnormally localized to the basal surface of micropapillary cancers,25 including those of bladder origin. Although not yet formally established, it seems likely that a mechanistic connection will be made between this sort of abnormal phenotype, the early submucosal infiltration, and early access to lymphatics that characterize the clinical course of these cancers. In our experience, the unusual finding of a bladder cancer that is stage pT1N1 is nearly always associated with micropapillary histology. Micropapillary bladder cancer has been reported to occur infrequently. In the case series from Sweden,19 based on a population-based registry, the incidence of bladder cancer cases was 0.7%. The referral-biased report from Mexico City,23 put the incidence at 38/630 (6%). In a similarly biased registry at MD Anderson Cancer Center, we find 162 cases, for an incidence of 4.2%. Of course these reports come from very different genetic and environmental contexts, and it is certainly possible that the incidence various according to geographic region. The clinical management of micropapillary bladder cancer should take into account the very real possibility that it will be clinically understaged and can grow rapidly. Thus, we advocate “early” cystectomy for any tumor that invades the lamina propria, and would certainly urge that any patient with cT1 disease or greater after a trial of intravesical therapy be guided to cystectomy without delay. For patients with locally advanced disease, the prognosis is worse than for patients with conventional TCC. In the MD Anderson Cancer Center experience, patients with cT3b or cT4a disease treated with a combination of systemic chemotherapy and surgery (in any sequence) had an inferior overall cure rate compared to patients with conventional TCC. In the metastatic setting, both the response rate and the overall survival of patients with micropapillary cancer are below our historical expectation. Despite aggressive application of combination chemotherapy, patients with micropapillary cancers continue to have a relatively poor outcome. This has been true with methotrexate/vinblastine/doxorubicin/cisplatin (MVAC), with ifosfamide-based combinations, and gemcitabine/cisplatin. It is important to recognize, however, that although the aggregate results are inferior to those obtained for patients with conventional TCC, there are many patients who do well, and therefore we continue to offer conventional therapy to patients in this subset.
ADENOCARCINOMA Glandular metaplasia is fairly common within the urothelium, and the appearance of cystitis glandularis and cystitis cystica are familiar consequences of chronic infection, inflammation, or irritation (as with urolithiasis). In the transformed state, focal areas of glandular differentiation are reasonably common in invasive, nonpapillary TCC. To our knowledge, such focal areas of glandular differentiation do not have any clinical significance with respect to the natural history or response to commonly used systemic agents. Furthermore, it is not unusual to see a remnant of glandular differentiation in a cystectomy specimen following neoadjuvant chemotherapy for
22
GENITOURINARY CANCER
(a)
(b)
(c)
(d)
Figure 3 Microscopic features of rare forms of urothelial carcinoma. (a) Micropapillary variant of urothelial carcinoma composed of small nests of atypical epithelial cells arranged in distinct lacunar spaces. (b) Plasmacytoid and signet ring cell variant of urothelial carcinoma. Note loosely arranged oval cancer cells with distinct eosinophilic cytoplasm and eccentric nuclei. Some of the cells have a large mucin-containing cytoplasmic vacuole displacing the nucleus. (c) Mucinous adenocarcinoma composed of nests of cancer cells in lakes of extracellular mucin. (d) Colonic variant of bladder adenocarcinoma composed of large glandular structures with stratified nuclei.
a high-grade, nonpapillary TCC, even if no adenocarcinomatous element was appreciated prior to therapy. These relatively common clinical scenarios highlight the morphologic repertoire of urothelial histology, and illustrate the difficulty of defining precisely what is meant by “adenocarcinoma of the bladder”. In this section we confine our discussion to cancers that exhibit an adenocarcinoma histology as the dominant pattern, and without recognizable TCC. There are many variations of adenocarcinoma encountered within the bladder (see Figure 3). Most authors (including the World Health Organization classification26 of bladder tumors) include mucinous, signet ring, enteric type, hepatoid,
and clear cell (formerly “mesonephric”) as recognizable subsets. In addition, an adenoid cystic pattern can also be seen, particularly in the context of transformation of a preexisting cystitis cystica. Tumors showing mixtures of these patterns are the rule. Enteric-type histology is especially encountered among cancers arising in a urachal remnant, and these are taken up separately in the section “Urachal Cancer” below. Nonetheless, villous adenoma27 and enterictype adenocarcinoma do rarely occur in the bladder proper. The immunophenotype of these cancers tends to overlap that seen in colon cancer,28 and most produce carcinoembryonic antigen (CEA). The diagnostic dilemma for the pathologist
UNCOMMON CANCERS OF THE BLADDER
is evident, since it could be important to distinguish a colon primary involving the bladder from a primary bladder tumor, and at present, this can only be done by excluding a colon primary by traditional clinical means. Most adenocarcinomas arising in the bladder proper are of the mucinous or signet ring variety. The male-to-female ratio is at least 2 : 1, and the age at onset is very similar to that seen in conventional TCC. Bladder exstrophy, a rare developmental anomaly affecting 1 in 50 000 births, is a well established risk factor,29 as are other nonphysiologic states such as a urinary diversion leaving the bladder in place. Intestinal metaplasia long-antedating the appearance of carcinoma is typical in these contexts. Adenocarcinomas also arise in the context of preexisting cystitis cystica (and glandularis), and sometimes in association with schistosomiasis (although squamous histology is more common in the latter context). It is typical for mucinous adenocarcinomas to be diffusely infiltrative, and to present with irritative symptoms out of proportion to the cystoscopic findings. Cross-sectional imaging typically shows diffusely thickened vesical walls, and frank linitis plastica is well known, especially in the subset with predominantly signet ring histology. We have encountered patients not only with linitis plastica, but also with involvement of the seminal vesicles, and even extension down both spermatic cords. Thus, while the cystoscopic appearance may be unimpressive, the examination under anesthesia is typically striking. The clear cell variant of bladder adenocarcinoma is quite rare, and is also clinically distinct. These cancers usually arise in females (at least 2 : 1 female predominance), and the median age is younger. More than half of the reported cases appear to arise from the urethra or periurethral glands. They typically express CA125, and there are other lines of evidence supporting an etiology from M¨ullerian rests.30 These cancers tend to be quite responsive to taxane-based therapy, such as used for epithelial ovarian cancer. The optimal clinical management of adenocarcinoma is of course not settled. The stage is the most important prognostic factor, and unfortunately most patients with adenocarcinoma present with locally advanced disease (cT3 or higher). Therefore, for many patients, neoadjuvant systemic therapy is a reasonable consideration, since the historical outcome for surgery alone in this setting is poor. Most patients at our center are, in fact, treated with combination chemotherapy followed by radical surgery. We have seen responses with a variety of chemotherapy regimens. In the metastatic setting, we find many examples of patients showing excellent response to chemotherapy, including that to standard TCC therapy. Nonetheless, the overall response rate is lower than is seen with conventional TCC, and survival exceeding 2 years is uncommon.
SQUAMOUS CARCINOMA Predominantly squamous carcinoma of the bladder is most commonly encountered in the setting of schistosomiasis in the Middle East, a context that is outside the scope of this chapter.
23
In the West, focal areas of squamous differentiation are commonly encountered among patients with invasive, nonpapillary TCC. As is the case for focal glandular differentiation, we know of no clinical significance to this finding. By contrast, pure squamous carcinomas are uncommon, and they show a very distinctive clinical expression. The most common setting for (non–schistosomiasis-related) squamous cancer is that of chronic irritation, typically either from urolithiasis (especially stag horn calculi) or from chronic indwelling catheters among patients with paraplegia, or neurogenic bladders from diseases such as multiple sclerosis. It is typical to see keratinizing squamous metaplasia, often with dysplasia, in areas adjacent to these cancers. Surgery is the mainstay of therapy for squamous cancers. Local control is often a bigger problem than distant progression, in rather marked contrast to the situation with conventional TCC. Sensitivity to chemotherapy is universally reported to be lower for squamous cancers than for conventional TCC, which further reinforces the importance of primary surgical management. Unfortunately, when these cancers are recurrent or metastatic, the expectations of chemotherapy are limited. However, there are certainly patients who enjoy excellent responses, and therefore it is difficult to assess the risk-to-benefit ratio of a trial of chemotherapy. In our limited experience, we find the combination of gemcitabine (900 mg m−2 over 90 minutes), cisplatin (50 mg m−2 ), and ifosfamide (1000 mg m−2 ) given every 14 days to be the most attractive regimen in the absence of a clinical trial. Interestingly, pulmonary metastases from squamous carcinoma of the bladder tend to cavitate, behavior not typical of other histologies. Our experience in trying to deliver chemotherapy in the special context of patients with paraplegia (or other diseases causing neurogenic bladder) is uniformly unsatisfying, and we cannot endorse use of conventional regimens such as MVAC in this setting.
LYMPHOEPITHELIOMA The descriptive term lymphoepithelioma was originally applied to a distinctive tumor occurring in the nasopharynx composed of a poorly differentiated epithelial component showing syncytial features, and a second component consisting of a prominent lymphoid stromal infiltration. In the nasopharynx, these cancers were found to be associated with Epstein-Barr virus, but no such association has been found when this pattern is seen in the bladder. These cancers were also found to be remarkably sensitive to radiotherapy. More recently, morphologically similar cancers have been reported from a great many sites, including the thyroid, skin, cervix, lung, and the gastrointestinal tract. The first report of this variant histology within the bladder came from Harvard investigators in 1991.31 Subsequently, case series from MD Anderson Cancer Center,32 Sweden,33 Spain,34 and Greece35 have appeared, but the total number of published cases is still far less than 100. There do not seem to be any particularly distinctive features of the clinical presentation of patients with lymphoepithelioma. The clinical importance of recognizing this subset
24
GENITOURINARY CANCER
is, first, to be aware of the potential for misdiagnosing this entity as an extranodal lymphoma, and second, to recognize the significantly better prognosis enjoyed by those with this pattern as the predominant (or exclusive) histology. In several series, including our own experience, these cancers have been noted to be more chemosensitive and more radiosensitive than conventional TCC. Strikingly however, patients with only “focal” expression of a lymphoepithelioma pattern have a poor prognosis. This clinical observation was first made in the series of 11 patients reported from MD Anderson registry.32 Currently, we find an additional 18 patients in the MD Anderson registry. This additional experience confirms the aggressive nature of this subset (and also the excellent prognosis for those with predominant or exclusive lymphoepithelioma).
URACHAL CANCER It has been estimated that there is one case of urachal carcinoma for every 600 treated bladder cancer patients. The urachus is a vestigial structure, which while important in some species, has no role in the development of humans. The precursor of the urachal ligament initially arises from the cloaca at the terminal end of the hindgut. During embryogenesis, the cloaca divides forming the urogenital sinus, which develops into the bladder and sex organs, and the anorectal canal, which becomes the rectum. The bladder is formed from the medial portion of the urogenital sinus. Superior to this, the lumen of the allantois is obliterated to form the urachus. By adulthood, the urachus coalesces with the obliterated umbilical arteries to form the ligamentum commune. While the urachal ligament most frequently connects with the dome of the bladder, it can also attach to the posterior or anterior bladder wall, usually in the midline.36 A remnant lumen may persist in the wall of the bladder as a tiny tubular or cystic structure, and can communicate with the lumen of the bladder in up to one-third of adults.37,38 Columnar cells, glandular islands, and transitional cell epithelium may be present in such a urachal remnant.39 When malignancy if found to arise in such a remnant, the histology is overwhelming enteric-type adenocarcinoma. Two theories have been proposed for the development of urachal tumors. One is that these adenocarcinomas originate from enteric rests left behind from the cloaca during embryological development. This would explain the histological resemblance to adenocarcinomas of the rectum. An alternate hypothesis is that these tumors arise from metaplasia. Supportive evidence includes the occurrence of adenocarcinomas in exstrophic bladders despite transitional epithelium at birth, the occasional development of other enteric-type tumors in the ureter and renal pelvis that are not of cloacal origin, and the observation of adenocarcinoma arising from cystitis glandularis. Whatever the details of their origin, it is clear that these cancers have a clinical expression that is quite distinct from typical urothelial cancer. No risk factors have been identified, and in particular, smoking and other environmental factors that figure so prominently in typical TCC seem not to be relevant. Patients with urachal cancer are typically much
younger, with a reported median age of 47–57 years at diagnosis, with many cases reported in the third and fourth decades.40,41 In addition, these cancers occur equally in men and women (male-to-female ratio for TCC is about 3 : 1), and they display markedly less susceptibility to cisplatin-based chemotherapy. The majority of urachal tumors display enteric-type histology, resembling adenocarcinomas of the colon and rectum. These tumors often have glandular structures with mucin production; colloid and/or signet ring cell histology may be present. More rarely, sarcomatoid, squamous, and transitional cell histology have been reported.39,40 Remnants of normal or focally ulcerative surface epithelium may overlay the tumor. Normal epithelium overlying the tumor strongly supports the diagnosis of urachal carcinoma. However, the destruction of this layer by tumor can make the distinction between urachal and nonurachal bladder adenocarcinoma difficult. The presence of cystitis cystica or cystitis glandularis transitioning to malignancy favors a diagnosis of an adenocarcinoma of the bladder proper, as opposed to that of urachal origin. Most patients have locally advanced disease at diagnosis, usually presenting with gross hematuria and irritative voiding symptoms, but often presenting with no urinary complaints at all. Patients may report voiding mucoid material, a feature consistent with the typical histology. Erythema and umbilical discharge have also been reported, and we have seen patients initially diagnosed with an “umbilical infection”. The presence of a cystic midline mass with calcifications at the bladder dome on radiographic imaging is nearly pathognomonic. As a practical matter, all patients with enteric-type adenocarcinoma involving the bladder dome should be regarded as having urachal cancer until proven otherwise. Nonetheless, it is important to recognize that these tumors can occur all along the urachal ligament, and may produce a palpable mass anywhere from the umbilicus to the symphysis. While involvement of the bladder is frequently present, it is not a requirement for the diagnosis. The majority of patients present with locally advanced disease, with tumor that invades the bladder wall. The diagnosis is typically made by cystoscopy and biopsy. In addition to the location in the bladder dome and the unusual histology, an important clue to the recognition of a urachal origin is the typical finding of tumor in the muscularis propria with unremarkable urothelium overlying the cancer. By contrast, adenocarcinomas arising from the urothelium grow from “the inside to the outside”, and are frequently associated with the presence of urothelial dysplasia or even focal areas of recognizable transitional cell cancer. The only other important differential consideration in the differential diagnosis is “drop metastasis” from an ovarian or upper gastrointestinal (or pancreatic) primary tumor, although these tend to involve the cul-de-sac and not the bladder dome. Invasion from a urachal primary into the large or small bowel is fairly common, and we have seen several cases of “multifocal colon cancer” or “bladder cancer metastatic to the colon” that turned out to be urachal cancers eroding into the gut at one or more locations. As with colon cancer, evaluation of serum tumor markers may be helpful, especially in the context of evaluating
UNCOMMON CANCERS OF THE BLADDER
response to therapy. We have found CEA, CA125, and CA19-9 to be helpful in some patients. As with other tumor types, elevation of CA125 should raise suspicion for the presence of peritoneal carcinomatosis, which is extremely common in patients with urachal cancers. Appropriate surgical management of urachal carcinoma requires that the diagnosis be made preoperatively, on the basis of recognition of this possibility in the appropriate clinical setting. Cross-sectional imaging is the key to recognizing the diagnosis. On CT, urachal carcinoma often appears as a low attenuation mass at the dome of the bladder, typically in the midline or slightly to one side. Because of the relatively high recurrence rate following the treatment of this disease, en bloc resection of the umbilicus, urachus, overlying peritoneum and posterior rectus fascia lateral to the medial umbilical ligament, bladder, and pelvic lymph nodes is now the standard operation. The recognition that urachal tumors are predominantly extravesical and not associated with a field defect suggested to surgeons that in most cases even bulky tumors could be completely resected with adequate margins by an en bloc dissection with only a partial cystectomy.40,42 Contemporary series report that neither local recurrence nor outcome is threatened by this approach. Rather, survival is more tightly linked to the stage at presentation, the presence of lymph node metastasis, and the ability to achieve a negative surgical margin than with the performance of a partial or total cystectomy. A radical cystectomy is indicated during salvage surgery to treat a positive surgical margin or to remove an inadequately controlled urachal ligament – that occurs when the diagnosis was not made preoperatively. In a series of patients referred to MD Anderson Cancer Center41 it was remarkable that only 19 of 35 patients undergoing primary surgical management had en bloc resection of the urachal ligament and umbilicus. The importance of proper surgical management was reinforced by the finding that, 13 of the 16 long-term survivors reported in this series were treated with en bloc resection including umbilectomy. Unfortunately, patients with nodal or peritoneal involvement discovered at surgery have a median survival of about 25 months, and demonstrate a clinical course that is virtually indistinguishable from that of patients with clinically apparent metastases at diagnosis. In view of this finding, and the demonstrated benefit of perioperative chemotherapy for colorectal cancer, the use of adjuvant or neoadjuvant chemotherapy for urachal cancer would seem to be a reasonable consideration. Unfortunately, there are essentially no data bearing on this point directly, so we are left to extrapolate from our experience with other enteric adenocarcinomas. Since we do have some systemic therapies with clinically relevant response rates (see below) it seems appropriate to discuss adjuvant therapy with patients at particularly high risk of recurrence, including those with tumor in the lymph nodes or on the peritoneal surface, or in the setting of inadequate surgery, including the presence of positive margins and in the setting where the urachal ligament was not controlled. Not surprisingly, few long-term survivors have been observed once metastases develop. The most frequently
25
involved sites include bone, lung, liver, lymph nodes, and the brain. Peritoneal carcinomatosis is common, especially in the setting of positive surgical margins and when peritoneal implants are present at cystectomy. Historically, chemotherapy has had little impact in the treatment of urachal cancer. This is particularly so in the context of chemotherapy regimens traditionally employed for TCC. More recently, responses have been reported in the setting of 5-fluorouracil based chemotherapy regimens.41,43 Currently, we are enrolling patients on a phase II trial using combination chemotherapy with 5-fluorouracil, leucovorin, gemcitabine, and cisplatin. Results in the first 20 patients show objective response in just over one-third of patients. In keeping with the clinical manifestations of this disease being so closely related to colorectal cancer, we have observed anecdotal responses in patients treated with capecitabine and irinotecan based regimens, and to the antiepidermal growth factor antibody cetuximab.
SARCOMA Sarcomas are quite uncommon in the urinary tract. Many patients referred to our institution with tumors diagnosed as sarcomas turn out to have epithelial components on review, and thus are best classified as carcinosarcomas, as described above. Nonetheless, sarcomas that appear to arise within the bladder without a detectable epithelial component do occur. To our knowledge, prior radiotherapy is the only recognized risk factor, and the typical delay of two or three decades from radiation exposure to secondary cancer seems to apply. In adults, the most commonly reported histologic subtype is leiomyosarcoma,44 although in our own registry, osteogenic sarcoma is the most common. Of course, many tumors display areas with more than one pattern of differentiation. Aside from the setting of prior pelvic radiotherapy, there are no characteristic features of the clinical presentation. It is our impression that patients with sarcoma are more susceptible to tumor implants in the urethra after a transurethral resection (TUR), but this has not been formally studied, nor reported by other centers. Clinical management follows principles of sarcoma management in other sites. In general, surgery is the mainstay of therapy. If the primary tumor is quite large, and it shows a histology for which a reasonable response to chemotherapy can be anticipated (such as osteogenic sarcoma), then neoadjuvant chemotherapy is given. As expected, outcome is primarily driven by stage.
REFERENCES 1. Dinney CPN, et al. Focus on bladder cancer. Cancer Cell 2004; 6: 111 – 6. 2. Baschinsky DY, et al. Carcinosarcomas of the urinary bladder – an aggressive tumor with diverse histogenesis. A clinicopathologic study of 4 cases and review of the literature. Arch Pathol Lab Med 2002; 124: 1172 – 8. 3. Lopez-Beltran A, et al. Carcinosarcoma and sarcomatoid carcinoma of the bladder: clinicopathological study of 41 cases. J Urol 1998; 159: 1497 – 503.
26
GENITOURINARY CANCER
4. Perret L, et al. Primary heterologous carcinosarcoma (metaplastic carcinoma) of the urinary bladder: a clinicopathologic, immunohistochemical, and ultrastructural analysis of eight cases and a review of the literature. Cancer 1998; 82: 1535 – 49. 5. Gronau S, et al. Immunohistomorphologic and molecular cytogenetic analysis of a carcinosarcoma of the urinary bladder. Virchows Arch 2002; 440: 436 – 40. 6. Halachmi S, et al. Genetic alterations in urinary bladder carcinosarcoma: evidence of a common clonal origin. Eur Urol 2000; 37: 350 – 7. 7. Iczkowski KA, et al. Inflammatory pseudotumor and sarcoma of urinary bladder: differential diagnosis and outcome in thirty-eight spindle cell neoplasms. Mod Pathol 2001; 14: 1043 – 51. 8. . Cheng L, et al. Molecular genetic evidence for a common clonal origin of urinary bladder small cell carcinoma and coexisting urothelial carcinoma. Am J Pathol 2005; 166: 1533 – 9. 9. Quek ML, et al. Radical cystectomy for primary neuroendocrine tumors of the bladder: the University of Southern California experience. J Urol 2005; 174: 93 – 6. 10. Siefker-Radtke AO, et al. Evidence supporting preoperative chemotherapy for small cell carcinoma of the bladder: a retrospective review of the M. D. Anderson cancer experience. J Urol 2004; 172: 481 – 4. 11. Abbas F, et al. Small cell carcinoma of the bladder and prostate. Urology 1995; 46: 617 – 30. 12. Grignon DJ, et al. Small cell carcinoma of the urinary bladder. A clinicopathologic analysis of 22 cases. Cancer 1992; 69: 527 – 36. 13. Walther PJ. Adjuvant/neo-adjuvant etoposide/cisplatin and cystectomy for management of invasive small cell carcinoma of the bladder. J Urol 2002; 167(suppl): 285, (abstract 1124). 14. Lohrisch C, et al. Small cell carcinoma of the bladder: long term outcome with integrated chemoradiation. Cancer 1999; 86: 2346 – 52. 15. Sejima T, Miyagawa I. “Successful” chemo- and radiotherapy prior to radical cystectomy does not necessarily correlate with clinical course in small cell carcinoma of the bladder. Urol Int 2005; 74: 286 – 8. 16. Sved P, et al. Small cell carcinoma of the bladder. BJU Int 2004; 94: 12 – 7. 17. Henderickson M, et al. Uterine papillary serous carcinoma: a highly malignant form of endometrial adenocarcinoma. Am J Pathol 1982; 6: 93 – 108. 18. Amin MB, et al. Lymphoepithelioma-like carcinoma of the urinary bladder. Am J Surg Pathol 1994; 15: 466 – 73. 19. Johansson SL, Borghede G, Holmang S. Micropapillary bladder carcinoma: a clinicopathological study of 20 cases. J Urol 1999; 161: 1798 – 802. 20. Maranchie JK, et al. Clinical and pathological characteristics of micropapillary transitional cell carcinoma: a highly aggressive variant. J Urol 2000; 163: 748 – 51. 21. Samaratunga H, Khoo K. Micropapillary variant of urothelial carcinoma of the urinary bladder; a clinicopathological and immunohistochemical study. Histopathology 2004; 45: 55 – 64. 22. Nassar H. Carcinomas with micropapillary morphology: clinical significance and current concepts. Adv Anat Pathol 2004; 11: 297 – 303. 23. Alvarado-Cabrero I, et al. Micropapillary carcinoma of the urothelial tract: a clinicopathologic study of 38 cases. Ann Diagn Pathol 2005; 9: 1 – 5.
24. Luna-More S, et al. Invasive micropapillary carcinoma of the breast. A new special type of invasive mammary carcinoma. Pathol Res Pract 1994; 190: 669 – 74. 25. Nassar H, et al. Pathogenesis of invasive micropapillary carcinoma: role of MUC1 glycoprotein. Mod Pathol 2004; 17: 1045 – 50. 26. World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of the Urinary System and Male Genital Organs. Eble JN, et al. (eds) Lyon, France: IARC Press, 2004. 27. Tamboli P, Ro JY. Villous adenoma of the urinary tract: a common tumor in an uncommon location. Adv Anat Pathol 2000; 7: 79 – 84. 28. Tamboli P, et al. Colonic adenocarcinoma metastatic to the urinary tract versus primary tumors of the urinary tract with glandular differentiation: a report of 7 cases and investigation using a limited immunohistochemical panel. Arch Pathol Lab Med 2002; 126: 1057 – 63. 29. Nielsen KK. Adenocarcinoma in exstrophy of the bladder – the last case in Scandinavia? A case report and review of literature. J Urol 1983; 130: 1180 – 2. 30. Mai KT, et al. Multicentric clear cell adenocarcinoma in the urinary bladder and the urethral diverticulum: evidence of origin of clear cell adenocarcinoma of the female lower urinary tract from Mullerian duct remnants. Histopathology 2000; 36: 380 – 2. 31. Zukerberg L, Harris NL, Young RH. Carcinomas of the urinary bladder simulating malignant lymphoma. A report of 5 cases. Am J Surg Pathol 1991; 15: 569 – 76. 32. Amin MB, et al. Lymphoepithelioma-like carcinoma of the urinary bladder. Am J Surg Pathol 1994; 18: 466 – 73. 33. Holmang S, Borghede G, Johansson SL. Bladder carcinoma with lymphoepithelioma-like differentiation: a report of 9 cases. J Urol 1998; 159: 779 – 82. 34. Lopez-Beltran A, et al. Lymphoepithelioma-like carcinoma of the urinary bladder: a clinicopathologic study of 13 cases. Virchows Arch 2001; 438: 552 – 7. 35. Constantinides C, et al. Lymphoepithelioma-like carcinoma of the bladder. BJU Int 2001; 87: 121 – 2. 36. Schubert GE, Pavkovic MB, Bethke-Bedurftig BA. Tubular urachal remnants in adult bladders. J Urol 1982; 127: 40 – 2. 37. Begg RC. The urachus: anatomy, histology, and development. J Anat 1930; 64: 170. 38. Hammond G, Yglesias L, Davis J. Urachus, its anatomy and associated fasciae. Anat Rec 1941; 80: 271. 39. Sheldon CA, et al. Malignant urachal lesions. J Urol 1984; 131: 1 – 8. 40. Henly DR, Farrow GM, Zincke H. Urachal cancer: role of conservative surgery. Urology 1993; 42: 635 – 9. 41. Siefker-Radtke AO, et al. Multimodality management of urachal carcinoma: the M. D. Anderson Cancer Center experience. J Urol 2003; 169: 1295 – 8. 42. Herr HW. Urachal carcinoma: the case for extended partial cystectomy. J Urol 1994; 151: 365 – 6. 43. Logothetis CJ, Samuels ML, Ogden S. Chemotherapy for adenocarcinomas of bladder and urachal origin: 5-fluorouracil, doxorubicin, and mitomycin-C. Urology 1985; 26: 252 – 5. 44. Mills SE, et al. Leiomyosarcoma of the urinary bladder. A clinicopathologic and immunohistochemical study of 15 cases. Am J Surg Pathol 1989; 13: 480 – 9.
Section 1 : Genitourinary Cancer
3
Urethral Cancer Oscar E. Streeter Jr and David I. Quinn
INTRODUCTION Primary carcinoma of the urethra is rare, representing less than 0.1% of all genitourinary neoplasms,1 – 3 or less than 80 patients each year by 1998 American Cancer Society estimate.4 Only 2000 patients have been reported since the mid-1800s. Because of its low incidence, no institution has been able to amass enough patients to evaluate different therapeutic options and develop consistent treatment strategies. Management decisions are therefore based on retrospective reports. The following discussion will review the important aspects of natural history, epidemiology, prognosis, and clinical management in men and women.
ANATOMY The anatomy and histology of the urethra differ greatly between males and females, and this leads to differences in pathological presentations. In the female, the urethra is a 4 cm long tubular conduit that courses obliquely anteroinferiorly from the internal urethral meatus through the urogenital diaphragm to the external urethral meatus. Multiple paraurethral glands of Skene (derivatives of the urogenital sinus and homologous to the prostate in males) secrete mucous material that provides urethral lubrication during sexual intercourse.5 By convention, in the female, the distal third of urethra is called the anterior urethra, while the proximal two-thirds is called the posterior urethra.6 The proximal third of the female urethra is lined with transitional cell epithelium, with the distal two-thirds lined with stratified squamous epithelium The male urethra, on the other hand, is divided into prostatic, membranous, and penile segments.7 The prostatic urethra is surrounded by the prostate, where the posterior wall of the urethra forms an elevation, the verumontanum (colliculus seminalis). The midline of the male urethra has an opening, the utriculus prostaticus, which is the rudimentary male homolog of the uterus.8 The urethra is located within the urogenital triangle and pierces the superficial and deep perineal spaces of the pelvic floor. Cancers of the anterior urethra preferentially drain into superficial inguinal lymph nodes. The posterior urethra (prostatic, membranous, and bulbar segments in the male
and the proximal two-thirds of the urethra in the female) generally drains into pelvic lymphatic channels.9 Lymphatic drainage of the posterior urethra in the female is to the pelvic lymph nodes, while those of the anterior urethra drain into the superficial and deep inguinal lymph nodes.6 Lymphatic drainage of the bulbomembranous urethra in the male is to the pelvic lymph nodes, while those of the penis are to the superficial and deep inguinal nodes.6
EPIDEMIOLOGY Unique among the urologic neoplasms, cancer of the urethra is more common in women than men, with a ratio of 4 : 1.10 Cases have been reported in all age-groups, but most present in the sixth decade of life or later.11 The ethnic incidence of adenocarcinoma of the urethra in women has been generally held to be higher among Caucasians than African-Americans or Hispanics, though one report differs in that opinion.12 The MD Anderson experience of Garden et al. was that 19 of 34 (56%) patients with adenocarcinoma were black. Of 22 cases of adenocarcinoma of the urethra in women reviewed, 9 were classified histologically as clear cell adenocarcinoma, and 13 were classified as columnar/mucinous adenocarcinoma.13 Our experience at the University of Southern California (USC) is that urethral adenocarcinoma is more common in African-American women than in other ethnic groups. No racial predisposition has been noted in men. Because of the small number of cases reported, it is difficult to ascertain significant etiologic agents. Even so, chronic inflammation, due to either infection or irritation, seems to play some role in the development of squamous cancers. A history of venereal disease, urethritis, multiple urethral instrumentation, and strictures will commonly be found in men.14,15 Human papillomaviruses (HPV) of genotype 16 and 18 are associated with human urogenital tumors.16 – 19 Although proliferative lesions such as papillomas, adenomas, and leukoplakia may also be important, especially in women, few associations have been substantiated.20 Heavy metal exposure, especially with arsenic, and the presence of urethral diverticulae appear to be important predisposing factors to adenocarcinoma of the bulbomembranous urethra in selected female populations.21 – 23
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
28
GENITOURINARY CANCER
PATHOLOGY Benign Lesions of the Urethra In the older surgical literature, before the widespread use of estrogen replacement therapy in postmenopausal women, 90% of the reported urethral masses in females were caruncles.24 Though these lesions are often asymptomatic, they can produce pain or hematuria in estrogen-deficient women. Caruncles histologically exhibit hyperplastic squamous epithelium with underlying submucosal vascularity, fibrosis, and inflammation.25 There are very few reports in the literature of benign tumors of the urethra. They include nephrogenic adenomas, papillomas, paragangliomas, amyloidosis, hemangiomas,26 fibromas, leiomyomas27 and sarcoidosis. Nephrogenic adenomas of the urinary tract are the most reported benign tumors of the urethra. Most nephrogenic adenomas occur in the urinary bladder (55%), less often in the urethra (41%), with 4% occurring in the ureters. The remaining nephrogenic adenomas occur in the prostate, but stain negative for prostatic specific antigen (PSA).28 These lesions represent benign metaplastic proliferations of urothelium, occurring mostly in males.29 Usually there is some associated trauma.30 In females, it can occur in a urethral diverticulum.31 – 35 The most common microscopic pattern is tubular, with the cystic pattern seen in 72% of the cases. Eosinophilic secretions are found in 75% of cases and basophilic secretions in 25%.22 These lesions do not transform into a malignancy but can recur after transurethral resection.36 Papillomas of the urethra are rare, occurring mainly in men.37 Paragangliomas outside the adrenals occur in only 10%, and only rarely in the urinary tract. The bladder is the most common site, the ureter is less frequently involved, and these are seen very rarely in the urethra.38 The cases are usually in elderly men and often discovered as an incidental finding. Grossly, they are polypoid or pedunculated lesions. Most occur at or proximal to the verumontanum and measure up to 15 mm in maximum dimension. There is no immunoreactivity for PSA or prostatic acid phosphatase. Neuroendocrine cells are normal cellular constituents in the bladder, prostate, and prostatic urethra, including Von Brunn’s nests and glandular metaplasia.39,40 Another benign lesion is amyloidosis of the urethra.41 When found, a workup for systemic amyloidosis should be performed since it is such a rare lesion, with only 18 reported cases.42 If localized, these lesions can spontaneously regress.43 Urethral capillary hemangiomas have been reported rarely in the world literature, with less than 20 cases documented.44 They have a tendency to recur locally, although the histological pattern of vascular sinuses and hemorrhages is clearly benign. Sex hormone cycles influence the clinical manifestations of this tumor in the female urethra in 7% of patients.45 There are only two reports of urethral leiomyomas presenting in pregnancy with measurable estrogen receptors (ER),46,47 although leiomyomas can occasionally occur without pregnancy in females.48 Grossly, it appears as a fleshy tumor, at times pedunculated. Microscopically, abundant muscle
fibers are seen and the cells have elongated nuclei with mitosis seen. ERs have been found in the male urethra in asymptomatic, as well as in symptomatic patients undergoing transurethral resection of the prostate (TURP) due to symptomatic infravesical obstruction, and also in patients with bladder cancer using the Abbott estrogen receptor immunocytochemical assay (ER-ICA) monoclonal assay for immunohistochemical demonstration of ERs. However, ER has not been demonstrated in the bladder. Bodker et al. noted that the difference in ER expression between the bladder and the prostatic urethra cannot be explained embryologically, since both the bladder and urethra are derived from the urogenital sinus. They proposed instead that the periurethral glands might have contributions from the M¨ullerian ducts, which in the female give rise to the ER-positive uterine tissue.49 Out of 21 reported patients with leiomyoma of the urethra, there is only one detailed case of ER-positive staining occurring in a male.50 Solitary urethral sarcoidosis is a rare occurrence,51 requiring workup for systemic sarcoidosis. An infection often seen in the urethra is condyloma acuminata.52 – 59 It is often associated with bladder or penis involvement.60 – 64 It is an important entity because it is a premalignant lesion.65,66
Malignant Lesions Carcinoma of the urethra is a rare tumor, with 1200 cases reported in females and 600 in males. In females, most are squamous cell carcinomas (55%), with 18% adenocarcinoma, 16% being transitional cell carcinomas (TCCs), and the remaining 11% being a mixture of rare tumors.8 In males, 80% of the tumors are squamous cell carcinomas, with 15% being TCCs, and the remainder adenocarcinomas. Most urethral cancers in the male are located in the bulbomembranous urethra (60%), with 30% in the penile urethra, and the remainder in the prostatic urethra.6 Gross Appearance
The gross appearance of urethral cancers may be papillomatous, mucosal thickening or induration, or a small spreading or superficial ulcer. As the lesion becomes more advanced, it will appear as an indurated, annular constricting, or fungating lesion.8 Microscopic Appearance
Squamous cell carcinoma (SCC) of the urethra usually has an area of focal keratinization that identifies it as SCC (see Figures 1–3). TCC of the urethra is indistinguishable from the appearance seen in TCC of the bladder (see Figure 4). Thus, one may note a papillary appearance, an in situ pattern, a papillary and infiltrating tumor, or an infiltrating carcinoma of any grade. A thorough examination of the whole urothelium must be performed for sites of concurrent malignancy, especially in the case of previous cystectomy for bladder carcinoma as there is the potential for a broadly based field defect of the entire mucosa. The adenocarcinoma of the urethra requires careful workup of possible extraurethral source.67,68 There is a frequently reported clear cell variant of adenocarcinoma, exhibiting
URETHRAL CANCER
Figure 1 Gross pathological specimen (longitudinal) of a squamous cell carcinoma of the proximal urethra (segmental resection). “A” represents the distal margin of the tumor. “B” represents the polypoid-appearing squamous cell carcinoma. “C” represents the proximal margin of the tumor. “D” represents periurethral soft tissue. “E” represents squamous metaplasia.
Figure 2 Cross-sections of the pathological specimen of the squamous cell carcinoma of the proximal urethra as seen in Figure 1. “A” represents the periurethral soft tissue. “B” represents cavities of the corpus cavernosum and corpus spongiosum. “C” represents the urethral opening.“D” represents the polypoid-shaped squamous cell carcinoma. The unlabeled specimen on the right represents the tumor at the level of the bulbomembranous urethra.
glycogen-rich cells. The clear cell variant can stain positive with antibodies to prostate-specific antigen and prostatic acid phosphatase.69 – 78 There is one case report of an adenocarcinoma of the urethra that had histological characteristics identical to a colon primary tumor (see Figure 5). On urethroscopy it produced mucin which, when this villous growth was seen in the bulbomembranous urethra, was thought to be a remnant of the anesthetic jelly used in the endoscopy procedure.79 Adenocarcinoma of the female urethra accounts for 10% of all urethral cancers, with the derivation of adenocarcinoma either from Skene’s glands and producing PSA or from urethritis transiting through intestinal metaplasia and producing mucin and not PSA. Dodson et al.71 evaluated 13 primary adenocarcinomas of the female urethra and found two histological groups: columnar/mucinous11 and clear cell.2 The
29
Figure 3 Micrograph of Figure 2, polypoid squamous cell carcinoma of the urethra. The white arrow represents the area of keratinization. The black arrow represents the polypoid portion of the tumor. Running along the bottom of the micrograph is a tongue of squamous metaplasia.
Figure 4 Micrograph of a high-grade transitional cell carcinoma of the urethra. The black arrow represents the lamina propria layer. The white arrow represents the TCC.
conclusion is that female urethral adenocarcinoma has more than one tissue of origin, with a minority arising from the Skene’s glands, accounting for less than 0.003% of all genital tract malignancies in females.70 Another review supporting more than one tissue of origin of female urethral adenocarcinoma was produced by Murphy et al. who studied 12 formalin-fixed archival tissue cases of primary female urethral adenocarcinoma, using the antibody mAbDas1 (formerly 7E12H12) developed for use against a unique colonic epithelial epitope and which is reactive in areas of intestinal metaplasia. Of the 12 cases, nine were columnar/mucinous adenocarcinoma, two clear cell adenocarcinoma, and one a cribiform pattern resembling adenocarcinoma of the prostate. All columnar/mucinous adenocarcinomas reacted positively with the mAbDas1 antibody but did not react with the PSA antibody. The tumor with a cribiform pattern reacted strongly with PSA but did not react with mAbDas1. The two clear cell adenocarcinomas did not react with either antibody.
30
GENITOURINARY CANCER
with a history of urinary incontinence, overflow, dysuria, and hematuria. In 19 gm of prostatic tissue obtained from a transurethral resection, there was a variety of morphologies. The specimen contained squamous metaplasia and poorly differentiated carcinoma that had clear cells, suggesting glandular pattern. Paget cells were seen in the basal portion of the prostatic urethra.116
INVESTIGATION Symptoms and Findings on Physical Examination
Figure 5 Micrograph of an adenocarcinoma of the urethra. The arrow is pointing to infiltrating colonic-type glands.
Benign urethral specimens demonstrate strong reactivity to the mAbDas1 antibody in areas of urethritis glandularis, while normal, inflamed urethral mucosa and TCC did not react.80
The female patient usually presents with hemorrhagic spotting, dyspareunia, and on pelvic examination a palpable urethral mass. Most tumors arise in the anterior urethra. The rest are in the proximal two-thirds or posterior urethra.6 The male patient usually presents with difficulty voiding and/or a bloody urethral discharge. A palpable nodule is felt in the urethra on examination of the penis and provides a rough estimate of the extent of the disease.6,8 Approximately 20% of patients present with palpably enlarged inguinal lymph nodes representing metastases. There is only one documented report of a patient presenting with sudden onset urinary retention and acute renal failure, later found to have a leiomyoma of the urethra.117
Rare Histologies Rare tumors of the urethra include non-Hodgkin’s lymphoma, reported in only eight well-documented cases, all females. The lesion appears as a firm meatal epithelium mass that may resemble a caruncle.81 Even rarer is an extramedullary plasmacytoma of the urethra, reported in only five patients.82 In these cases, patients do not necessarily develop multiple myeloma as seen in solitary plasmacytomas of the bone. There is only one reported case of primary carcinoid tumor of the urethra. The lesion grossly appeared as ventral palpable mass in the midpenile urethra and at the fossa navicularis. It was thought to be a TCC of the urethra. After 15 months with several intervening procedures, the patient developed facial flushing, sweating, diarrhea, loss of weight, and weakness. On examination, the patient was found to have an enlarged liver which, when palpated, triggered waves of pain, flushing, and sweating. An open liver biopsy confirmed a carcinoid malignancy that prompted review of all previous pathology, including specimens from the segmental resection, a recurrence treated by penile urethrectomy, bladder papilloma recurrence, and palpable lymph nodes from a fourth admission. The original urethral pathology was rereviewed with argentaffin stains and electron microscopy, and it was determined that it was consistent with a carcinoid primary.83 There are also many case reports of melanoma of the urethra, with pathological features similar to those found at more conventional sites.84 – 112 Paget’s disease of the urethra usually occurs as part of an extension from vulvar, penile, or bladder cancer.113 – 115 There is only one case of primary Paget’s disease of the urethra. It occurred in a 58-year-old African-American male seen at the Medical College of Virginia. The man presented
Staging The American Joint Commission on Cancer (AJCC) Staging System (see Table 1) is consistent with that of the International Union against Cancer (UICC), which makes reporting of treatment and survival more consistent in the literature worldwide.118
Endoscopic Examination Most diagnoses are made initially by cystourethroscopy and biopsy, using either a standard rigid cystoscope or a flexible instrument. Care must be taken not to traumatize the lesion, which may cause significant hemorrhage and make pathological interpretation more difficult. One must also be careful not to confuse a mucin-producing tumor with the anesthetic jelly used in endoscopy.79
Radiological Studies The gold standard for identifying the urethral lesion is still retrograde urethrography.6,8 Pelvic computed tomography (CT) is useful in identifying enlarged pelvic or retroperitoneal nodes, although the specificity and sensitivity of pelvic node imaging are somewhat limited. Multicoil magnetic resonance imaging (MRI) has enhanced the ability to visualize abnormalities of the female and urethra and periurethral tissues, contributing to improved surgical planning.5 Positron emission tomography (PET) scan has been used to evaluate any biological activity in pelvic or retroperitoneal nodes, although formal reporting of the accuracy of the test in this context has been somewhat limited. A PET scan also may obviate the need for a bone scan since uptake will often be seen in the cortex or medullary cavity of bones in instances of osseous metastatic disease.
URETHRAL CANCER Table 1 Clinical-pathologic staging for urethral cancer.118
Primary tumor (T) (men and women) Tx T0 Ta Tis T1 T2 T3
T4
Primary tumor cannot be assessed No evidence of primary tumor Noninvasive papillary, polypoid, or verrucous carcinoma Carcinoma in situ Tumor invades subepithelial connective tissue Tumor invades any of the following: corpus spongiosum, prostate, periurethral muscle Tumor invades any of the following: corpus cavernosum, beyond prostate capsule, anterior vagina, bladder neck Tumor invades other adjacent organs
Urothelial (transitional cell) carcinoma of the prostate Tis pu Tis pd T1 T2 T3
Carcinoma in situ, involvement of the prostatic urethra Carcinoma in situ, involvement of the prostatic ducts Tumor invades subepithelial connective tissue Tumor invades any of the following: prostatic stroma, corpus spongiosum, periurethral muscle Tumor invades any of the following: corpus cavernosum, beyond prostatic capsule, bladder neck (extraprostatic extension)
Regional lymph nodes (N) Nx N0 N1 N2
Regional lymph nodes cannot be assessed No regional lymph node metastasis Metastasis in a single lymph node, 2 cm or less in greatest dimension Metastasis in a single lymph node more than 2 cm in greatest dimension, or in multiple lymph nodes
Distant metastasis (M) Mx M0 M1
Distant metastasis cannot be assessed No distant metastasis Distant metastasis
Stage Grouping Stage 0a Stage 0is
Stage I Stage II Stage III
Stage IV
Ta Tis Tis pu Tis pd T1 T2 T1 T2 T3 T3 T4 T4 Any T Any T
N0 N0 N0 N0 N0 N0 N1 N1 N0 N1 N0 N1 N2 Any N
M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1
From AJCC 2002 Cancer Staging Manual, 6th Edition.
PROGNOSIS AND NATURAL HISTORY The 5-year overall survival after a diagnosis of urethral carcinoma ranges from 32 to 51% (see Table 2). Clinical and pathological stages predict the prognosis: patients with stage I or II disease at diagnosis have a 50–83% 5-year overall survival rate, compared to 0–42% for those with stage II to IV disease (see Table 2). The presence of local invasion and/or lymph node metastases has consistently been associated with a poorer outcome. In addition, involvement of the bulbomembranous (proximal) urethra or the entire length of urethra has a worse prognosis than involvement
31
limited to the distal segment. Adenocarcinoma has a poorer survival rate compared to either transitional or squamous cell carcinoma in most series. In contrast, tumor grade, while it may have predictive significance as a single variable in some series, is not significant when other factors such as stage are incorporated, although this may reflect the statistical limitations of the small numbers of case studies. SCCs and TCCs tend to progress by spread to local lymph node groups, typically the inguinal nodes, then paraaortic lymph nodes, and subsequently the lungs and other visceral organs. Adenocarcinoma tends to have a slightly different pattern of spread, with early involvement of the peritoneum and associated ascites even in the absence of lymph node involvement. Subsequently, patients may develop lung (pleural), liver, and bone metastases.
TREATMENT Surgery Surgery in the Female Patient
The anatomical presentation of the tumor determines the treatment approach. In the more common anterior urethral lesions (distal 1/3), where the goal is curative, a cure rate of 75% at 5 years can be obtained from either surgery or radiation, depending on the histology and extent of disease.6 Laser resection and fulguration can treat urethral tumors that are restricted to the mucosa and tumors that are very superficial.10,126 The advantage of the laser over local excision is the absence of bleeding, faster healing, and the possible avoidance of hypertrophic scar associated with deeper penetration seen in surgery. However, there is a higher chance of recurrence, but the laser resection can be repeated.127 For anterior urethral tumors of low clinical stage that are small and visible, local excision is all that should be necessary, with transurethral resection reserved for lesion in the submucosa (Tis or T1 tumors). Partial urethrectomy, based on clinical situation, can be performed on Tis, T1, and T2 lesions of the anterior urethra.10 Total urethrectomy is used with T2/T3 lesions where bladder preservation is possible.10,128 For tumors of the posterior urethra, the overall survival rate is only 10–17% at 5 years. Patients presenting with involvement of both posterior and anterior urethra also have a poor prognosis. Surgery involves an anterior pelvic exenteration, urinary diversion, and ilioinguinal dissection.6 For defects that are too large for primary closure, the gracilis musculocutaneous flap provides healthy tissue with its own blood and nerve supply to close the wound.129 If the tumor is a Skene’s (periurethral) gland carcinoma, and preoperatively the PSA is elevated, postoperative levels of PSA should decrease after surgical removal.70 Surgery in the Male Patient
As in the female, the anatomical presentation of the tumor determines the treatment approach where the overall survival rates are 22% at 5 years for penile urethra tumors, and only 10% at 5 years for those of the bulbomembranous urethra.6,130 Surgery for penile urethra tumors involves
59.5 59.2 36 – 92 NR 125 1 – 336 1 (2) 29 (63) 15 (33) 1 (2) 18 (39) 28 (61)
21 (49) 22 (51) 42 83 36
61 NR 23 – 78
50 NR 5 – 156
0 19 (83) 3 (13) 1 (4)
15 (65) 8 (35)
6 (26) 17 (74) 41 83 31
46 46 male only
1958 – 1996 Surgery
1979 – 1990 Surgery
23 23 male only
MSKCC NY120
MDACC Houston119
9 (43) 12 (57)
4 (36)
11 (52) 1 (5) 1 (5) 7 (64)
42.1 38 5 – 96
59.2 32 – 80 NR
21 11 : 10
1980 – 1996 Neoadjuvant RT + CT, surgery
BAKCC Detroit21
6 (43) 8 (57)
5 (35) 9 (65)
6 (43) 5 (36) 3 (21) 0
55.6 36 5 – 160
53 52.6 28 – 72
21 14 : 7
1988 – 2001 Surgery
NCKUH Taiwan23
0
50
9 (50) 9 (50) 33
9 (50) 9 (50)
9 (50) 5 (28) 4 (22) 0
NR NR 18 – 70
NR 58 45 – 72
18 36 : 18 only female data reported
1991 – 2000 Surgery, CT, RT
TMH Mumbai121
40 (41) 57 (59) 41
34 (35) 63 (65)
34 (35) 40 (41) 21 (22) 2 (3)
NR 105 20 – 337
63 NR 36 – 89
97 97 female only
1955 – 1989 RT
MDACC Houston13
13 (30) 31 (70) 42
20 (47) 23 (53)
23 13 5 3
NR 99 30 – 282
NR 67 37 – 89
44 44 female only
1959 – 1995 RT with or without surgery
WUSM St Louis122
22
78
19 (31) 42 (69) 32
25 (35) 40 (56)
25 (35) 28 (39) 11 (15) 8 (11)
NR 85 0 – 384
60 59 21 – 84
72 72 female only
1958 – 1994 Surgery or RT
MSKCC NY123
8 (24) 26 (76) 58
14 (41) 20 (59)
6 (18) 15 (44) 13 (38) 0
NR 84 21 – 325
67 NR 30 – 80
34 34 female only
1961 – 1990 RT
PMH Toronto124
42
59
27 (51) 26 (49) 51
29 (55) 24 (45)
14 21 15 0
NR 174 19 – 337
63 63 36 – 92
53 53 female only
1948 – 1999 Surgery
Mayo Clinic Rochester125
MDA, University of Texas M.D. Anderson Cancer Center; MSKCC, Memorial Sloan Kettering Cancer Center; BAKCC, Barbara A. Karmanos Cancer Center; NCKUH, National Cheng Kung University Hospital; TMH, Tata Memorial Hospital (India); WUSM, Washington University School of Medicine; PMH, Princess Margaret Hospital; RT, radiation therapy; CT, chemotherapy; NR, not reported. Note: Numbers inside the parenthesis indicate percentages.
Age (years) Median Mean Range Follow-up (months) Mean Median Range Histological type Adenocarcinoma Squamous Transitional Other Tumor location Distal Proximal or entire (Bulbomembranous /Prostatic) Stage I – II III – IV 5-year overall survival (%) 5-year overall survival – stage I – II (%) 5-year overall survival – stage III – IV (%)
Characteristic Number Gender: M:F
Predominant primary local therapy
Hospital/Cancer Center (Reference No.)
Table 2 Summary table for several contemporary series of urethral carcinoma.
32 GENITOURINARY CANCER
URETHRAL CANCER
local excision, partial or total penectomy. Local excision can provide long-term disease control in distal tumors when followed by adjuvant radiation therapy.131 – 134 A more established procedure for distal tumors is partial penectomy. The advantage of this procedure is a 2 cm proximal margin while preserving sufficient penile length to allow urination in a standing position with excellent disease control.130 The ideal patient for radical penectomy is a patient with T2 disease (involvement of corpus spongiosum) whose tumor does not extend proximal to the midbulb or a patient with a distal neoplasm. If there is evidence of T3 disease, en bloc resection with cystoprostatectomy is the best surgical option.130 En bloc resection of the penis, bilateral ilioinguinal node dissection, cystoprostatectomy, and urinary diversion should be performed if surgery is the treatment option for carcinoma of the bulbomembranous urethra.
Radiotherapy Radiotherapy in the Female Patient
Early lesions can be treated with interstitial irradiation with excellent results (75% 5-year results), with surgery used for failures or persistent tumors. The technique employs afterloading brachytherapy catheters utilizing iridium-192 (Ir-192) to form a volume implant in a circular pattern around the urethral orifice, with placement based on tumor volume, with the use of MRI for pretreatment planning.135 The treatment is carried out using a disposable Syed-Neblett template with blind-end stainless steel (17-gauge) of Flexguide catheters. After verification, simulation films are taken with dummy seeds, a low-dose Ir-192 treatment is delivered between 60 and 70 Gy at the rate of 60–120 cGy hour−1 to the periphery of the implant when used alone. For patients with bulky disease, the whole pelvis is treated with external beam irradiation for control of subclinical disease to a total of 45–50 Gy in 5 to 5.5 weeks. The portals cover the inguinal, external, and internal iliac lymph nodes. If the inguinal nodes are involved, they are boosted to 60 or 65 Gy with separate electron beams. There is debate on whether the legs should be frog-legged to flatten the dose to the perineum or whether they should be left in a closed position, and the choice is left to the treating radiation oncologist. Confluent moist desquamation is inevitable, which may require a treatment break of 1 or 2 weeks, with topical care with aluminum acetate (Domeboro’s) solution and silver sulfadiazine (Silvadene) 1% cream. Bolus or tissue compensators can reduce this complication. An additional 20–35 Gy is delivered by afterloading interstitial implant using a modified disposable Syed-Neblett template for a total dose of 70–80 Gy.135 Follow-up consists of physical examination and cystoscopy every 3 months, with biopsy at the first visit to ensure there is no residual disease.136 For tumors of the anterior meatus, the 5-year survival rates are reported to be 100%.137,138 In a study of 62 female patients with primary carcinoma of the urethra treated with combined radiation therapy, 42 patients (67.7%) had tumors of the anterior urethra and 20 (32.3%) had tumors that involved the posterior urethra. The overall survival rate was 64.5%, with anterior urethral carcinomas having
33
a higher 5-year survival rate (71.4%). Patients with posterior carcinoma have a 5-year survival of only 50%.139 Followup consists of physical examination and cystoscopy every 3 months, with biopsy at the 3-month point to see if there is residual disease. For patients who are medically inoperable with urethral carcinoma, treatment can be delivered with external beam radiation therapy and high-dose rate (HDR) intracavitary brachytherapy.140 Georgetown University reported on four women with locally advanced urethral cancer who received whole pelvis external beam radiotherapy to a planned dose of 45 Gy in 1.8 Gy fractions using AP-PA fields. The anterior field included the inguinal nodes and was not opposed by the posterior field. The inguinal fields were boosted separately with electron fields. A HDR afterloader with a 10 Ci Ir192 source was used for boost therapy. The treatment was carried out by running the source down a modified 20F Foley catheter in a vaginal cylinder. This was used to displace the mucosa of the posterior wall. A shield was placed posteriorly within the cylinder to provide further protection to the posterior vaginal wall and rectum. Patients received three or four HDR intracavitary implants, with a dose of 7 Gy isodose cloud encompassing the tumor with margin. The total dose was 66–70 Gy. All three patients had a functioning urethra afterwards. Two died of complications of metastatic disease at 22 and 25 months, with one dying of an unrelated medical problem 12 months later. The remaining patient was diseasefree at 55 months with only urethral meatal telangiectasia and occasional urinary urgency. Radiotherapy in the Male Patient
Radiation therapy in distal urethral lesions provides organ preservation. Early lesions, as in females, can be treated with intracavitary therapy with or without interstitial needles implanted around the urethral meatus. For more advanced lesion, a technique can be employed using parallel-opposed fields with the penis suspended vertically by a urethral catheter141 or by using a tissue-equivalent mold. However, this technique is not useful to treat inguinal nodes. This technique can be combined with an intracavitary boost using a modified 20F Foley catheter. All patients develop a brisk skin reaction and swelling, which should subside in 2 to 4 weeks. Chronically, patients can develop urethral meatal strictures. Radiation therapy for primary non-Hodgkin’s lymphoma of the urethra can be the primary mode of treatment for early stage disease.81 Plasmacytoma of the urethra can be treated with radiation therapy with a low dose of pelvic irradiation (41.4 Gy), with a long-term disease-free survival (more than 12 years).82
Cytotoxic Chemotherapy Chemotherapy was previously considered only for urethral cancers with inguinal or pelvic node involvement, where the survival rates at 5 years are 10 to 30% or for patients presenting with metastatic disease. The drugs most commonly used have been cisplatin (CDDP), bleomycin, and methotrexate (MTX).6 Single agents can provide a useful palliative response for some patients.142 The overall approach
34
GENITOURINARY CANCER
to locally advanced disease (T3 or greater) and patients with posterior urethral tumors has evolved because of data showing extremely poor outcomes for patients treated with single modality local treatments. Patients commonly relapse in a pattern where loss of local disease control and development of distant metastases are contemporaneous or only separated by a few months. This better understanding of the disease has led to the use of chemotherapy with or without radiation in the neoadjuvant setting for patients having surgery, or as primary therapy for those with distant spread who need local disease control for quality of life and morbidity benefit. Urethral carcinomas are generally responsive to platinumbased cytoxic regimens. Neoadjuvant strategies in this disease show high complete response rates to chemotherapy either alone or in combination with radiation therapy.21 [Editorial Note: Despite evidence of initial objective responses, in more advanced tumors, these responses are often shortlived.] As with surgery and radiation therapy, there are no prospective evaluations of response and outcome or clinical trials of cytotoxic therapy in urethral carcinoma because of the rarity of the disease. On this basis, chemotherapeutic strategies are derived either from experience in urethral cancer or by paradigm extrapolation from more common cancers occurring in the pelvis with similar histologies. Some oncologists model therapy for squamous cell urethral carcinoma on therapies developed for cervical, vulvar, and anal carcinomas, for urethral transitional carcinoma on that used for upper tract and bladder TCC, and for adenocarcinoma on that used for rectal carcinoma: • Current standard therapy for SCC of the cervix or vulva usually involves CDDP (with or without 5 fluorouracil – 5-FU) given concurrently with radiation therapy, while that for anal SCC is either 5-FU and mitomycin C or CDDP and mitomycin C given concurrently with radiation.143 – 145 • In urothelial cancer, the MVAC (Methotrexate, Vinblastine, Adriamycin (doxorubicin), and Cisplatin) regimen is a long time standard for adjuvant or neoadjuvant treatment.146 A recent randomized phase III trial run by the Eli Lilly Company compared MVAC with gemcitabine and cisplatin (GC) in patients with metastatic TCC and found response and survival to be virtually identical but with less toxicity in the GC arm.147 Although likely to be relevant to the management of urethral cancer, this study did not specifically address tumors at this site. In patients with localized bladder cancer, concurrent single agent CDDP and RT result in good local disease control and bladder preservation in up to 70% of patients, with surgical salvage for the remaining 30%.148 In addition, the activity of three drug regimens incorporating platinum, gemcitabine, and a taxane (GCT) has produced impressive response rates in advanced bladder cancer, including cases with uncommon histological types such as squamous cell or adenocarcinoma.149,150 Whether GCT is superior to GC in advanced TCC awaits the outcome of a phase III trial recently completed by the European Organization for the Research and Treatment of Cancer and the Southwest Oncology Group.151 • Rectal adenocarcinoma is now routinely treated with radiation therapy and 5-FU before surgery.152,153 Several trials
are testing whether the addition of new platinums such as oxaliplatin or biological agents targeted at angiogenesis or the epidermal growth factor receptor might further improve outcomes.154,155
Combined Chemoradiotherapy Chemoradiotherapy has been explored in patients with tumors that are locally advanced or involve the proximal urethra.156,157 The radiation therapy dose is a minimum of 30 Gy with a maximum dose of 55 Gy. The use of concurrent cytoxic agents in this setting is commonplace, but a wide selection of drugs has been reported. Use of regimens involving CDDP, infusional 5-FU and/or mitomycin C as single agents or combination depends upon the familiarity of the oncologist with the specific regimens. Our approach at the University of Southern California has been to use weekly CDDP at a dose of 20 mg m−2 of body surface area in combination with 5-FU 350–450 mg m−2 day−1 infused over 4 days every third week. This regimen is used, depending upon patient tolerance, for 6 weeks concurrently with radiation therapy. The common side effects of this approach include inflammation of the skin, bladder, urethra, vagina, and rectum, dehydration from diarrhea, and electrolyte imbalance from CDDP. These are all manageable, provided the clinicians involved in providing treatment are diligent in assessing for these effects. Once radiation therapy is complete, we normally give another 6 weeks of chemotherapy with CDDP and 5-FU to complete the therapeutic program.
Chemotherapy for Metastatic Disease CDDP and 5-FU have been considered a standard in this disease for many years. Recently, we have utilized a regimen combining CDDP or carboplatin with gemcitabine and a taxane. We have found this combination tolerable, but the small case numbers preclude formal comparison with the utility and efficacy of CDDP and 5-FU. While significant disease shrinkage and clinical improvement are common, complete response of all disease amounting to remission is rare and all patients relapse and eventually die, usually from their cancer. In this setting, newer biological agents typically targeting tyrosine kinases are theoretically attractive. Until we have some indication of the response of these patients in formal clinical trials, or are able to better profile patients to determine their response, indiscriminate use of these agents in patients with urethral cancer remains unwise.
SUMMARY Although the urethra comes embryologically from the same urogenital sinus as the bladder, its tumors are more variable, though rare. Treatment also is more variable, based on anatomical location of the tumor, which determines overall survival. Urethral cancers are usually reported in small case series and therefore there is no clear standard of care. When possible, if the possibility of cure is not compromised, organ preservation should be the primary goal, and this may be effected by the use of combined modality therapies.
URETHRAL CANCER
REFERENCES 1. Srinivas V, Khan SA. Urethral cancer. Hosp Pract (Off Ed) 1985; 20(5): 147 – 8, 52 – 4. 2. Srinivas V, Khan SA. Female urethral cancer – an overview. Int Urol Nephrol 1987; 19(4): 423 – 7. 3. Srinivas V, Khan SA. Male urethral cancer. A review. Int Urol Nephrol 1988; 20(1): 61 – 5. 4. Landis SH, et al. Cancer statistics, 1998. CA Cancer J Clin 1998; 48: 6 – 29. 5. Prasad SR, et al. Cross-sectional imaging of the female urethra: technique and results. Radiographics 2005; 25: 749 – 61. 6. Harty JI. Decision Making in Oncology: Evidence Based Management. New York: Churchill Livingstone, 1997. 7. Terry PJ, Cookson MS, Sarosdy MF. Carcinoma of the urethra and scrotum. In Raghavan D, et al. (eds) Principles and Practice of Genitourinary Oncology. Philadelphia, Pennsylvania: LippincottRaven Publishers, 1997: 347 – 354. 8. Mostofi FK, Davis CJ, Sesterhenn IA. Carcinoma of the male and female urethra. Urol Clin North Am 1992; 19: 347 – 58. 9. Carroll PR, Dixon CM. Surgical anatomy of the male and female urethra. Urol Clin North Am 1992; 19(2): 339 – 46. 10. Narayan P, Konety B. Surgical treatment of female urethral carcinoma. Urol Clin North Am 1992; 19(2): 373 – 82. 11. Beduschi MC, Wishnow KI, Oesterling JE. Urethral carcinoma: diagnosis and staging. In Oesterling JE, Richie JP (eds) Urologic Oncology, 1st ed. Philadelphia, Pennsylvania: W. B. Saunders, 1997: 561 – 571. 12. Meis JM, Ayala AG, Johnson DE. Adenocarcinoma of the urethra in women. A clinicopathologic study. Cancer 1987; 60(5): 1038 – 52. 13. Garden AS, Zagars GK, Delclos L. Primary carcinoma of the female urethra. Results of radiation therapy. Cancer 1993; 71(10): 3102 – 8. 14. Kaplan GW, Buckley GJ, Grayhack JT. Carcinoma of the male urethra. J Urol 1967; 98: 365. 15. Bostwick DG, Lo R, Stamey TA. Papillary adenocarcinoma of the male urethra. Case report and review of the literature. Cancer 1984; 54: 2556 – 63. 16. Wiener JS, Liu ET, Walther PJ. Oncogenic human papillomavirus type 16 is associated with squamous cell cancer of the male urethra. Cancer Res 1992; 52(18): 5018 – 23. 17. Wiener JS, Walther PJ. The association of oncogenic human papillomaviruses with urologic malignancy. The controversies and clinical implications. Surg Oncol Clin N Am 1995; 4(2): 257 – 76. 18. Wiener JS, Walther PJ. A high association of oncogenic human papillomaviruses with carcinomas of the female urethra: polymerase chain reaction-based analysis of multiple histological types. J Urol 1994; 151(1): 49 – 53. 19. Cupp MR, et al. Detection of human papillomavirus DNA in primary squamous cell carcinoma of the male urethra. Urology 1996; 48(4): 551 – 5. 20. Vesa Llanes J, et al. Inverted papilloma of the prostatic urethra. Histopathogenetic considerations. Arch Esp Urol 1994; 47(10): 1022 – 4. 21. Gheiler EL, et al. Management of primary urethral cancer. Urology 1998; 52(3): 487 – 93. 22. Oliva E, Young RH. Nephrogenic adenoma of the urinary tract: a review of the microscopic appearance of 80 cases with emphasis on unusual features. Mod Pathol 1995; 8(7): 722 – 30. 23. Tsai YS, et al. Experience with primary urethral carcinoma from the blackfoot disease-endemic area of South Taiwan: increased frequency of bulbomembranous adenocarcinoma? Urol Int 2005; 74(3): 229 – 34. 24. Marshall PC, Uson AC, Melicow MM. Neoplasms and caruncles of the female urethra. Surg Gynecol Obstet 1960; 110: 723 – 33. 25. Dmochowski RR, et al. Benign female periurethral masses. J Urol 1994; 152: 1943 – 51. 26. Hayashi T, Igarashi K, Sekine H. Urethral hemangioma: case report. J Urol 1997; 158(2): 539 – 40. 27. Scholl AJ, Braasch WF. Primary tumors of the urethra. Ann Surg 1922; 76: 246 – 59.
35
28. Young RH. Nephrogenic adenomas of the urethra involving the prostate gland: a report of two cases of a lesion that may be confused with prostatic adenocarcinoma. Mod Pathol 1992; 5(6): 617 – 20. 29. Baghavan BS, et al. Nephrogenic adenoma of the urinary bladder and urethra. Hum Pathol 1981; 12: 907 – 16. 30. Carcamo Valor PI, et al. Nephrogenic adenoma of the upper and lower urinary tract. Apropos of 22 cases. Arch Esp Urol 1992; 45(5): 423 – 7. 31. Klutke CG, Akdman EI, Brown JJ. Nephrogenic adenoma arising from a urethral diverticulum: magnetic resonance features. Urology 1995; 45(2): 323 – 5. 32. Materne R, et al. Apropos of a case of nephrogenic adenoma in a urethral diverticulum in a woman. Acta Urol Belg 1995; 63(4): 13 – 8. 33. Medeiros LJ, Young RH. Nephrogenic adenoma arising in urethral diverticula. A report of five cases. Arch Pathol Lab Med 1989; 113(2): 125 – 8. 34. Miyake O, et al. A case of nephrogenic adenoma in the female urethral diverticulum. Hinyokika Kiyo 1990; 36(10): 1189 – 92. 35. Pamplona M, et al. Nephrogenic adenoma in urethral diverticulum in women. Actas Urol Esp 1990; 14(4): 277 – 8. 36. Berger BW, et al. Nephrogenic adenoma: clinical features and therapeutic considerations. J Urol 1981; 126(6): 824 – 6. 37. Ojea Calvo A, et al. Inverted papilloma of the urethra. Actas Urol Esp 1993; 17(3): 193 – 5. 38. Boyle M, Gaffney EF, Thurston A. Paraganglioma of the prostatic urethra. A report of three cases and a review of the literature. Br J Urol 1996; 77(3): 445 – 8. 39. Freedman SR, Goldman RL. Normal paraganglia in the human prostate. J Urol 1975; 113: 874 – 5. 40. Kiernan M, Gaffney EF. The endocrine-paracrine cells of Von Brunn’s and glandular metaplasia in the supramontanal prostatic urethra. Histopathology 1975; 16: 367 – 70. 41. Branson AD, et al. Localized amyloidosis of the urethra: report of a case. J Urol 1969; 101(1): 68 – 70. 42. Provet JA, et al. Primary amyloidosis of urethra. Urology 1989; 34(2): 106 – 8. 43. Brown RD, et al. Localized amyloidosis of the urethra: diagnostic implications and management. J Urol 1988; 140(6): 1536 – 8. 44. Roberts JW, Devine CJ Jr. Urethral hemangioma: treatment by total excision and grafting. J Urol 1983; 129(5): 1053 – 4. 45. Fry M, et al. Leiomyoma of the female urethra. J Urol 1988; 140(3): 613 – 4. 46. Kato T, et al. Urethral leiomyoma expressing estrogen receptors. Int J Urol 2004; 11(7): 573. 47. Kesari D, et al. Estrogen receptors in a urethral leiomyoma presenting in pregnancy [letter]. Int J Gynaecol Obstet 1994; 47(1): 59 – 60. 48. Dasan JC, Rao K, Nalini V. Leiomyoma of the female urethra – a clinical curiosity. Int J Gynaecol Obstet 1989; 28(4): 381 – 3. 49. Bodker A, et al. Estrogen receptors in the human male bladder, prostatic urethra, and prostate. An immunohistochemical and biochemical study. Scand J Urol Nephrol 1995; 29(2): 161 – 5. 50. Ohtani M, et al. Leiomyoma of the male urethra. Eur Urol 1982; 8(6): 372 – 3. 51. Ho KL, Hayden MT. Sarcoidosis of urethra simulating carcinoma. Urology 1979; 13(2): 197 – 9. 52. Bissada NK, Redman JF, Sulieman JS. Condyloma acuminatum of male urethra. Successful management with 5-fluorouracil. Urology 1974; 3(4): 499 – 501. 53. Brenner M, et al. Intraurethral condyloma acuminatum: current management. J Am Osteopath Assoc 1983; 82(8): 611 – 5. 54. Buscemi ML, Silber LM, Hanna MK. Intraurethral condylomas acuminata in previous hypospadias repair. Urology 1985; 26(4): 398 – 9. 55. Cervigni M, et al. Acute urethral obstruction due to condylomata acuminata. Obstet Gynecol 1991; 78(5 Pt 2): 970 – 2. 56. Cetti NE. Condyloma acuminatum of the urethra: problems in eradication. Br J Surg 1984; 71(1): 57. 57. Kesner KM. Extensive condylomata acuminata of male urethra: management by ventral urethrotomy [see comments]. Br J Urol 1993; 71(2): 204 – 7. 58. Mininberg DT, Rudick DH. Urethral condyloma accuminata in male children. Pediatrics 1976; 57(4): 571 – 3.
36
GENITOURINARY CANCER
59. Redman JF, Meacham KR. Condyloma acuminata of the urethral meatus in children. J Pediatr Surg 1973; 8(6): 939 – 41. 60. Asvesti C, et al. Multiple condylomata of the urethra and bladder disclosing HIV infection. Ann Urol 1991; 25(3): 146 – 9. 61. Benoit G, et al. Presence of papilloma virus type 11 in condyloma acuminatum of bladder in female renal transplant recipient. Urology 1988; 32(4): 343 – 4. 62. Bissada NK, Cole AT, Fried FA. Extensive condylomas acuminata of the entire male urethra and the bladder. J Urol 1974; 112(2): 201 – 3. 63. Pompeius R, Ekroth R. A successfully treated case of condyloma acuminatum of the urethra and urinary bladder. Eur Urol 1976; 2(6): 298 – 9. 64. Wallin J. 5-Fluorouracil in the treatment of penile and urethral condylomata acuminata. Br J Vener Dis 1977; 53(4): 240 – 3. 65. Libby JM, Frankel JM, Scardino PT. Condyloma acuminatum of the bladder and associated urothelial malignancy. J Urol 1985; 134(1): 134 – 6. 66. Noronha RF, Sundaram M. Are intraurethral condylomata premalignant? Br J Urol 1984; 56(5): 546 – 7. 67. Llarena Ibarguren R, et al. Urethral metastases of prostatic adenocarcinoma. Arch Esp Urol 1993; 46(9): 779 – 82. 68. Stragier J, et al. Adenocarcinoma of the rectum with a solitary metastasis to the urethra in a female. Eur J Surg Oncol 1994; 20(6): 696 – 7. 69. Assimos DG, O’Conor VJ Jr. Clear cell adenocarcinoma of the urethra. J Urol 1984; 131(3): 540 – 1. 70. Dodson MK, et al. Skene’s gland adenocarcinoma with increased serum level of prostate- specific antigen. Gynecol Oncol 1994; 55(2): 304 – 7. 71. Dodson MK, et al. Female urethral adenocarcinoma: evidence for more than one tissue of origin? Gynecol Oncol 1995; 59(3): 352 – 7. 72. Doria MI Jr, et al. Cytologic features of clear cell carcinoma of the urethra and urinary bladder. Diagn Cytopathol 1996; 14(2): 150 – 4. 73. Drew PA, et al. The histogenesis of clear cell adenocarcinoma of the lower urinary tract. Case series and review of the literature. Hum Pathol 1996; 27(3): 248 – 52. 74. Ebisuno S, Miyai M, Nagareda T. Clear cell adenocarcinoma of the female urethra showing positive staining with antibodies to prostatespecific antigen and prostatic acid phosphatase. Urology 1995; 45(4): 682 – 5. 75. Hull MT, et al. Glycogen-rich clear cell carcinoma of the urethra: an ultrastructural study. Ultrastruct Pathol 1987; 11(4): 421 – 7. 76. Oliva E, Young RH. Clear cell adenocarcinoma of the urethra: a clinicopathologic analysis of 19 cases. Mod Pathol 1996; 9(5): 513 – 20. 77. Seballos RM, Rich RR. Clear cell adenocarcinoma arising from a urethral diverticulum. J Urol 1995; 153(6): 1914 – 5. 78. Tanabe ET, Mazur MT, Schaeffer AJ. Clear cell adenocarcinoma of the female urethra: clinical and ultrastructural study suggesting a unique neoplasm. Cancer 1982; 49(2): 372 – 8. 79. Yachia D, Turani H. Colonic-type adenocarcinoma of male urethra. Urology 1991; 37(6): 568 – 70. 80. Murphy DP, et al. Female urethral adenocarcinoma: immunohistochemical evidence of more than 1 tissue of origin. J Urol 1999; 161(6): 1881 – 4. 81. Selch MT, et al. Primary lymphoma of female urethra: long-term control by radiation therapy. Urology 1993; 42(3): 343 – 6. 82. Mordkin RM, Skinner DG, Levine AM. Long-term disease-free survival after plasmacytoma of the urethra: a case report and review of the literature. Urology 1996; 48(1): 149 – 50. 83. Sylora HO, et al. Primary carcinoid tumor of the urethra. J Urol 1975; 114(1): 150 – 3. 84. Aragona F, et al. Primary malignant melanoma of the female urethra: a case report. Int Urol Nephrol 1995; 27(1): 107 – 11. 85. Arai K, et al. Primary malignant melanoma of the female urethra: a case report. Jpn J Clin Oncol 1993; 23(1): 74 – 7. 86. Ariel IM. Malignant melanoma of the female genital system: a report of 48 patients and review of the literature. J Surg Oncol 1981; 16(4): 371 – 83. 87. Block NL, Hotchkiss RS. Malignant melanoma of the female urethra: report of a case with 5-year survival and review of the literature. J Urol 1971; 105(2): 251 – 5.
88. Buckle AE. Primary malignant melanoma of the female urethra. Br J Surg 1969; 56(7): 548 – 50. 89. Calcagno L, et al. Primary malignant melanoma of male urethra. Urology 1991; 37(4): 366 – 8. 90. Clarke JF. Urethral metastasis from a presumed primary malignant melanoma presenting as postmenopausal bleeding. Proc R Soc Med 1975; 68(4): 227 – 8. 91. Fujimoto N, Oda M, Shimoe S. Primary malignant melanoma of the male urethra. Urol Int 1991; 47(3): 176 – 7. 92. Geelhoed GW, Myers GH Jr. Primary melanoma of the male urethra. J Urol 1973; 109(4): 634 – 7. 93. Godec CJ, et al. Melanoma of the female urethra. J Urol 1981; 126(4): 553 – 5. 94. Heslinga JM, Lycklama a Nijeholt GA, Ruiter DJ. Primary melanoma in the female distal urethra. Eur Urol 1986; 12(6): 446 – 7. 95. Iversen K, Robins RE. Mucosal malignant melanomas. Am J Surg 1980; 139(5): 660 – 4. 96. Iyer KM, et al. Primary malignant melanoma of the male urethra. Indian J Cancer 1974; 11(2): 213 – 4. 97. John G, et al. Primary malignant melanoma of the male urethra. Br J Urol 1992; 69(2): 212 – 3. 98. Katz JI, Grabstald H. Primary malignant melanoma of the female urethra. J Urol 1976; 116(4): 454 – 7. 99. Kim CJ, et al. Primary malignant melanoma of the female urethra. Cancer 1993; 71(2): 448 – 51. 100. Kokotas NS, Kallis EG, Fokitis PJ. Primary malignant melanoma of male urethra. Urology 1981; 18(4): 392 – 4. 101. Lopez JI, Angulo JC, Ibanez T. Primary malignant melanoma mimicking urethral caruncle. Case report. Scand J Urol Nephrol 1993; 27(1): 125 – 6. 102. Methfessel HD, Bettzieche H, Methfessel G. Melanoma of the female urethra. Zentralbl Gynakol 1983; 105(12): 796 – 800. 103. Miyauchi T, Maruoka M, Nagayama T. Malignant melanoma of the penis associated with von Recklinghausen’s neurofibromatosis: report of a case. Hinyokika Kiyo 1988; 34(4): 710 – 3. 104. Morrow CP, DiSaia PJ. Malignant melanoma of the female genitalia: a clinical analysis. Obstet Gynecol Surv 1976; 31(4): 233 – 71. 105. Pillai KR, et al. Cytodiagnosis of primary malignant melanoma of the female urethra – a case report [published erratum appears in Indian J Pathol Microbiol 1995 Apr; 38(2):138]. Indian J Pathol Microbiol 1995; 38(1): 103 – 8. 106. Primus G, et al. Early ‘invasive’ malignant melanoma of the glans penis and the male urethra. Report of a case and review of the literature. Eur Urol 1990; 18(2): 156 – 9. 107. Radhi JM. Urethral malignant melanoma closely mimicking urothelial carcinoma. J Clin Pathol 1997; 50(3): 250 – 2. 108. Rashid AM, Williams RM, Horton LW. Malignant melanoma of penis and male urethra. Is it a difficult tumor to diagnose? Urology 1993; 41(5): 470 – 1. 109. Sanz Velez JI, et al. Melanoma of female distal urethra: apropos of a case. Actas Urol Esp 1989; 13(1): 63 – 4. 110. Strzyzowski J. Malignant melanoma of the orifice of female urethra. Pol Przegl Chir 1976; 48(2 A): 289 – 90. 111. Sugaya K, et al. A case of amelanotic malignant melanoma of the female urethra. Jpn J Clin Oncol 1983; 13(2): 435 – 9. 112. Urso C, Taddei GL. Melanoma of the lower female urogenital tract (gynecological melanoma). Pathologica 1991; 83(1083): 29 – 34. 113. Begin LR, Deschenes J, Mitmaker B. Pagetoid carcinomatous involvement of the penile urethra in association with high-grade transitional cell carcinoma of the urinary bladder. Arch Pathol Lab Med 1991; 115: 632 – 5. 114. Lee RA, Dahlin DC. Paget’s disease of the vulva with extension into the urethra, bladder, and ureters; a case report. Am J Obstet Gynecol 1981; 140: 834 – 6. 115. Metcalf JS, Lee RS, Maize JC. Epidermotropic urothelial carcinoma involving the glans penis. Arch Dermatol 1985; 121: 532 – 4. 116. Salazar G, Frable WJ. Extramammary Paget’s disease: a case involving the prostatic urethra. Am J Clin Pathol 1969; 52(5): 607 – 12. 117. Leung YL, Lee F, Tam PC. Leiomyoma of female urethra causing acute urinary retention and acute renal failure. J Urol 1997; 158(5): 1911 – 2.
URETHRAL CANCER 118. Greene FL, et al. (eds). Urethra, 6th ed. New York: Springer-Verlag New York, LLC, 2002. 119. Dinney CP, et al. Therapy and prognosis for male anterior urethral carcinoma: an update. Urology 1994; 43(4): 506 – 14. 120. Dalbagni G, et al. Results of high dose rate brachytherapy, anterior pelvic exenteration and external beam radiotherapy for carcinoma of the female urethra. J Urol 2001; 166(5): 1759 – 61. 121. Thyavihally YB, et al. Primary carcinoma of the female urethra: single center experience of 18 cases. Jpn J Clin Oncol 2005; 35(2): 84 – 7. 122. Grigsby PW. Carcinoma of the urethra in women. Int J Radiat Oncol Biol Phys 1998; 41(3): 535 – 41. 123. Dalbagni G, et al. Female urethral carcinoma: an analysis of treatment outcome and a plea for a standardized management strategy. Br J Urol 1998; 82(6): 835 – 41. 124. Milosevic MF, et al. Urethral carcinoma in women: results of treatment with primary radiotherapy. Radiother Oncol 2000; 56(1): 29 – 35. 125. DiMarco DS, et al. Outcome of surgical treatment for primary malignant melanoma of the female urethra. J Urol 2004; 171(2, Pt 1,): 765 – 7. 126. Dann T, et al. Treatment of distal urethral cancer by laser coagulation. Urologe A 1989; 28(5): 296 – 9. 127. Staehler G, et al. The use of neodymium-YAG lasers in urology: indications, technique and critical assessment. J Urol 1985; 134(6): 1155 – 60. 128. Desgrandchamps F, et al. Total extended urethrectomy and transureteral continent cystostomy in the treatment of urethral melanoma in women. Report of a case. Prog Urol 1995; 5(5): 720 – 3. 129. Larson DL, Bracken RB. Use of gracilis musculocutaneous flap in urologic cancer surgery. Urology 1982; 19(2): 148 – 51. 130. Zeidman EJ, Desmond P, Thompson IM. Surgical treatment of carcinoma of the male urethra. Urol Clin North Am 1992; 19(2): 359 – 72. 131. Bracken RB, Henry R, Ordonez N. Primary carcinoma of the male urethra. South Med J 1980; 73(8): 1003 – 5. 132. Hopkins SC, Nag SK, Soloway MS. Primary carcinoma of male urethra. Urology 1984; 23(2): 128 – 33. 133. Mandler JI, Pool TL. Primary carcinoma of the male urethra. J Urol 1966; 96(1): 67 – 72. 134. Ray B, Canto AR, Whitmore WF Jr. Experience with primary carcinoma of the male urethra. J Urol 1977; 117(5): 591 – 4. 135. Micaily B, et al. Brachytherapy for cancer of the female urethra. Semin Surg Oncol 1997; 13(3): 208 – 14. 136. Klein FA, Ali MM, Kersh R. Carcinoma of the female urethra: combined iridium Ir 192 interstitial and external beam radiotherapy. South Med J 1987; 80(9): 1129 – 32. 137. Antoniades J. Radiation therapy in carcinoma of the female urethra. Cancer 1969; 24(1): 70 – 6. 138. Prempree T, Amornmarn R, Patanaphan V. Radiation therapy in primary carcinoma of the female urethra. II. An update on results. Cancer 1984; 54(4): 729 – 33. 139. Weghaupt K, Gerstner GJ, Kucera H. Radiation therapy for primary carcinoma of the female urethra: a survey over 25 years. Gynecol Oncol 1984; 17(1): 58 – 63.
37
140. Kuettel MR, et al. Treatment of female urethral carcinoma in medically inoperable patients using external beam irradiation and high dose rate intracavitary brachytherapy [In Process Citation]. J Urol 1997; 157(5): 1669 – 71. 141. Heysek RV, et al. Carcinoma of the male urethra. J Urol 1985; 134(4): 753 – 5. 142. Eisenberger MA. Chemotherapy for carcinomas of the penis and urethra. Urol Clin North Am 1992; 19(2): 333 – 8. 143. Flam M, et al. 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 1996; 14(9): 2527 – 39. 144. Green JA, et al. Survival and recurrence after concomitant chemotherapy and radiotherapy for cancer of the uterine cervix: a systematic review and meta-analysis. Lancet 2001; 358(9284): 781 – 6. 145. Ryan DP, Compton CC, Mayer RJ. Carcinoma of the anal canal. N Engl J Med 2000; 342(11): 792 – 800. 146. Sternberg CN, et al. M-VAC (methotrexate, vinblastine, doxorubicin and cisplatin) for advanced transitional cell carcinoma of the urothelium. J Urol 1988; 139(3): 461 – 9. 147. von der Maase H, et al. Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic bladder cancer: results of a large, randomized, multinational, multicenter, phase III study. J Clin Oncol 2000; 18(17): 3068 – 77. 148. Kachnic LA, et al. Bladder preservation by combined modality therapy for invasive bladder cancer. J Clin Oncol 1997; 15(3): 1022 – 9. 149. Hussain M, et al. Combination paclitaxel, carboplatin, and gemcitabine is an active treatment for advanced urothelial cancer. J Clin Oncol 2001; 19(9): 2527 – 33. 150. Hussain M, Vaishampayan U, Smith DC. Novel gemcitabinecontaining triplets in the management of urothelial cancer. Semin Oncol 2002; 29(1 Suppl 3): 20 – 4. 151. De Wit R. Overview of bladder cancer trials in the European Organization for Research and Treatment. Cancer 2003; 97(Suppl 8): 2120 – 6. 152. Bosset JF, et al. Enhanced tumorocidal effect of chemotherapy with preoperative radiotherapy for rectal cancer: preliminary results – EORTC 22921. J Clin Oncol 2005; 23(24): 5620 – 7. 153. Sauer R, et al. Preoperative versus Postoperative Chemoradiotherapy for Rectal Cancer. N Engl J Med 2004; 351(17): 1731 – 40. 154. Cunningham D, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 2004; 351(4): 337 – 45. 155. Hurwitz H, et al. Bevacizumab plus Irinotecan, fluorouracil, and Leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350(23): 2335 – 42. 156. Lutz ST, Huang DT. Combined chemoradiotherapy for locally advanced squamous cell carcinoma of the bulbomembranous urethra: a case report. J Urol 1995; 153(5): 1616 – 8. 157. Tran LN, Krieg RM, Szabo RJ. Combination chemotherapy and radiotherapy for a locally advanced squamous cell carcinoma of the urethra: a case report. J Urol 1995; 153(2): 422 – 3.
Section 1 : Genitourinary Cancer
4
Uncommon Cancers of the Prostate Scott T. Tagawa, Omid Hamid, Eila Skinner and Parvesh Kumar
INTRODUCTION The prostate is an exocrine organ present in all mammals, normally weighing 20–25 g in an adult male.1,2 It is derived from the urogenital sinus, with growth directed by dihydrotestosterone during the third month of fetal growth. Epithelial buds invade the mesenchyme to develop into the prostate, forming the various zones. There are three predominant cell types making up prostatic tissue: epithelial cells, basal/stem cells, and neuroendocrine cells. Stromal cells and prostatic tissue matrix are also important parts of the prostatic milieu for normal development and function, as well as malignant processes. Unlike some of the other accessory sexual glands in men, the prostate is frequently involved with hyperplasia and malignancy.1 The most common malignancy of the prostate, by far, is adenocarcinoma. This cancer, the most common in men–other than basal cell and squamous carcinomas of the skin–will affect approximately 232 000 men in the United States in 2005, resulting in over 30 000 deaths.3 However, there are other malignancies affecting this organ that are, as explicitly stated in the title, uncommon. This chapter in this edition of the textbook will combine and update three chapters from the previous edition, as well as add new information. We will review four major uncommon primary cancers of the prostate: small cell undifferentiated carcinoma (SCUC), sarcoma, transitional cell carcinoma (TCC), and lymphoma. In each section, we will review the pathology of the disease, discuss the clinical presentation and diagnostic points, review prognosis and treatment options, and conclude with our recommendations.
SMALL CELL UNDIFFERENTIATED (NEUROENDOCRINE) CARCINOMA OF THE PROSTATE Small cell undifferentiated carcinoma of the prostate (SCUCP) is an entity that has been characterized in detail only relatively recently.4 Historically, there was underreporting of this entity. In 1957, Sommers5 reported in an autopsy study of 109 patients with prostatic carcinoma that 14% had adrenal cortical hyperplasia, suggesting ectopic
production of adrenocorticotropic hormone (ACTH) and that one-third of patients in whom parathyroid tissue was examined had glandular hyperplasia, implying raised levels of immunoreactive parathyroid hormone (PTH)-like material. Until the 1980s, there had been no experimental model for the study of this disease. Our current knowledge is still based predominantly on published data consisting of preclinical studies, small clinical series, and case reports. With increased pathological sophistication and the availability of molecular diagnostic techniques, accompanied by the recognition that SCUCP is associated with a range of neuroendocrine markers, this entity is being recognized more frequently than the traditional incidence figure of 1%. Although this is a relatively uncommon entity, there is now sufficient published experience (in addition to extensive literature from bronchogenic small cell carcinoma) to allow a general plan of management to be defined.
Pathology SCUCP, initially reported in the literature, was associated with clinical syndromes suggesting ectopic hormone production, thus focusing histopathological study on particular morphological and functional features consistent with neuroendocrine differentiation. For more than 40 years, it has been recognized that endocrine –paracrine cells (argentaffin or argyrophil staining) are present in the normal and hyperplastic prostate gland.6 – 8 Just as SCUC of the lung was thought to arise from the so-called bronchial K cells,9 the histologically similar SCUCP was also considered to arise from this type of cell.10 Recent studies have indicated considerable overlap between both pulmonary and prostatic SCUC and their more common epithelial counterparts at both histological and ultrastructural levels.11,12 Indeed, a significant majority of SCUCP have presented as mixed tumors with a prominent epithelial component, or have emerged during or after the treatment of classical prostatic adenocarcinomas. Thus SCUCP and prostatic adenocarcinoma may have a common cell of origin. In approximately 50% of cases of prostate cancer, tumors are a combination of small cell carcinoma and adenocarcinoma.13 Further complicating the issue, carcinoid
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
UNCOMMON CANCERS OF THE PROSTATE
tumors (which are also characterized by neuroendocrine differentiation) also occur in the prostate, either in isolation or with coexistent adenocarcinoma, and may well form part of a continuum with SCUC as there is substantial histological overlap between the two patterns. The pathology of SCUCP reflects these considerations, and the characteristic morphology consists of sheets and nests of uniform cells almost devoid of specific cell-to-cell orientation. At best, there is some focal suggestion of perivascular or peripheral palisading along the epithelial–stromal interfaces, but more typically, no such arrangement is apparent. Microscopic or larger foci of tumor necrosis are prominent, as is the “streaming effect” of hematoxyphilic debris. The tumor infiltrates widely and diffusely in a lymphoma-like fashion with poorly circumscribed margins at the advancing edge. Lymphatic and blood vessel permeation are usually evident, and small tumor emboli may be seen at a distance from the infiltrating edge of the carcinoma. The stroma between the
39
tumor islands is immature and fibroblastic. Remnants of normal prostatic tissue may be seen, entrapped, or encircled by the carcinoma. Figure 1 represents a transrectal ultrasound (TRUS) guided biopsy of SCUCP. Cytologically, the small uniform cells have rounded hyperchromatic nuclei, coarse chromatin, and usually unapparent nucleoli. Mitotic figures are numerous, occurring at a rate of 5 to 10 per high-power field in most regions. The cytoplasm is scanty and, in many areas, the nuclei appear “naked”. Mucin stains are negative, while argyrophil granules (Grimelius technique) are variably present. Argentaffin granules are usually not observed (Fontana –Masson method). In the immunoperoxidase technique, the profile of tumor markers includes positive staining for neuroendocrine elements (neuron-specific enolase (NSE), bombesin, and chromogranin) and various polypeptide hormones (including ACTH, antidiuretic hormone (ADH), and corticotropin-releasing factor (CRF)), and negative staining
(a)
(b)
(c)
(d)
Figure 1 Low (a) and high-power (b) views of SCUCP, featuring atypical nuclear molding with increased nuclear: cytoplasmic ratios, coarse chromatin, and irregular nuclear membranes. Low (c) and high-power (d) views of prostatic sarcoma, featuring highly atypical elongated spindle cells characterized by anisocytosis, anisonucleosis, irregular nuclear contour, high nuclear: cytoplasmic ratio, presence of mitotic figures, and necrosis consistent with undifferentiated sarcoma (Courtesy of Rashida Soni, MD. University of Southern California – Kenneth Norris Jr Comprehensive Cancer Center).
40
GENITOURINARY CANCER
for classical markers of prostatic glandular differentiation (prostate-specific antigen (PSA) and prostatic acid phosphatase (PAcP)). The electron microscopic features of SCUCP are similar to those of SCUC occurring at other sites, and include typical small (140–250 nm), roughly circular neurosecretory dense-core granules.14,15 Small, well-formed desmosomes have been described in at least one clinical study,12 and this observation has been confirmed in xenograft studies. Most cases of SCUCP are associated with concurrent or antecedent prostatic adenocarcinoma (Table 1). In those tumors with a mixed pattern, the adenocarcinoma is of the moderately to poorly differentiated microacinar or cribriform type, with intermingling of tumor elements and transitional areas. We regard identification of an adenocarcinomatous component as an important confirmation that the tumor has arisen in the prostate, although it is not an absolute requirement for the diagnosis of SCUCP. In many instances, such confirmation has not been possible before autopsy has been performed, and hence the definitive diagnosis of
SCUCP has often not been achieved sufficiently early in the clinical course to be of benefit to the patient. It has been hypothesized that the neuroendocrine cells are admixed with adenocarcinoma and emerge as a hormonerefractory carcinoma after the androgen blockade. The SCUCP shows a spectrum ranging from a mixed adenocarcinoma with a neuroendocrine component to the extreme case of pure neuroendocrine cells. In those tumors with mixed elements of SCUCP and adenocarcinoma, the latter component usually retains immunoreactivity for PSA and PAcP in large measures. In as many as one-third of classical prostatic carcinomas, spanning the entire histological spectrum, occasional cells may contain argyrophil granules and will show immunostaining suggestive of neuroendocrine differentiation.26,27 In patients with a pure neuroendocrine tumor it is important to exclude the possibility that the lesion represents metastasis from a bronchogenic tumor.28,29 This can be a particular problem in the case of metastatic SCUCP with mediastinal lymph node involvement, although this is an
Table 1 Small cell undifferentiated carcinoma of the prostate clinical details.a
Presence of AdCa
Expression of PSA
Ectopic hormone production
Local involvement
Liver metastases
56 58 63 57 66 62 76 80 67 70 61
Yes Yes Yes Yes Yes Yes? Yes Yes Yes Yes No
? Yes ? ? No No ? Yes Yes Yes No
Yes Yes Yes Yes Yes Yes Yes No No Yes No
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
No No Yes Yes Yes No No No Yes No Yes
73
Yes
No
Yes
Yes
60 72 68 65 30 59 61 68 62 71 73 71 57 72 69 89 49 66 80 53 70 61
No Yes Yes Yes No Yes No Yes No Yes No Yes Yes Yes No Yes No No No No No No
No Yes ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
No No ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Age
Other sites of disease
Response to hormone therapy
Response to chemotherapy
B PE, Ad, B No B No No No No B B N, B
Yes Yes Yes NA Yes NA NA Yes No Yes No
NA NA NA NA NA NA NA No ? ? Yes
No
B, PE
No
NA
Yes No No Yes No Yes Yes No Yes Yes No Yes No No Yes No No No Yes No No No
No B N B Lu B, N, Lu N, Lu, O N, Lu N N, Lu, B N, Pa Lu, B N, O B No Lu, Pl N N Ad N, B Lu, B, V B
NA No ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
NA NA ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Sites at autopsy B, L, Pr, Sp, N ? Pr, L L, Lu, B L, Sp, GB, N Pr, Bl, N Pr Pr, N, B Lu, N, Pe Pr, N, B, L Lu, L, Sp, Ad, M, N, Br Pr, B, N, Pl, Pe, St, L Pr, L, N, Ad Br, L, Lu, Bl ? ? ? ? B ? ? B ? ? ? ? ? Lu-SA ? ? ? ? ? ?
References 16 16 17 18 19 14 20 12 12 12 21 22 23 24 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25 15,25
AdCa, adenocarcinoma; PSA, prostate-specific antigen; B, Bone; PE, pleural effusion; Ad, adrenals; N, nodes; Lu, lung; O, omentum; Pa, pancreas; Pl, pleura; V, vocal cord; L, liver; Pr, prostate; Sp, spleen; GB, gall bladder; Bl, bladder; M, bone marrow; St, stomach; Lu-SA, lung – spindle cell sarcoma; NA, not applicable. a Adapted from Raghavan D, Russell P. In Raghavan D et al. (eds) Textbook of Uncommon Cancer, 2nd ed. New York, NY: Wiley & Sons, 1999.
UNCOMMON CANCERS OF THE PROSTATE
extremely unusual clinical scenario. This can usually be excluded on clinical grounds, especially with the increased precision of pulmonary imaging technologies. Sometimes bronchoscopy is required to exclude the presence of an endobronchial primary tumor. Immunohistochemistry with markers such as thyroid transcription factor-1 (to indicate a lung primary) may sometimes be helpful in the clinical context. Expression of Tumor Markers
Polypeptide hormones constitute the major biochemical markers of SCUCP, exhibiting features characteristic of the endocrine –paracrine cells described above (so-called amine precursor uptake and decarboxylation or “APUD” cells). The production of these hormones is associated morphologically with argyrophil staining or, at an ultrastructural level, with the presence of the neurosecretory granules. A range of peptide hormones is produced, including the chromogranins, calcitonin, an immunoreactive PTH-like substance, and a thyroid-stimulating hormone (TSH)-like substance. However, even in classical adenocarcinoma of the prostate, hypercalcemia has been reported, which may be due to the production of an immunoreactive PTH-like substance.30 Further evidence for the overlap between classical adenocarcinoma and SCUCP has been provided by Pruneri et al.,31 who immunohistochemically documented the production of chromogranin A, chromogranin B, and secretogranin II in more than half of a series of 64 patients with classical adenocarcinoma of the prostate. They identified a correlation between these neurocrine hormones and the areas of poor differentiation, although in this small series they did not find obvious correlations between the expression of these peptides and the ultimate prognosis. To date, ACTH appears to be the most common ectopic hormone in SCUCP, usually identified only on the basis of biochemical, immunohistochemical, or endocrinological tests,14,18,32,33 although occasionally presenting with a variant of Cushing’s syndrome. ADH production has also been described in association with SCUCP,22,34 and there has been a case report of metastatic SCUCP that elaborated corticotropin.35 A xenografted SCUCP cell line from Raghavan et al. has been shown to produce low levels of bombesinlike immunoreactivity and of somatostatin.36 Neuron-specific enolase (NSE), an enzyme of anaerobic glycolysis that is present in neuroendocrine cells, has been extensively reported as a marker of SCUC of the lung. Its expression has been found to vary with both the proliferative state and the level of oxygenation of bronchogenic SCUC,37 and we do not regard it as a useful clinical marker for this tumor phenotype. There is discordance of expression between NSE and the classical prostate markers. For example, Ghandur-Mnaymneh et al.22 reported that NSE was expressed in pure SCUCP, none of which stained for PSA or PAcP, an observation confirmed by Ro et al.15 Conversely, Ro et al. have described one case of SCUCP that stained for PSA and PAcP but did not express NSE.15 In mixed tumors, the adenocarcinomatous components do not stain for NSE whereas 50% of tumors express this antigen in regions of SCUC differentiation.15,22
41
Other tumor markers have been reported in cases of SCUCP, although their clinical utility appears limited. For example, carcinoembryonic antigen (CEA) has been demonstrated in up to 25% of classical prostate cancers,38,39 but is not a specific marker. Its expression has been correlated with decreasing differentiation of the adenocarcinoma cells.38,40 In a detailed xenograft study, the authors of the previous chapter have shown marked heterogeneity of expression of CEA (representing about 25% of cells), in contrast to nearly uniform immunohistochemical staining of cells for NSE and epithelial membrane antigen (EMA).36,41 This further illustrates the limitations of CEA as a potential marker of SCUCP. Another marker, calcitonin, was evaluated retrospectively in 16 patients with widely metastatic small cell carcinoma of the prostate (5 pure SCUCP and 11 combined with adenocarcinoma). Nine of 16 (56%) had elevated serum calcitonin level (range 42–2654 pg mL−1 ), which chemically supported the diagnosis of SCUCP. Survival analysis by logrank test did not show a statistically significant prognostic value of serum calcitonin. However, patients with extremely high serum calcitonin level tended to have poor survival.42 EMA is expressed in normal prostate cells43 and is seen focally in up to 80% of prostate adenocarcinomas, although its presence does not correlate with histological grade.38,39 It is not found on cells of normal endocrine tissue or in classical neuroendocrine tumors, but has been reported to be expressed by certain SCUC of the lung.43 As noted previously, although PSA and PAcP are not usually expressed in foci of SCUCP, they may be detected immunohistochemically in coexistent regions of adenocarcinoma. By contrast, these markers have been reported in both the adenocarcinomatous and carcinoid components of prostatic carcinoid tumors.44 – 46 However, it is also possible that this distribution may be inaccurate, reflecting the artifacts of the immunohistochemical techniques used in the early 1980s, before the availability of monoclonal antibodies and some of the sophisticated controls that are routinely applied today. Cases of SCUCP have been described in which no ectopic hormone production has been detected.12,21,47 Furthermore, in several cases in which ectopic hormone production is thought to have occurred, the diagnosis has merely been presumptive, on the basis of biochemical changes (e.g. hypokalemic alkalosis) or autopsy data (adrenal hyperplasia).5,16,17 Clinical Application of Tumor Markers
In clinical practice, the role of these markers in the serum and in tissue has been restricted predominantly to diagnosis, and in particular to the more detailed characterization of undifferentiated tumors. The demonstration of increased serum levels of NSE or chromogranin may suggest the presence of occult elements of neuroendocrine differentiation in a patient thought to suffer from a classical high-grade adenocarcinoma of the prostate. Because early recognition of this clinical entity seems important (see below), one should have a low threshold for checking levels of these better-defined markers, which may indicate the presence of clinically important SCUC components. However, in contrast
42
GENITOURINARY CANCER
to the serial measurement of blood levels of α-fetoprotein and human chorionic gonadotropin in the management of germ cell and trophoblastic tumors, CEA and NSE appear to have little role in monitoring the clinical course of SCUCP as they often correlate poorly with the changing level of tumor burden. Angelsen et al.48 have studied the serum and tissue expression of neuroendocrine markers in a series of 22 cases of prostate cancer, with tissue specimens obtained by transurethral resection (and thus subject to some geographical selection bias). They correlated tissue and blood levels of PSA, NSE, chromogranin A, and chromogranin B, and found significant discordance of expression. In their study, none of the patients had elevated blood levels of NSE, despite the presence of NSE-positive tumor cells in 77% of the tumors. However, a positive correlation was identified between the number of chromogranin A staining cells and the serum values of chromogranin A, although the number of cases were relatively small.48 Other investigators have suggested that chromogranin A and/or NSE may be markers of prognosis in patients with adenocarcinoma of the prostate.49 – 52 However, clearly more prospective data is needed before clinical decisions should be made on this basis alone. The potential role of monitoring ectopic hormone levels has not yet been defined in this context. It is not clear whether their presence (or their corresponding clinical syndromes) acts as an adverse prognostic determinant per se, although several cases of SCUCP with ectopic hormone production have been characterized by short survival. However, this may simply be a function of tumor volume, with larger tumors producing sufficient quantities of these hormones to allow detection in the circulation. DNA Content and Cytogenetic Abnormalities
Chromosomal analysis of cell lines derived from prostatic adenocarcinomas has shown them to be mostly triploid or tetraploid, with one showing a pseudodiploid profile, as reviewed elsewhere.53 Direct karyotyping has been recorded for primary carcinomas and bone marrow metastases,54,55 with the most common abnormalities described in these studies including deletions of the long arm of chromosome 10, chromosome 7, structural rearrangements of chromosome 1, and loss or rearrangement of the Y chromosome. These chromosomal alterations did not correlate with the level of histological differentiation of the tumors. Studies of the xenografted line of SCUCP have revealed a hypodiploid karyotype with several structural alterations.56 The cytogenetic characteristics of bronchogenic SCUC57,58 differ from those of typical prostatic adenocarcinoma. Marked chromosomal heterogeneity has been reported in the bronchogenic tumors, with a wide range of consistent abnormalities, including deletion of a part of the short arm of chromosome 3, the presence of markers associated with chromosomes 1, 6, and 11, and the detection of minute and double minute chromosomes. Although the chromosome 1 markers were of a variable structure, the markers of chromosomes 6 and 11 frequently took the form of 6q− and 11p+.57 However, as noted subsequently, there is an interesting similarity
of the 1p+ marker chromosome in a model of SCUCP with that shown in a cell line of SCUC of the lung.56 Oncogenes
A detailed review of the expression of oncogenes, their possible mutation, rearrangement, or amplification in cancer tissue is beyond the scope of this review. To date, we are not aware of any studies of the expression of oncogenes in SCUCP. However, it is worth noting that some data are available with respect to typical prostatic adenocarcinoma. The ras oncogene is expressed in prostate cancer tissue, correlating directly with the degree of nuclear anaplasia and inversely with the level of glandular differentiation.59 Large amounts of ras and myc mRNA have been demonstrated in poorly and moderately differentiated prostatic cancer cell lines.60 Of interest, expression of c-myc correlates with suppression of c-kit function in SCUC of the lung.61 Whether similar functions operate in SCUCP has not yet been defined, notwithstanding the high levels of c-myc expression demonstrated in prostatic adenocarcinoma.60 Raghavan et al. contributed much to our understanding of the disease by establishing the UCRU-PR-2 xenograft line from a confirmed primary prostatic SCUCP and characterized it in serial passages in nude mice and in tissue culture.24,36,41 Immunohistochemical studies of xenografted tissues revealed heavy staining for CEA and NSE, and focal areas of expression of EMA. By contrast, PAcP, PSA, and ACTH were not detected by these methods,24 although radioimmunoassay revealed the production of ACTH by all tumor fragments tested and by cells in tissue culture.36 Calcitonin was not detected, but bombesin-like immunoreactivity was identified in low levels in some samples. Androgen and estrogen receptors were negative.24,36 Despite negative immunohistochemistry, the serum of tumor-bearing nude mice contained human PAcP, perhaps reflecting the high tumor:host mass ratio characteristic of the xenograft model.62 This discrepancy suggests that, if clinically relevant and technically feasible, more accurate assessment of hormone and enzyme content of SCUCP may be obtained by methods more sensitive than the more convenient immunohistochemistry. Flow cytometry showed UCRU-PR-2 to be consistently diploid.24 The karyotype was hypodiploid with nonrandom losses of chromosomes 6, 7, 10, and 13, and with structural rearrangements of chromosomes 1 and 2,56 demonstrating the characteristics of prostatic adenocarcinoma (deletion of chromosome 10) and a similarity of the 1p+ chromosome that has been reported in some samples of bronchogenic SCUC. They postulated that this line might exhibit stem cell function with variable differentiation, which could be elicited by implantation of tumors at different physical sites, but were unable to demonstrate any such changes in differentiation in tumors growing at different sites.41 Others have also developed xenografts and cell lines of neuroendocrine prostate cancers, which have retained the characteristic features (histology, neurosecretory granules, chromogranin expression) and an absence of estrogen and androgen receptor expression.63 These lines have been used to study the relationship between hormone suppression and outgrowth of neuroendocrine cells, and appear to suggest
UNCOMMON CANCERS OF THE PROSTATE
that the change in histology reflects an induction of neuroendocrine differentiation by hormonal deprivation.63 In addition, a transgenic mouse model of prostate cancer may lend itself to the detailed characterization of the histogenesis of SCUCP.64
Histogenesis To date, the histogenesis of SCUCP has been controversial. Different theories exist regarding its histogenesis. Neuroendocrine cells are scattered throughout the prostate gland and prostatic epithelium, including ducts and acini.65 Although neuroendocrine cells are a constant feature of the prostatic ducts and periurethral region, they seem to disappear from normal peripheral prostate tissue during childhood, and return at puberty, although the basis of this fluctuation is not known.66 It was suggested that SCUCP is derived from the neural crest line/APUD cell system.67 This was not confirmed by the embryological studies that showed an endodermal origin.68 Another common theory is the progression of typical adenocarcinoma to SCUCP as a product of final dedifferentiation according to the model of divergent differentiation.69 A third theory presents a direct stem cell origin for SCUCP based on the lack of immunohistological staining for PSA, lack of androgen receptor positivity, and high MIB-1 labeling.70 In the majority of reported cases, the identification of neuroendocrine elements is preceded by adenocarcinoma, often involving an interval of several years. Whether the production of polypeptide hormones is attributed to an endodermal or neuroectodermal origin, this index of functional differentiation is most easily explained by origin from a pluripotential stem cell. Whatever the histogenesis, it is clear that SCUCP is often intimately associated with elements of adenocarcinoma. Thus, the formulation of a plan of management, both diagnostic and therapeutic, should take into account the presence of tumor cells with a wide range of potential functional characteristics. Furthermore, the paradigms of management of classical adenocarcinoma of the prostate may require modification, as it appears likely that SCUCP has previously been underdiagnosed, and may well account for a proportion of “resistant” prostatic adenocarcinomas.
Clinical Presentation Extrapulmonary small cell carcinomas have presented a diagnostic and therapeutic challenge to oncologists since their first description by Duguid and Kennedy in 1930.71 Although sometimes misdiagnosed as metastatic small cell lung carcinoma, small cell anaplastic carcinomas share many of the same characteristics; they are extremely aggressive, metastasize early and often, and do not respond to many traditional chemotherapeutic regimens. Estimated to comprise only 0.1 to 0.4% of all common forms of cancer and 2 to 4% of all small cell carcinoma,72 rarity adds to difficulties in their diagnosis and treatment. The malignant capacity of SCUCP leads to metastasis more often to soft tissues, lung, liver, and lymph nodes than to bone. Dauge and Delmas found the degree of aggressive behavior directly proportional to the extent of
43
neuroendocrine differentiation.73 Metastasis to rare sites such as the pericardium has also been reported. In contrast to adenocarcinoma of the prostate, SCUCP can present with metastatic disease to the bone that is purely lytic. Patients with SCUCP more frequently present with metastatic disease, tend to have a greater proportion of brain metastases, and are younger than patients with prostatic adenocarcinoma. The aggressive nature of SCC is evident by a median survival of 5.2 months after primary diagnosis. Concomitant adenocarcinoma can be present elsewhere in the prostate gland of patients with SCUCP, and serum PSA and PAcP levels may be elevated in this disease. In the past, the prospective clinical diagnosis of SCUCP was rarely made, with the majority of cases diagnosed either or late in the course of the disease at autopsy. More recently, there has been an increasing level of prospective histological recognition of this entity, thus presenting the cases for management at a point at which treatment decisions can affect outcome. As shown in Table 1, there are many similarities to the presentation of classical adenocarcinoma of prostate, whether antecedent or concomitant adenocarcinoma is present. The population of patients consists of males, predominantly in the age range 60–80 years, often with features of local and distant involvement.12,14 – 16,18 – 21,23,53,74 – 76 The presenting symptoms include urinary outflow obstruction, frequency, perineal pain or discomfort, nocturia, hematuria, and occasionally, rectal symptoms due to the effects of the enlarged prostate. A common feature that distinguishes SCUCP from classical adenocarcinoma clinically is the paucity of osseous involvement. Although bone metastases may occur with SCUCP, it is more common for this entity to demonstrate a predominance of lymph node, pulmonary, hepatic, brain, and soft tissue involvement. Many patients present with neurological symptoms in association with unrecognized brain metastasis, rarely seen in adenocarcinoma of the prostate.25 McCutcheon et al. reviewed 38 patients with antemortem intracerebral metastasis found on a review of 7994 prostate cancer patients treated over an 18-year period, noting that SCC comprised 0.5% of the patient cohort but 26% of patients with brain metastasis.77 All patients with SCUCP originally presented with stage D disease in comparison to only half of the adenocarcinoma patients; they tended to have greater numbers of brain metastases, and were younger than patients with adenocarcinoma. Surprisingly, patients with brain metastases of pure adenocarcinoma had lower mean survival even though they originally presented at lower stage of disease than those with SCUCP. Paraneoplastic syndromes occur in 10% of patients with SCC of the prostate. These syndromes include thyrotoxicosis, inappropriate ADH production, hypercalcemia, and adrenal hyperfunction. Ectopic production of thyroxine causing thyrotoxicosis and ultimately death has been reported.78 In these instances, biochemical disturbances such as hypokalemic alkalosis or hyponatremia have required treatment if recognized in time. In one case, the presenting feature of hypokalemic alkalosis was an acute psychosis,17 although the usual metabolic abnormalities are more classically associated
44
GENITOURINARY CANCER
with clouding of consciousness, constipation, anorexia, and malaise.
Diagnosis Perhaps the most important prerequisite for the correct diagnosis of SCUCP is initial awareness of the entity. In the patient with an unusual presentation of prostate cancer, or with the presence of an apparent paraneoplastic syndrome, the appropriate physical examination and biochemical tests, such as measurement of NSE, chromogranin A, calcitonin, or ACTH may help establish the diagnosis. Electrolyte abnormalities, including hyponatremia, and the presence of hypercalcemia may suggest the diagnosis. It should be noted that patients with ectopic ACTH production may not show classical Cushing’s syndrome, and may appear cachexic and pigmented. PSA level should also be measured in the blood as it may reflect the presence of a component of adenocarcinoma. In fact, from xenograft studies, it appears likely that SCUCP itself may release low levels of PAcP.24 The demonstration of modest elevations of PSA level, in the presence of a large tumor load with extensive metastases, should raise the suspicion of SCUCP as the diagnosis. In recent times, we have tended not to routinely measure PAcP level, although occasionally its presence will increase the diagnostic suspicion of prostatic origin of an undifferentiated tumor. Complete staging of this tumor is required, along with a bone scan and computed tomography (CT) scan of the chest, abdomen, and pelvis with intravenous and oral contrast. The presence of lytic bone metastases will increase suspicion of the presence of metastatic neuroendocrine elements. Magnetic resonance imaging (MRI) scans may increase the level of definition of the prostatic primary tumor and local extension, but has no clear advantage over CT scan. The usefulness of positron emission tomography (PET) has been described for small cell lung cancer,79,80 but has not been evaluated in SCUCP. However, based upon limited information available in small cell carcinoma of other sites,81,82 as well as proven uptake in small cell lung cancer (SCLC), PET may be of use in the staging of SCUCP, particularly in suspected limited stage disease. If definitive local therapy is planned, or in the presence of neurological symptoms, an MRI scan of the brain and occasionally a bone marrow aspiration and biopsy should also be considered before defining the plan of management.
Treatment Untreated SCUCP may be rapidly progressive and fatal. However, despite its aggressiveness and grave prognosis, SCUCP is exquisitely sensitive to cytotoxic chemotherapy and radiation. Early introduction of chemotherapy may afford a survival advantage. Effective palliation can be achieved, and in some series treatment has achieved a 2-year survival of 20% of patients.78 Considered rare, transient remissions have been seen in patients who have received multimodality therapy with chemotherapy, surgery, and radiation. In view of the infrequent early diagnosis of this condition, it is not possible to define the single “optimal” approach
to the management of localized SCUCP.53,74 However, in small cell cancers of other sites, including the lung and urinary bladder, primary surgery as monotherapy has had very poor results and has largely been abandoned. Integral to any approach is the early diagnosis of SCUCP and the definition of the extent of disease, as outlined above. Patients with SCUCP that appears to be localized to the prostate are highly likely to have occult metastatic disease. Thus a multimodal approach to a patient who is otherwise healthy is logical. Such an approach may include both systemic therapy with cytotoxic chemotherapy (with or without hormone therapy) and possibly local therapy with external beam radiation or radical prostatectomy. Possible approaches include: 1. Radical radiotherapy, in an attempt to achieve local control and perhaps cure. 2. Radical prostatectomy, to control the adenocarcinomatous elements and to reduce the bulk of the primary tumor (less relevant if the patient has previously presented with adenocarcinoma of the prostate, with the current syndrome representing SCUCP). 3. Concurrent or sequential combination cytotoxic chemotherapy, administered to control both local disease and systemic micrometastases in a fashion analogous to that employed for bronchogenic SCUC.83 Local therapy is generally reserved for the healthy patient with apparently localized disease, and should usually be preceded by systemic chemotherapy. Clinical studies in bladder SCC have suggested the superiority of this approach to one beginning with surgery.84 Median cancer-specific survival (CSS) for resectable patients treated with initial cystectomy was 23 months, with 36% disease-free at 5 years (similarly poor survival as patients treated with primary surgery in our own institution85 ). Patients receiving neoadjuvant chemotherapy did not reach the median CSS (p = 0.026) with CSS at 5 years of 78%. No cancer-related deaths were observed beyond 2 years, though no similar comparison has been made for SCUCP. Unfortunately, many cases of SCUCP are not recognized on biopsy or clinical presentation, but are rather diagnosed on the radical prostatectomy specimen. In such cases adjuvant chemotherapy will generally be offered, based on the experience in SCC of the lung. With regard to the optimal dose of radical radiotherapy, there is controversy as to whether a lower dose (40–50 Gy over 4–5 weeks) is appropriate (based on the marked radiosensitivity of SCUC at other sites), or whether a more conventional, higher dose (75.6+ Gy at 1.8 Gy per fraction) should be applied, analogous to prostatic adenocarcinoma. Just as in the treatment of adenocarcinoma of the prostate, definitive conformal radiation therapy using either 3-D conformal radiotherapy (CRT) or intensity modulated radiation therapy (IMRT) should be used to limit toxicity to surrounding critical organs (i.e. rectum and bladder). We believe that the latter approach should be used in view of the potential for admixture of elements of SCUC and adenocarcinoma in these tumors. No controlled studies of adjuvant cytotoxic chemotherapy have been carried out in cases of localized SCUCP, and
UNCOMMON CANCERS OF THE PROSTATE
it is not possible to make any specific recommendations at present. In view of the rarity of the disease, it is unlikely that the appropriate randomized studies will be completed. Thus, any decision regarding the use of adjuvant cytotoxic chemotherapy should be influenced substantially by considerations such as the age and general fitness of the patient, geographical accessibility, available facilities, and patient preferences. There is only a relatively scanty literature regarding the management of extensive SCUCP with chemotherapy. The cytotoxic agents that are active against bronchogenic SCUC, including cyclophosphamide, vincristine, doxorubicin, cisplatin, carboplatin, etoposide, and paclitaxel86 produce objective responses in metastatic SCUCP, although specific objective response rates have not been defined clearly for single agents.21,53,74 The use of three- or four-drug combination regimens yields objective response rates of 60–75% in SCUC of the lung.86,87 The available literature suggests that the response to chemotherapy for SCUCP may be lower than that for bronchogenic tumors, perhaps in view of the heterogeneity of SCUCP with the admixture of elements of adenocarcinoma. Galanis et al. retrospectively reviewed 81 cases of extrapulmonary small cell carcinoma treated between 1974 and 1994 at a single institution.88 Tumor sites included the head and neck, genitourinary, gastrointestinal, and gynecological systems. Among patients treated initially with surgical resection, 75% (30/40) relapsed with a median disease-free survival of 6 months despite adjuvant chemotherapy and/or radiotherapy. The majority of recurrences (80%) occurred in distant sites with median survival of 18 months. Fiveyear overall survival was reported in 13%. The best predictor of long-term disease-free survival after local therapy was disease localized to primary organ at presentation. For patients with localized disease, chemotherapy with radiotherapy showed equal efficacy to surgery. Metastatic disease had a response rate of 73% to platinum-based therapies with a medium duration of response of 8.5 months. The authors concluded that surgery can be curative in organ-confined disease but due to the high rate of systemic recurrence and the chemoresponsiveness of the tumor, platinum-based adjuvant regimens must be considered. In a single institution phase II trial of combination therapy consisting of doxorubicin, etoposide, and cisplatin in patients with SCUCP, Papandreou questioned the belief in threeor four-drug regimens. When 38 patients were treated with this regimen, a 61% response rate was seen with no complete responses. While toxicity was substantially greater than historical cisplatin-etoposide regimens, notwithstanding the serious problems of historical controls, median time to progression of 5.8 months and median survival of 10.5 months failed to improve upon the historical outcome.89 Although there is no specific evidence to support improved survival, chemotherapy can provide palliation, and complete responses have been reported.21 We have usually treated such cases with etoposide and carboplatin or cisplatin, using relatively conventional doses,83,87 and anticipate a response rate of 50–60%. A common approach has been to use etoposide at a dose of 360–500 mg m−2 , delivered
45
over 3–5 days per cycle, with cisplatin at a total dose of 60–100 mg m−2 or carboplatin at an AUC dose of 5.0–6.0, each cycle being repeated every 21–28 days. Although newer agents, such as gemcitabine, taxanes, and irinotecan appear to have a role in the management of bronchogenic SCUC, their role has not yet been defined for SCUCP. There are no data for the use of targeted biologic agents. Much more complex and controversial is the decision regarding the role for initial hormonal therapy. Where the patient has a histological diagnosis of concurrent adenocarcinoma and SCUCP or high circulating blood level of PSA, it is worth considering the role of castration as part of initial treatment. We usually use the gonadotropin releasing hormone (GnRH) agonists in this situation, as they are potentially reversible if there is no evidence of hormone response by the tumor. The role of hormonal manipulation should not be trivialized. In several cases summarized in Table 1, sustained objective initial remission was achieved by bilateral orchiectomy or by the use of systemic estrogens (without cytotoxic chemotherapy). As many of these cases of SCUCP were diagnosed post hoc, it is not clear whether the impact of hormonal manipulation at initial presentation was directed toward pure elements of adenocarcinoma or whether SCUCP itself truly responded to the hormonal therapy. It should also be emphasized that patients treated initially by hormonal manipulation alone for SCUCP should be monitored very closely. Some of the available case reports describe rapid deterioration and death within a few weeks after the institution of hormonal treatment. If a patient with SCUCP fails to respond to hormonal manipulation, cytotoxic chemotherapy should be applied. In the management of prostatic adenocarcinoma, we usually allow a minimum of 2 months before assessing the response to hormonal manipulation (unless the patient is rapidly deteriorating). By contrast, the patient with SCUCP should be reviewed within 2–4 weeks, depending upon the extent and severity of the disease. On an empirical basis, we will often begin treatment with cytotoxic chemotherapy simultaneously with initiation of hormonal therapy or without it at all if there is a high index of suspicion (or histological confirmation) of SCUCP. It is again emphasized that these principles are derived from our clinical practice and explicit empiricism, rather than representing the results of structured, randomized trials. The specific implication of the diagnosis of ‘carcinoid’ of the prostate is not completely clear, although there is a general consensus that classical prostate carcinoids are less aggressive in their natural history than SCUCP.19,76 If the diagnostic distinction can be made between prostatic carcinoid and SCUCP, analogous to the situation that applies to tumors of the thorax, there is a greater tendency to apply localized treatment (in particular, surgical resection) for prostatic carcinoids. In carcinoids at other sites, cytotoxic chemotherapy appears to have a substantially lesser role than for SCUC, and thus we have been less inclined to use systemic chemotherapy for prostatic carcinoids without evidence of spread. In the workup of such tumors of the prostate, octreotide scanning can elucidate sites of
46
GENITOURINARY CANCER
metastatic disease. Somatostatin analogs may be useful in controlling systemic carcinoid symptoms and have shown antiproliferative actions on human prostate cancer cell line LNCaP.90 Therapy with somatostatin analogs has been shown to inhibit tumor growth in vitro with evaluable responses seen in patients.91
Prognosis Traditionally, the prognosis for SCUCP has been dismal. Although the majority of patients in the literature have survived less than 6 months from the time of documentation of SCUCP, it should be noted that most of them did not receive cytotoxic therapy and have been diagnosed with advanced disease. More recently, with the earlier recognition of SCUCP, we have seen prolonged survival among patients who have been treated aggressively with definitive local treatment combined with systemic chemotherapy.53,74 In a retrospective review of 180 patients with genitourinary small cell carcinoma,92 primary surgical therapy was the only parameter that predicted survival in SCUCP. Median survival was 7 months in the 60 patients reviewed. The neuroendocrine component of these carcinomas was resistant to antiandrogen therapy, possibly because of lack of estrogen and testosterone receptors. Hormonal manipulation and systemic chemotherapy did not have a significant impact on the natural history of this disease. Patient and tumor characteristics were not survival determinants on subgroup analysis. No further conclusions could be made because of the nature of the study. Given the rarity of SCUCP it is unclear whether the late diagnosis of this disease is ultimately responsible for the poor outcome of its patients or whether the presence of neuroendocrine elements per se confers a worse survival, with conflicting data from different series.25,93 The definition of median survival is complicated by the fact that several of the reported cases have apparently had biphasic tumors with elements of adenocarcinoma and SCUC. Thus some cases with prolonged survival may have reflected the sustained initial responses to hormonal manipulation because of an initial dominance of adenocarcinoma. The rapid decline after documentation of SCUCP may have reflected a selection process because of (i) the outgrowth of SCUC after initial treatment, (ii) failure to apply the most appropriate treatment for a SCUC, or (iii) an altered biological state of the tumor, consequent upon initial hormonal therapy.
Conclusions and Recommendations The biology of SCUCP is still incompletely understood, although our evolving recognition of the mechanisms of cell cycle regulation and gene control is providing insights into the disease. Morphologically and functionally it bears a closer resemblance to bronchogenic SCUC than to classical prostatic adenocarcinoma. However, in many patients, there is also a substantial overlap with the characteristics of adenocarcinoma, either due to the presence of concurrent elements of this histological subtype or as an inherent biological function of the SCUC cell. The use of multimodality therapy is considered to be standard although no clear consensus exists
on the ideal approach for localized or locally advanced disease. The abundance of case reports, heterogeneity of data, and failure of clinical diagnosis of this entity have led to difficulty in determining its optimal treatment. There is a clear consensus only on the belief that the prognosis is poor with this histology. Although rare, one of the most important aspects of clinical management is the need to consider this entity early in the differential diagnosis of a poorly differentiated prostatic neoplasm as it appears that early recognition, with the concomitant use of cytotoxic chemotherapy is associated with the potential for long-term survival and even with the possibility of ultimate cure. Given the above information, strong consideration should be given to a referral to an experienced center with multidisciplinary care. Even in the setting of localized disease, systemic therapy should be considered. In all cases, if available, participation in a clinical trial is suggested.
SARCOMAS OF THE ADULT PROSTATE When compared with other malignancies, sarcomas are uncommon. In the United States, an estimated 9420 new cases of soft tissue sarcoma will be diagnosed in 2005.3 Worldwide estimates are more difficult to make. The World Health Organization (WHO) tracks cancer statistics by anatomical site (not by histology). Since soft tissue sarcomas may occur in almost any anatomical site, specific numbers for sarcomas are unknown. However, based on extrapolation of estimates from several countries, the worldwide incidence is approximately 30 cases per million in the population (International Agency for Research on Cancer (IARC) data). Primary prostatic occurrences of this relatively uncommon cancer are even more rare, and only case reports and small retrospective series exist. Although cancer of the prostate is one of the most common malignancies, the fraction of these that are sarcomas is exceedingly rare. During a period in which 31 882 deaths from prostatic malignancy occurred, 35 cases of prostatic sarcoma were identified.94 Schmidt presented a single institution’s experience over the 40-year period from 1933 to 1973 in which 12 cases of prostatic sarcoma were identified among over 5000 cases of adenocarcinoma of the prostate (0.24%).95 Investigators from Memorial Sloan-Kettering Cancer Center reported that 2.7% of soft tissue sarcomas seen over a 7-year period arose from the genitourinary tract, but only 10 of 1583 (0.6%) arose from the prostate.96 This section will review the pathology of prostatic sarcomas, provide information relevant to clinical situations, and review treatment options with the authors’ recommendations.
Pathology One of the difficulties in evaluating sarcomas of rare sites, especially when reviewing historical data, is with regard to the pathologic classification of this malignancy. Sarcomas may be classified on the basis of their biological behavior, tissue of origin, site of occurrence, histologic appearance and/or grade, immunohistochemical profile, molecular profile, and/or microarray analysis. Therefore within the rare
UNCOMMON CANCERS OF THE PROSTATE
entity of sarcoma of the prostate, there may be many more subclassifications. Amongst sarcomas of the adult prostate, based on traditional classification schemes, leiomyosarcoma appears to be the most common followed by rhabdomyosarcomas (RMS).95,97,98 This is in comparison with malignant fibrous histiocytoma and liposarcoma being the two most prominent historical subtypes of soft tissue sarcoma overall. Figure 1 describes histology of prostatic sarcoma. The difficulty and importance of sarcoma classification should not be overlooked. Several studies have demonstrated that pathologists with specific sarcoma experience at specialized centers will have different interpretations from local pathologists.99 – 102 These differences in interpretations may have prognostic implications, and may be especially important in the context of clinical trials.101,102 Historical case reports and series describing the frequency of sarcoma subtypes should be viewed critically, as many of the series that compiled the frequency of pathologic diagnosis were based on the use of light microscopy alone. The use of immunohistochemical techniques, and to a lesser degree electron microscopy, has permitted a more accurate identification of sarcoma subtypes than was possible even as recently as a decade ago. For example, equivocal diagnoses of the most common sarcomas of the prostate, RMS and leiomyosarcoma, can often be confirmed by the demonstration on electron microscopy of the presence of thick and thin filaments or Z-band material,103,104 or immunohistochemical positivity for antigens highly specific to muscle tissue, such as desmin, myoglobin, the M subunit of creatine phosphokinase (CPK), skeletal muscle myosin, or muscle-specific actin.105 Furthermore, some of the sarcoma subtypes have been found to be highly associated with individual cytogenetic abnormalities,106,107 thereby providing another potential means to arrive at a more definitive diagnosis. For example, t(l2;16)(ql3;pl1) has been identified in myxoid liposarcoma. The presence of this translocation would support this diagnosis over that of the myxoid variants of malignant fibrous histiocytoma, a distinction that is often difficult to make. Other associations include the t(X;18)(pll.2;qll.2) with synovial sarcoma, and the t(2;13)(q37;ql4) with alveolar RMS. In contemporary times, investigators have shown that, in some cases of RMS, there is a conversion to homozygosity within tumor tissue of a constitutionally heterozygous region at band 11pl3.108 This event implies the unmasking of a recessive mutant allele by elimination of a balancing wildtype homolog, with the resultant effect that a gene product essential to differentiation is not produced. As such data can also be used to distinguish one sarcoma subtype from another, continuing dissection of sarcoma at the molecular level will allow more precise diagnosis of these disorders, as well as a better understanding of the histogenesis and pathogenesis of these malignancies. Recently, the use of microarray analysis has been developed to characterize tumors by gene expression patterns.109,110 This technology permits simultaneous analysis (profiling) of multiple gene expression markers. Alveolar RMS, Ewing’s sarcoma, gastrointestinal stromal tumors, and synovial sarcoma have all been characterized.111 – 114 This
47
technology also reveals the drawback of less-precise histologic subtyping of sarcomas, such as malignant fibrous histiocytoma, which has variable gene expression profiles.110 In addition, potential therapeutic molecular targets may be discovered.114 With the above caveats in mind when reading the historical literature of sarcoma, and in particular, rare sarcomas such as primary sarcoma of the prostate, several subtypes have been described. Many of the histological subtypes of sarcoma that are more commonly found at other locations in the body have also been described in the prostate.95,97,98,115 – 122 In this chapter, we will limit our specific discussions to include some of the more common sarcomas of the adult prostate (leiomyosarcoma and RMS), a tumor rarely seen outside of the prostate in adult males (cystosarcoma phyllodes), and some other interesting tumors (carcinosarcoma and postradiotherapy prostatic sarcoma) before discussing the broader clinical aspects of sarcoma of the prostate. Leiomyosarcoma
Although leiomyosarcoma is the most common histology to affect the prostate, we will not dwell too much on it specifically since most of the general information on prostatic sarcoma (see below) applies to this entity. Most of the larger series of adult prostatic sarcoma have found this histology to be the most common.95,96,98 Cheville et al. reviewed 23 cases of leiomyosarcoma of the prostate from a single institution (treatment plus pathologic consults) from 1929 to 1994, of which 14 had clinical follow-up information.120 Patients ranged in age from 41 to 78 years (mean 61), all presenting with urinary obstruction. During the follow-up period (mean 19 months, range 2–72), 10 of 14 died, with a mean of 22 months after diagnosis. Three of four patients still alive had less than 5 months of follow-up at the time of reporting. The mean time to metastatic development was 10.3 months, with the lungs the most common site of metastases. This report is consistent with previous investigators reporting median survival of 16–49 months.95,97,98 Although adult RMS is infrequently described in the literature,123 – 126 it is probably even less common than the literature would suggest. In a Finnish study involving 880 cases of soft tissue sarcoma among adults older than 40 years, 25 of these were originally identified as RMSs.127 However, when these 25 cases were retrospectively evaluated by histology, immunochemistry, and electron microscopy, with strict criteria for the diagnosis of striated muscle differentiation, only 2 of the 25 cases were convincingly demonstrated to be RMS. Both tumors arose in the urogenital region, one in the prostate and the other in the bladder, and both patients died within 3 months of diagnosis. Of interest, 30 pleomorphic sarcomas in this series were also evaluated to determine whether any case of RMS was misdiagnosed, but none of the latter was found.127 Smith and Dehner reviewed the records of the Armed Forces Institute of Pathology and found 55 primary sarcomas of the prostate (including pediatric patients).97 Twenty three of the 55 patients (42%) were classified as having RMS. Of these, 17 were embryonal RMS (median age 16), 4 alveolar, and 2 other. Waring et al. published a case series of adult
48
GENITOURINARY CANCER
prostatic embryonal RMS in a clinicopathologic review.128 Six cases from the post-1958 literature (after reclassification of RMS) met criteria of patients at least 18 years of age, having prostate as the origin of disease, and adequate clinical information; the authors added three new cases from Australia. The median age was 31 years (mean 39) and median survival was 8 months from diagnosis. Cystosarcoma Phyllodes
Cystosarcoma phyllodes is a tumor rarely described in men. Among men with this tumor, most have a prostatic site of origin. Fewer than 30–40 cases of cystosarcoma phyllodes of the prostate have been described.129 – 143 It resembles its namesake in the female breast; both are described as having a foliated (leaflike) ductal component, as well as a sarcomatous stromal pattern. The clinical course of the few prostatic cases also seems to parallel that of the breast. Specifically, they are characterized by their slow growth, local invasion of surrounding tissues, and uncommon tendency to metastasize. However, this histology in the prostate has occasionally been shown to be capable of aggressive behavior and cause rapid death.132,133,142 Histologic features that appear to distinguish a more aggressive lesion include the degrees of cellularity, nuclear pleomorphism, and mitotic activity. The age range for presentation of cystosarcoma phyllodes of the prostate is broad, from the fourth through the eighth decades. There is no agreement regarding what histologic criteria comprise prostatic cystosarcoma phyllodes. Some investigators have attempted to distinguish its nonaggressive subset, that is, cases with few or no mitotic figures, variable cellularity, and small numbers of atypical nuclei, by designations such as “phyllodes tumor of atypical prostatic hyperplasia”131 or “atypical fibromuscular hyperplasia”.134 Gaudin et al. reviewed 22 cases of proliferative stromal lesions of the prostate from two centers.136 Their cases of phyllodes tumors were placed in the category of “prostatic stromal proliferation of uncertain malignant potential” as opposed to prostatic stromal sarcoma based on the degree of stromal cellularity and the presence of mitotic figures, necrosis, and stromal overgrowth. Other spindle cell tumors of the prostate, both rare and benign, such as pseudosarcomatous spindle cell lesions, blue nevi, leiomyomas, and inflammatory pseudotumors, must also be distinguished from cystosarcoma phyllodes and other malignant sarcomas of the prostate.135 The somewhat arbitrary distinction between “benign” and aggressive lesions may have therapeutic implications. Local resection of a “benign” lesion may be the only requirement for cure, whereas radical surgery with or without the use of other treatment modalities may be appropriate curative strategies for the more aggressive manifestations of the disease (see section on “Treatment”). Recently, Bostwick et al. reviewed 23 cases of phyllodes tumor of the prostate and provided long-term follow-up.144 The patients ranged in age from 25 to 86 years at diagnosis and had a mean followup of 7.2 years (0.3–25 years). Overall, tumors recurred in 15 of the 23 patients (65%) with a trend for increased recurrence rate in the higher-grade tumors (89 vs 50%, one-sided t-test
p = 0.05). There was also a trend for longer overall survival based on tumor grade (p = 0.06). Carcinosarcoma
Another unusual histology found in the prostate, but less exclusive to it than adult RMS and cystosarcoma phyllodes, is carcinosarcoma. This mixture of malignant epithelial and malignant mesenchymal elements has been described in only 20–25 cases.95,145 – 158 Carcinosarcoma is a difficult lesion to diagnose. It must be distinguished from (i) a primary carcinoma that has become anaplastic and spindle shaped, (ii) a primary carcinoma that has produced nonmalignant sarcomatous changes in the stroma, and (iii) the “collision” of a separately occurring carcinoma and a sarcoma.95 Mostofi and Price have defined carcinosarcoma as a malignant mixed tumor with carcinomatous and sarcomatous components, the latter consisting of neoplastic cartilage or bone.159,160 Indeed, in the absence of malignant heterologous elements, such as bone, cartilage, or striated muscle, it may be difficult to ascertain mesenchymal differentiation of cells that do not have an epithelial appearance.151 Many theories have been proposed to explain the origin of carcinosarcoma, although none has been proven. These include the derivation of the tumor from an uncommitted stem cell with a capacity for both epithelial and mesenchymal differentiation; the ability of an adenocarcinoma to induce a metaplastic response that subsequently becomes malignant; derivation from a mesodermal vestige within the prostate; or the collision of two tumor types that develop independently.145 Of interest in this regard, Chung et al., inoculating specific epithelial and fibroblast cell lines into either adult male syngeneic rats or athymic nude mice, induced the development of tumors that resembled carcinosarcoma.160 In this model, the investigators were able to establish that the fibroblasts influenced the epithelial cells rather than the reverse – namely, that the expression of tumorigenic phenotypes by the epithelial cells was regulated epigenetically by fibroblasts through a paracrine pathway. The symptoms and signs experienced by a patient with carcinosarcoma are not distinctive from those experienced by a patient with either adenocarcinoma or sarcoma of the prostate. The age range overlaps that of prostatic adenocarcinoma. Cases have been reported among patients in their fifth through ninth decades.145,146,158 PAcP or PSA can be slightly elevated or normal. The metastatic pattern parallels that of sarcoma more closely than that of prostatic adenocarcinoma, with a predominance of local regrowth, and nodal and soft tissue progression.152 The reported mean survival is approximately 21 months, with a median survival of approximately 1 year.145,146 The largest series reported on 21 patients treated or seen in pathology consultation from a single institution.158 Patients ranged between 50 and 89 years of age at diagnosis. The 5-year CSS was 41% with a 7-year survival of 14%. During the follow-up period, 18 of the 21 patients died, with one other patient lost to follow-up with known metastatic disease present. There was no difference in survival by treatment modality.
UNCOMMON CANCERS OF THE PROSTATE
Radiation-associated Sarcoma
Another important topic is about sarcomas developing in a prostate in a previous therapeutic radiation field. The impact of radiation exposure on the development of a subsequent malignancy has been described historically, and therapeutic radiation also carries with it a risk of secondary malignancy, with an incidence of sarcoma in 0.8% of described women receiving adjuvant radiotherapy for breast cancer.161 The specific risk of radiation-associated sarcoma in the prostate is unknown. A retrospective single institution study estimated the risk with long-term follow-up to be 0.03–0.8% with a multitude of sites and reasons for the radiotherapy.162 The latency period reported in that study was 12 years, consistent with that reported from other institutions.163 In a review of patients treated surgically at a single institution for radiationinduced soft tissue sarcomas, 14% of the cases developed after radiotherapy for prostate cancer (third most frequently after breast cancer 29% and lymphoma 15%).164 The histology of these sarcomas has been reported as variable (with the caveats described at the beginning of this section even more important), with malignant fibrous histiocytomas and osteosarcomas among the most frequent.165 The majority of the cases reported in the literature tend to be high grade and often associated with a poor prognosis.164 – 166 Postradiation sarcomas of the prostate do not appear to behave differently than similar sarcomas in other sites when stratified for grade. Over 20 cases have been reported.164,167 – 173 Canfield et al. reported a case of sarcoma developing after brachytherapy for adenocarcinoma of the prostate.173 They concluded from their review of the literature that many of the cases, especially those with short latency periods, may in fact be carcinosarcoma or dedifferentiation of adenocarcinoma. The clinical presentation of radiation-associated prostatic sarcomas does not appear to be significantly different from that of other prostatic sarcomas, although pain may be a more prominent symptom after radiation than in de novo cases.
Clinical Characteristics Some specifics regarding the clinical presentation and biologic behavior of certain subtypes of prostatic sarcoma are described above. However, many of the symptoms of clinical presentation are similar, despite differences in histologic diagnosis. Obstructive symptoms are the most common across the board in men presenting with sarcoma of the prostate. These obstructive symptoms may range from mild to acute urinary obstruction and its consequences. Other relatively common symptoms include hematuria, pyuria, and dysuria. Other reported clinical signs and symptoms include urinary incontinence, changes in bowel habits, weight loss, fever, edema, and pain. The disproportionate presentation of pain in those cases of sarcoma occurring after radiation has been discussed above. The most common sites of metastatic disease include lymph nodes, lungs, and bones. Waring et al. reported clinical characteristics of the disease differentiating it from adenocarcinoma of the prostate, including a younger age at diagnosis, higher frequency of a suprapubic mass, more lymph nodal metastases, presence of osteoclastic metastases, and rapid clinical course in their review of
49
adult embryonal RMS.128 However, since any space occupying lesion of the prostate can cause similar subjective and objective evidence of disease, none of them is distinctive to prostatic sarcoma. It has been suggested that there may be differences found by digital rectal exam. Sarcoma may be smooth, tense, and symmetrical; or soft and balloonlike; or smooth and firm; these consistencies can be distinguished from the fixed, hard, and irregular consistency that is usually characteristic of the gland in adenocarcinoma.94,95,119 Nonetheless, the consistency of prostatic sarcoma mimics that of prostatic adenocarcinoma in the majority of cases; only approximately 20% of sarcomas have these unique features at palpation.97 An important point is that it is unusual for children or young adults to have an abnormal prostate or to present with symptoms of urinary obstruction, and this clinical picture should alert the examiner to the possibility of sarcoma. The differential diagnosis of an abnormal prostate examination finding in the older patient includes benign prostatic hypertrophy, cysts or abscesses of the prostate, cysts or neoplasms of the seminal vesicles and M¨ullerian duct remnants, tuberculosis prostatitis, sarcomas or TCCs of the bladder invading the prostate, or metastatic lesions to the prostate. Tissue diagnosis is usually achieved by conventional means, that is, by transrectal biopsy under ultrasound guidance. In general, there are few radiologic characteristics of prostatic malignancies that may be diagnostic in differentiating an adenocarcinoma from another malignancy.174 However, the characteristics of sarcomas may give some clues. Sarcomas are often associated with marked prostatic enlargement in comparison to other prostatic malignancies. Apart from this general difference, imaging may play a more clinical role in staging, and following up treatment than in diagnosis of the primary lesion. There have been several reports of a few cases of different imaging modalities of prostatic sarcoma. The most common, and probably the most clinically useful in the right context are ultrasonography, CT, and MRI. Both pelvic and transrectal ultrasonography results have been described in prostatic sarcomas. Pelvic ultrasonography has proved to be useful in pediatric pelvic RMSs,175,176 but is probably not as useful as other imaging modalities in adults. Transrectal ultrasonography has been reported to be helpful in assessing characteristics of the primary tumor, as well as following up treatment in adult prostatic sarcoma.170,177 Computed tomography has demonstrated utility in pelvic malignancies, but is rarely diagnostic of a prostatic sarcoma, with the exception of the unique appearance of cystosarcoma phyllodes.131 Figure 2 demonstrates typical CT appearance of a sarcomatous prostate. MRI may be superior to CT in tumors of soft tissue and bone, but both modalities may be complimentary178,179 and findings on MRI of prostatic sarcoma have been described.180 Of interest, in three of four cases described by Bartolozzi et al., MR images showed the site of origin to be the central area of the prostate, with compression of a clearly recognizable peripheral zone, which may be relevant in differentiating sarcomas from carcinomas.
50
GENITOURINARY CANCER
Figure 2 Computed tomography images of malignant fibrous tumor of the prostate. Notice the enlarged heterogeneous prostate with areas of central necrosis. The seminal vesicles are pushed posteriorly with the mass pushing into bladder base without invasion. The trigone is elevated and rectum spared with tumor extending into the pouch of Douglas, consistent with a prostatic sarcoma (Courtesy of William Boswell, MD. University of Southern California – Kenneth Norris Jr Comprehensive Cancer Center).
Prognosis Prognosis of individual categories of sarcoma of the prostate has been described above. Since there have been no formal prospective studies validating prognostic factors in prostatic sarcoma, we can only extrapolate based on historical cases and relevant factors from sarcoma in other anatomical sites. However, the same limitations described above regarding difficulties in classifying sarcomas apply even to some prospective studies of sarcoma, particularly when it comes to analysis by histologic type. The site of origin of sarcomas has been described as a prognostic factor in soft tissue sarcomas overall, with nonextremity sites of origin having higher risk of death.181,182 This fact alone implies that prostatic sarcomas will have a poorer prognosis than the same counterpart found on an extremity. Other factors that have been shown to be prognostic in soft tissue sarcomas are grade, tumor size, depth, age, gender, histology, presence of lymph nodal metastases, and margin positivity after surgery.181,183 – 187 Only the tumor grade has consistently stood out as a significant prognostic factor on multivariate analysis across studies, keeping in mind the caveats in grading tumors mentioned previously.
The presence or absence of distant metastases is also prognostic.188 More recently, DNA content, Ki67, and P-gp staining have been shown to be prognostic.189 It remains to be seen what the impact of microarray analysis will be on clinical outcome. Kattan’s postoperative nomogram may be clinically very useful in soft tissue sarcoma overall, but is of unproven utility in prostatic sarcoma.181 Keeping in mind the above discussion, the prognosis of sarcoma of the prostate is variable. Patients presenting with metastatic disease or large high-grade lesions may have a rapidly progressive course and a short survival. Other patients may have an indolent course, with or without aggressive therapy. The factors mentioned above, in addition to other patient characteristics and preferences must be kept in mind when developing a treatment plan (see below). Young patients with high-grade prostatic sarcomas tend to have a poor prognosis in spite of aggressive treatment.
Treatment Surgery continues to be the primary modality of therapy for localized soft tissue sarcomas. However, there are reports of
UNCOMMON CANCERS OF THE PROSTATE
treatment with primary radiotherapy. Multimodality therapy has also been used (and advocated) for localized disease. Systemic chemotherapy has been used for metastatic disease. Because of the lack of prospective data, only extrapolation from general sarcoma data and anecdotal case reports are available for guidance. We will discuss the available data and give our recommendations, particularly with reference to specific clinical circumstances. The largest amount of data about prospective treatment comes from soft tissue sarcomas of the extremities. For this disease, complete surgical excision with clean margins has produced the best long-term results, particularly with regard to recurrence-free survival.183,187 Postoperative radiotherapy has been added to surgery, and appears to decrease local recurrence rates, particularly with high-grade lesions.190 Preoperative radiotherapy has also been studied, although there are no published prospective randomized controlled trials looking at survival in this setting. Retrospective multivariate analysis has found no significant difference in outcome between preoperative and postoperative radiotherapy.191 Complete resection is also the most important factor in outcome for retroperitoneal sarcomas, although complete resection of nonextremity sarcomas is more problematic, leading to higher recurrence rates. In a study of 307 soft tissue sarcomas of patients with extremity primaries (twothirds of whom had a limb-sparing procedure), 31% suffered a local or distant recurrence; by contrast, 41% of patients with trunk primaries and 47% with retroperitoneal primaries had recurrences.192 Series reporting higher resectability rates show higher survival rates than those with lower resectability rates, and complete resection remains a laudable goal, if it can be accomplished without excessive morbidity. Complete resection has not been uniformly defined, but usually refers to the resection of all gross disease; microscopic residual disease often remains. Among 63 patients with primary retroperitoneal sarcomas, Jaques et al. demonstrated a 5-year median survival for completely resected sarcomas, but survival of only 2 and 1 years respectively, for those partially resected or unresectable.193 Glenn et al. has reported a 43% 3-year survival (median follow-up 29 months) among completely resected primary retroperitoneal sarcomas.194 The results of complete resection of nonextremity sarcomas in other locations are even better. In a study of 176 adults with primary soft tissue sarcomas of the head and neck, patients with negative margins had 85% survival at five years versus a 28% survival among those whose margins were positive.195 In another study of 57 patients with head and neck, breast, and trunk sarcomas, completely resected disease resulted in a 77% 3-year survival (median follow-up, 35 months).194 Among 50 patients with primary colorectal sarcomas, complete resections led to a median survival of 174 months, whereas a less than complete resection achieved a median survival of only 12 months.196 Despite the aggressive surgical approaches demanded by complete resection of nonextremity sarcomas (en bloc removal of organs is commonly required to accomplish this goal), approximately 50% of patients will have a recurrence.186,193,194,196 Even in this setting, it appears that continuing attempts to resect recurrent disease completely
51
may prove beneficial in regard to prolongation of survival. In a study of 88 recurring retroperitoneal soft tissue sarcomas, complete resections were possible in 49, 47, and 33% of first, second, and third recurrences.193 Patients whose recurrences were completely resected had a 48-month median survival as opposed to that of 21 and 15 months for patients with partially resectable or unresectable malignancies. In a similar analysis among patients with soft tissue sarcomas at all locations except the viscera, 66 of 107 patients with an initial recurrence were rendered disease-free by surgery.192 Their actuarial survival at 3 years was 51% compared to a median survival of 7.4 months among patients not rendered diseasefree. Thirty (45%) of the 66 remained disease-free at the time of the report (median follow-up 28 months), but the other 36 had recurrence. Sixteen (44%) of these 36 were again rendered disease-free; their survival was 18 months, compared to 6 months for those whose disease could not be completely resected. However, only 4 of these 16 remained disease-free at follow-up. The use of radiotherapy as an adjunctive treatment following complete resection, because of the common presence of microscopic residual disease and local recurrence, is often considered. There are few data in nonextremity sarcoma to support this practice, but sarcomas are known to be moderately radiosensitive. As in the case of extremity sarcomas, we do not recommend adjuvant radiotherapy for “low-grade” lesions that have been resected with negative surgical margins. In cases of “high-grade” sarcomas, we recommend conformal radiation therapy (either 3-D CRT or IMRT) to either 60 Gy (with negative margins of resection) or at least 66–70 Gy (with positive margins of resection) at 2 Gy per fraction. In addition, for low-grade sarcomas with positive surgical margins, adjuvant CRT to at least 66–70 Gy level at 2 Gy per fraction should be considered. In general, RMSs require significantly lower doses of radiation depending upon the response to initial chemotherapy (40 Gy with a complete response to chemotherapy, or at least 50 Gy if less than a complete response is achieved) and margin status after resection (40 Gy with negative margins, or at least 50–55 Gy with positive margins). Case reports suggest that radiotherapy has a useful palliative role for the prostatic sarcoma that cannot be completely resected. However, its curative potential without the use of surgery is anecdotal. Surgery alone seems acceptable for low-grade lesions (e.g. cystosarcoma phyllodes), if clear margins can be obtained, although metastatic cystosarcoma phyllodes has been treated successfully with chemotherapy in an anecdotal report.140 The role of chemotherapy will be discussed in further detail below. The reader must be aware, however, that the foundation from which these recommendations are built is not firm. Data on nonextremity sarcomas have been collected retrospectively, often over several years. Different studies may generate dissimilar results for several reasons. Factors that vary between studies include the skills of the participating surgeons, surgical practices and supportive care over time, staging techniques, histopathological classification, adjunctive therapies to surgery, the differing mix of prognostic factors, and follow-up intervals. Even within studies, attempts
52
GENITOURINARY CANCER
to dissect some of these influences through the use of multivariate analyses have been limited. In addition, for the determination of the effect of surgery on recurrent disease, the influential role of lead-time bias needs to be considered. Even if these interpretive difficulties could somehow be obviated, it remains clear that there is significant room for therapeutic improvement and innovation, and a continuing need for well-structured clinical trial design in this context. From a therapeutic perspective, the prostatic sarcoma that has been most critically examined is RMS, although the majority of cases happen in the pediatric age range. There have been many contributors to the development and evaluation of therapeutic strategies in RMS, but the Intergroup Rhabdomyosarcoma Study (IRS) Committee, which receives patient data from centers around the world and has coordinated trials since 1972, has been the most instrumental. As adult prostatic RMS is an uncommon manifestation of an uncommon problem, we will only briefly review the salient therapeutic findings. The first Intergroup Rhabdomyosarcoma Study (IRSI) approached bladder –prostate sarcomas with a primary extirpative surgery, if feasible, followed by radiotherapy (50–60 Gy) and chemotherapy (vincristine, dactinomycin, and cyclophosphamide).197 On the basis of the data suggesting that the bladder salvage rate was higher if chemotherapy and radiotherapy were applied initially, followed by surgery only if malignancy remained, one of the major objectives of IRS-II, III, and IV was to continue to improve disease control and survival with increased bladder preservation. Bladder preservation rates improved from 25 to 60% in IRS-II to IRS-III.198,199 In IRS-IV, 82% of patients with nonmetastatic RMS survived for 6 years and 55 of 88 patients retained their bladder without relapse.200,201 Unfortunately, the favorable results in the pediatric age-group have not carried over to patients diagnosed with RMS or other prostatic sarcomas beyond puberty, who have a poor prognosis in spite of multimodal therapy. The role of chemotherapy for soft tissue sarcomas in the setting of both adjuvant therapy and metastatic disease has been extensively reviewed and is only briefly discussed here.202 A broad range of traditional cytotoxic agents, including doxorubicin, ifosfamide and cyclophosphamide, methotrexate, cisplatin and carboplatin, dacarbazine, actinomycin D, and etoposide, have demonstrable activity against soft tissue sarcomas, with reported objective response rates for advanced disease in the wide range of 0–40%. No optimal agent has been defined, although there is a practical current consensus that doxorubicin and ifosfamide should be regarded as the anchors of treatment. Although there is no consensus regarding the respective merits of single-agent and combination chemotherapy, it is generally agreed that combination chemotherapy regimens yield higher objective response rates (60–70% vs 0–40%), but without clear evidence of survival benefit. In current practice, the CYVADIC and MAID regimens, incorporating cyclophosphamide or ifosfamide in combination with doxorubicin and dacarbazine, are widely used for advanced soft tissue sarcomas, although there is considerable controversy as to whether the toxicity of these regimens is justified by the modest reported outcomes.
At the University of Southern California it has been standard practice when choosing to give combination chemotherapy for soft tissue sarcoma outside a trial to use the combination of doxorubicin and ifosfamide with mesna prophylaxis (AIM) without dacarbazine to reduce toxicity while attempting to maintain the efficacy of combination chemotherapy. With respect to adjuvant therapy, controversy persists regarding its role, even in the treatment of more common extremity sarcomas. Prostatic sarcomas should be regarded as “deep” tumors,188 irrespective of stage and grade, when considering the role of adjuvant chemotherapy. The literature on adjuvant chemotherapy of sarcomas is dominated by series including extremity sarcomas, although these reports have often included nonextremity tumors as well,203 – 205 making it more difficult to give specific recommendations regarding prostatic sarcomas. When nonextremity sarcomas have been treated exclusively in adjuvant trials, there has been no clear demonstration of prolonged survival, although one trial did show an improvement in disease-free survival.206 However, each of these studies was conducted with such small number of patients, possibly used suboptimal chemotherapy, and/or suffered from such low statistical power, that the issue of adjuvant therapy for nonextremity sarcoma should more properly be considered unresolved. Given the limitations of conventional cytotoxic singleagent and combination regimens, newer agents and approaches have been applied in both the adjuvant and metastatic setting. Epirubicin, an analog of doxorubicin with possibly reduced cardiotoxicity, may have a role in the treatment of soft tissue sarcomas (as may liposomal preparations of doxorubicin). Taxanes and gemcitabine have also been assessed in this context. Hyperthermia and intraoperative photodynamic therapy are being combined with other modalities in an attempt to improve response. Preoperative chemotherapy has also been studied, but remains unproven in this context. In theory, the attractions of this strategy include: (i) the reduction of requisite surgery or conversion of a nonresectable mass to an operable lesion, (ii) sparing of nonresponders from further ineffective chemotherapy in the adjuvant setting, (iii) identification of patients with a poor prognosis, thereby stimulating a change in approach, (iv) possible radiosensitization of tumor tissues, and (v) possible effect on occult metastatic deposits before completion of definitive local therapy. In addition, regimens of high-dose chemotherapy are being evaluated in soft tissue sarcomas, although we are unaware of specific protocols for patients with prostatic sarcomas. As we move toward a molecular basis of diagnosis, targeted therapies may come into play.182 Thus far, the only targeted therapy to have proven clinical benefit is imatinib (and possibly newer derivatives) for gastrointestinal stromal tumors.
Recommendations As with all cases of rare tumors, consideration should be given to referral to an experienced center with multidisciplinary care. Pathologic review of the available tissue by an experienced, specialized pathologist is important in guiding further therapy. In general for soft tissue sarcoma of
UNCOMMON CANCERS OF THE PROSTATE
the prostate, based on the data for other nonextremity sarcomas, it appears reasonable to approach resectable prostatic sarcomas with initial aggressive surgery. Complete resection of all gross disease should be the primary goal, as lesser resections do not appear to prolong survival. This usually includes cystoprostatectomy, and often, total pelvic exenteration. On the basis of the data for high-grade extremity sarcoma demonstrating the equivalence of limb-sparing surgery followed by radiation therapy to amputation, it is not unreasonable to adopt a similar approach in high-grade prostatic sarcomas, that is, resection of all gross disease with an attempt at “clear” margins followed by radiation therapy. We recommend CRT (either 3-D CRT or IMRT) to deliver high doses of adjuvant irradiation (60–70 Gy depending upon the margin status and grade of sarcoma). Significant doses of radiation therapy can be delivered to the prostate bed often with a tolerable degree of side effects. For low-grade tumors, such as low-grade leiomyosarcomas or cystosarcoma phyllodes, surgical resection is probably all that is warranted, but local recurrences are very common. For recurrences, complete resection of disease should be considered, whether the recurrence is local or distant. Given the morbidity associated with these procedures, resection should be attempted only if the surgeon believes that all gross disease can be successfully removed. Adjuvant chemotherapy programs for sarcomas have been associated occasionally with a prolonged disease-free interval and rarely with a longer survival. Although the prolongation of disease-free status is not an unreasonable goal, the inconsistency of this result, especially in the setting of nonextremity sarcoma, suggests that the role of adjuvant chemotherapy for prostatic sarcoma should be confined to the investigational setting. Patients with good performance status and locally advanced or metastatic prostatic sarcoma are candidates for chemotherapy, although its benefits, if any, are likely to be modest. Despite increases in toxicity without clear survival benefit, in patients in whom control of metastatic disease followed by resection is a possibility, combination chemotherapy is recommended. Until new agents are found, anthracyclines and ifosfamide, with or without dacarbazine, remain the drugs of choice. These limitations emphasize the importance of continued innovative approaches and the entry of eligible patients into well-structured clinical trials.
TRANSITIONAL CELL CARCINOMA OF THE PROSTATE Primary urothelial transitional cell carcinoma (TCC) most commonly arises in the bladder, but may arise from the transitional cell epithelium of the upper tracts, posterior urethra, or distal excretory prostatic ducts. Solitary TCC of the prostate is a rare entity, much less common than the approximate 63 000 cases3 of bladder cancer expected in the United States in 2005 (or the 356 000 cases207 worldwide). Much more commonly, the prostate is involved either as a synchronous or a metachronous lesion to one in the urinary bladder. Certain aspects of TCC of the prostate continue to be confusing, namely, staging and pathogenesis. Some
53
of this confusion is derived from the historically imprecise nomenclature for this tumor. Melicow and Hollowell are generally given credit for the first description of urothelial cancer involving the prostate in 1952,208 although earlier authors may have examined prostates with evidence of TCC.209,210 Bowen’s disease was used to describe 30 cases of urothelial carcinoma in situ (CIS). Three patients with CIS of the prostate were found incidentally after open prostatectomy. Ortega et al. used the term Paget’s disease for in situ TCC of the prostate.211 The association of CIS of the urinary bladder and that of the prostatic urethra was also noted.212 Franks and Chesterman presented a case similar to the previous ones and stressed the use of the term “cancer in situ” rather than the eponymic titles Bowen’s or Paget’s.213 Ende et al. hypothesized the origin of these cancers from the prostatic ducts as they enter the prostatic urethra,214 later confirmed by others.215,216 In 1972, Johnson et al. demonstrated the first cases of noncontiguous TCC of the bladder and the prostate.217 Seemayer et al. raised concern about patients with CIS of the bladder and the subsequent development of prostatic ductal TCC, which according to them may occur quickly and silently.218 Many of the early case reports note the presence of infection or cystitis without frank bladder tumors in association with TCC of the prostate. Grabstald theorized that some of these lesions actually represented unappreciated CIS of the bladder.219 In this section we will review primary TCC of the prostate. Primary TCC is defined as carcinoma arising from the transitional cell epithelium of the prostatic structures without upper tract or bladder cancer.220 Much more commonly, the epithelium of the prostatic urethra or ducts is involved secondary to a bladder cancer.217 Any review of the literature is confounded by the admixture of primary and secondary TCCs. In many cases of primary TCC, the extent of histological investigation of the bladder is uncertain.221 Also confusing the situation are some early reviews, which considered TCC of the prostate to be more closely related to adenocarcinoma of the prostate than to a TCC elsewhere in the urinary system.222 It is common for these reports to include a history of hormonal therapy without response.
Incidence The true incidence of this rare disease is difficult to determine. As noted above, the diagnosis of this entity in some historical reports is doubtful. There may be an increasing frequency of the finding of TCC of the prostatic structures,223 – 225 although this may instead reflect an increased pathological awareness or an emphasis on urothelial mapping with inclusion of prostatic sampling. A 9-year pathologic experience of 2724 cases of prostatic cancers revealed 122 cases (4.5%) of TCC of the prostate.226 Of these, 46 (1.7%) were thought to be solitary lesions without TCCs of the urinary bladder. Other studies report that primary TCC of the prostate comprises 1 to 5% of all primary prostatic neoplasms.214,222,227 – 229 As discussed previously, TCC of the prostate occurs in strong association with that of the urinary bladder. In an effort to determine the incidence of bladder TCC-derived
54
GENITOURINARY CANCER
involvement of the prostate, a number of authors have performed pathological evaluation of the prostate in specimens removed for treatment of bladder cancer. Pagano et al. reported on 72 of 562 (13%) patients with pathologically staged prostatic involvement after cystectomy for T1 to T4 disease.230 Others have found an incidence of 12–43%,231 – 233 confirmed by our own institution’s experience of 29.2% (143 of 489 cases).234 A retrospective pathological review by Kirk et al. reported a 55% incidence of TCC invasive to the prostate, dysplasia, or CIS of the prostatic ducts in cystoprostatectomy specimens.235 These authors conclude that the high incidence of these malignant or premalignant abnormalities portends a poor prognosis for nonoperative therapy. A retrospective review of a 20-year experience at the Mayo Clinic notes that 23% of prostatic TCC were true primary lesions without disease elsewhere in the urinary tract.227
Pathogenesis and Etiology It is likely that the risk factors for the development of transitional cell cancer of the prostate are the same as that of the risk factors for urothelial cancer elsewhere in the urinary system. These presumptive carcinogenic agents include tobacco smoke, aromatic amines (β-naphthylamine, benzidine), analgesics (especially phenacetin-containing analgesics), cyclophosphamide, radiation, and chronic inflammation or infection.236 Several less-established associations have been proposed. Intact RNA of human papilloma virus type 6 has been detected in grade 1 TCCs of the distal urethra;237 investigators have postulated a role of this virus as an agent for the development of urothelial carcinomas arising from transitional cell epithelium as analogous to condylomata arising from squamous epithelium. A role for chronic irritation from cystitis or urethritis has been proposed but not proven.215 Asbestos has been reported as a risk factor in a case report.238 CIS is the putative precursor lesion, but other changes have been identified. Ullmann and Ross presented nine cases with abnormal epithelial changes of the periurethral prostatic glands of patients with benign bladders.239 These abnormalities included hyperplasia, hyperplasia with atypia, and CIS. Although they were not able to show temporal progression of these lesions to frank invasive TCC, they have been proposed as potential precursor lesions. Johnson et al. identified three mechanisms of development: (i) direct extension from an invasive bladder lesion, (ii) prostatic urethral implantation from a lesion elsewhere in the urinary tract, and (iii) de novo development from the prostatic urothelium.217 Other patterns of involvement include contiguous intraepithelial spread and metastatic spread. Supporting evidence for intraepithelial spread of transitional carcinoma is provided by prostate-mapping studies of cystoprostatectomy specimens.240,241 The authors of the chapter in the previous edition and colleagues examined eight cases of transitional cell CIS of the seminal vesicles.241 The finding of CIS of the seminal vesicles is very highly associated with carcinoma of the bladder and prostate, and it is always found with transitional cancer of the prostatic ducts. As the seminal vesicles contain columnar and no transitional epithelium,
this is most consistent with mucosal spread via the prostatic ducts (in the absence of evidence of transmural penetration from the bladder).241
Pathology The vast majority of primary prostatic carcinomas are adenocarcinomas, with several variants (ductal, mucinous, sarcomatoid, endometrioid, neuroendocrine, and signet ring).242 – 244 Grossly, adenocarcinoma and TCC of the prostate are indistinguishable.222,245 The gland may be firm, irregular, fixed, and enlarged.245,246 TCC involving the prostatic urethra, ducts, and acini is generally easily distinguishable from adenocarcinoma, and follows the same pathologic criteria of urothelial cancer elsewhere.244 TCC is typically moderately to poorly differentiated and often found with significant chronic inflammation but not particularly associated with squamous metaplasia.245,247 Prostatic urethral dysplasia has been associated with TCC of the bladder.248 There is a strong association of TCC of the prostate and CIS of the bladder. The study by Wishnow and Ro noted that 100% (25 of 25 cases) of their patients with TCC of the prostate had multifocal CIS of the bladder.249 Johnson noted a substantial but lesser association (70%).217 Prout et al. evaluated the significance of transitional cell CIS of the urinary bladder with and without invasive cancer.250 In their patients who underwent cystectomy for CIS of the bladder, 66% had carcinoma of the prostatic urethra, ducts, or stroma. A more recent study by Cheville et al. found a 76% concordance with historical or contemporaneous CIS of the urinary bladder.251 Although secondary involvement of the prostate with invasive TCC of the bladder is not the subject of this chapter, we refer the reader to studies documenting the significant risk of urethral recurrence, a discussion, which may be mistaken for prostatic TCC.251 – 253 As may be expected by the high incidence of prostatic adenocarcinoma in the population most at risk for urothelial carcinoma, there are a number of cases of concomitant prostatic transitional cell and incidental adenocarcinomas. In the large retrospective review by Cheville et al. 4 of 50 (8%) patients had prostate adenocarcinoma.251 Other studies note a larger coincidence of 20–50%, which is comparable to the rates of incidental adenocarcinoma in the general population.217,254 – 256 High-grade TCCs may be difficult to distinguish from poorly differentiated adenocarcinoma.226 Nuclear characteristics are often useful; urothelial cancer cells tend to have larger, more pleomorphic nuclei with more irregularly distributed and coarsely granular chromatin.247 The determination may also be facilitated by the use of immunohistochemical stains for PSA and PAcP; however, these may be false-negative in a poorly differentiated adenocarcinoma.257 Squamous cell carcinoma occurs rarely in the prostate and appears indistinguishable from high-grade TCC. Stains for keratin are essential to make this differentiation.247 Transitional cell cancer involving the prostate tends to be of high grade.226,258 In a series of 110 cases of TCC of the prostate examined by Goebbels et al. there were no specimens with grade 1 histology, 24 with grade 2, and 86
UNCOMMON CANCERS OF THE PROSTATE
prostates with grade 3 urothelial carcinoma.226 Fifty percent of cases have prominent perineoplastic inflammation. This is in contradistinction to prostatic adenocarcinoma, which lacks inflammation.226 When the prostate is involved with TCC, there exists a substantial risk of distant metastasis. In one study, 20% of patients had metastases to bone and lung at presentation.254 The presence of invasive (into prostatic stroma) TCC has been associated with a 100% risk of also having metastatic disease in one study.249
Clinical Features The diagnosis of TCC of the prostate has been made in males from age 7 years to the very elderly, with a mean age at diagnosis in the eighth decade.259,260 Many patients complain of irritative voiding symptoms such as urgency, frequency, dysuria, or hematuria.217 In those patients with concomitant bladder cancer, these symptoms may be referable to the vesical disease.254,261 A number of studies report patient presentations with obstructive voiding complaints.214,217,262 However, concomitant benign prostatic hyperplasia (BPH) is common in this age-group, and the etiology of symptoms may be difficult to distinguish. Because the age at presentation tends to be similar to that of prostatic adenocarcinoma, there is a significant coincidence of prostate cancer. Occasionally, TCC is diagnosed in younger patients.220,263 Primary presentation with nasal skin metastasis from suspected primary TCC of the prostate has been described.264 Prior to the advent of PSA serum markers, TCC occasionally presented as a hormone-refractory “adenocarcinoma”.222,262,265 Signs of prostate involvement with TCC may include prostate nodularity and firmness.217
Diagnosis Cytology is currently a cornerstone for diagnosis and surveillance of urothelial carcinomas but voided cytology alone will not distinguish bladder from prostate origin. Epstein reported prostate TCC in a patient with a history of bladder cancer diagnosed by prostatic cytologic aspirate.266 TRUS results of 221 patients with carcinoma of the prostate were reviewed retrospectively, including two cases of confirmed TCC involving the prostate.267 These two patients had identifiable hypoechoic lesions involving both the anterior and posterior quadrants, which were indistinguishable from those found on TRUS of the adenocarcinomatous prostate. In another study, TRUS was unable to identify TCC invasive to the prostatic urethra.268 This same study identified hypoechoic lesions in five of seven patients with stromal invasion and in 100% of patients with involvement of the ejaculatory ducts. TCC of the prostate from bladder cancer may also be identified using CT scan.269 Algaba et al. reviewed their cystoscopic findings of five patients with TCC involving the prostate.254 Four of these had no abnormalities; one had findings related to the concomitant bladder cancer but a negative prostatic urethra. A prospective review by Montie et al. noted that only 1 of
55
10 patients with TCC of the prostate had cystoscopic evidence of the cancer.270 This patient had a visible papillary lesion of the prostatic urethra. Incidental prostatic urothelial carcinoma has been identified by needle biopsy for presumed adenocarcinoma,271 as well as by transurethral or open prostatectomy for benign prostatic hypertrophy.225 Albert et al. described a diagnosis made after transperineal needle biopsy performed for evaluation of an abnormal digital rectal examination.225 As there is considerable coincidence of prostatic and bladder involvement with urothelial carcinoma and the preoperative identification of the malignancy in the prostate could potentially alter management, adequate biopsy of the prostate is a necessary component for staging of primary bladder cancer.272 – 274 This is especially true for high-risk patients such as those with CIS or with cancer near the bladder neck. Rikken et al. performed endoscopic cold cup prostatic urethral biopsies prior to mitomycin C therapy for bladder carcinoma.275 They noted a CIS or superficial cancer incidence of 27%. Complete pathology was not obtained. This is within the range of incidence from several studies, but slightly lower than the frequency of TCC of the prostate reported by some authors.232,270 A prospective evaluation of three biopsy techniques, transperineal needle core biopsy, transrectal fine needle aspiration of the prostate, and transurethral resection biopsy was performed prior to radical cystoprostatectomy and wholemount 4–5 mm step-section pathological evaluation.270 The results indicate a 20% diagnostic accuracy with perineal needle biopsy and a 40% accuracy with transrectal aspiration. Transurethral resection biopsy correctly identified prostatic involvement in 90% of cases, was 83% correct in defining ductal involvement, and was 40% accurate for prostatic stromal invasion. The authors’ recommended technique was to perform deep resectoscope biopsy of the prostate at the 5 and 7 o’clock urethral positions. This technique was confirmed to be the most accurate by Sakamoto et al., who performed prostatic mapping of cystoprostatectomy specimens.276 They noted that the finding of carcinoma in the prostatic ducts at the 5 and 7 o’clock positions at the level of the verumontanum should raise suspicion of a deeper involvement. Although there exists a theoretic risk of implantation after resection of a bladder tumor and prostate biopsy, Laor et al. found no increased risk of subsequent development of a new prostatic TCC implant.277 Controversy exists over the best staging system for TCC of the prostate. Some authors have used traditional staging systems for adenocarcinoma of the prostate, while others have used similar ones from TCC of the bladder. To be practical, the staging system must predict prognosis. Much of the argument over staging has revolved around secondary involvement of the prostate rather than the primary prostatic TCC.228,230,234
Prognosis The discussion of prognosis is often made difficult by the reporting of cases in which prostatic involvement may have been shown to be truly secondary by today’s diagnostic
56
GENITOURINARY CANCER
standards. The literature has been confounded with discussions of urothelial TCC with additional prostatic involvement rather than true primary prostatic TCC. The prognosis of patients diagnosed with TCC of the prostate has traditionally been very poor; they commonly have metastatic disease at diagnosis.246,254,271,278 – 281 Laplante and Brice note that “if invasion into the prostate gland is proved, the prognosis is so poor that radical cystectomy for cure appears unreasonable”.272 This may reflect the overall historically poor prognosis of patients with advanced-stage bladder cancer. Recent studies indicate more optimistic survival after aggressive therapy of the associated primary bladder cancer, although these reports often describe bladder TCC with prostatic involvement.282 Cheville et al. reviewed 50 cases of primary prostatic TCC, including CIS of the prostatic urethra and prostatic ducts (bladder CIS was included, but current or historical invasive bladder TCC was excluded).251 All patients were treated with radical cystoprostatectomy. Overall 5-year survival was 40%. Patients were broken into four categories of disease with different survival: CIS only (62% 5-year survival), prostatic stromal invasion (35% 5-year survival), extraprostatic extension (0% 5-year survival), and lymph node metastases (30% 5-year survival). Only CIS survival (38% of the patients reported) was statistically significant from the others.
Management For many years the treatment of TCC with hormone ablation was reported, always with unsuccessful results and numerous studies have documented this inefficacy.228,246 The value of intravesical bacillus Calmette-Guerin (BCG) for the treatment of TCC of the prostate remains controversial. Several authors have investigated the use of BCG as initial or as definitive therapy for CIS of the prostatic urethra.283 – 287 Orihuela et al. used BCG intravesical therapy as treatment for 15 patients who had superficial TCC of the bladder and mucosal involvement of the prostate.286 They noted successful treatment in 87% of patients after a mean follow-up of 37 months. They considered the presence of prostatic urethritis and granulomas after the treatment to be a sign of prostatic exposure to the BCG and indicative of adequate therapeutic effect. Lamm considered intravesical BCG therapy to be the treatment of choice for superficial TCC of the prostatic urethra.288 The question remains as to how adequately the prostatic urothelium is exposed to therapeutic BCG. Transurethral resection (TUR), particularly of the bladder neck, has been recommended to facilitate exposure of the lesion to the bacillus;286,289,290 some authors do not recommend TUR prior to intravesical treatment.283 Intravesical chemotherapy for primary prostatic TCC has not been reported to our knowledge. A report by Lockhart and colleagues noted a 36% recurrence rate in the prostatic urothelium after intravesical mitomycin C therapy for superficial bladder TCC.291 However, the prostates were not sampled prior to therapy and this may in fact represent untreated lesions rather than recurrent lesions. Droller and Walsh reported progression to prostatic involvement in three patients undergoing intravesical chemotherapy.292
Transurethral resection of transitional cell lesions of the prostate has been used; early reports noted reasonable survival rates with this method for transitional and mixed cancers.214,278 Shenasky and Gillenwater recommended transurethral resection with repeat surveillance cystourethroscopy as a definitive therapy for superficial lowgrade urothelial cancer of the prostate.293 These authors recommended aggressive therapy for high-grade lesions or those invasive into the parenchyma. Some authors have reported results of radiation therapy for TCC of the prostate. On the basis of the aggressiveness of this tumor, Kopelson et al. suggested that radiotherapy is the treatment of choice.294 Schellhammer et al. noted 20% 5-year survival of patients with stromal invasion who underwent preoperative radiotherapy followed by radical cystoprostatectomy.231 Frazier and colleagues found no difference in the local recurrence rate or survival of patients with and without preoperative radiation treatment.295 Occasional reports of success with primary systemic chemotherapy have been published.246,278,279,296 – 298 On the basis of historical data, it is likely that the most efficacious regimen for unresectable TCC of the prostate would be methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC).224,299 However, other regimens are becoming increasingly common. Many oncologists now regard the combination of gemcitabine and cisplatin as standard for advanced urothelial TCC in place of MVAC, based on probable noninferiority with less toxicity when given in the advanced setting.300 Other regimens also look promising based on phase II studies301,302 and we await the results of phase III studies.
Conclusions and Recommendations TCC of the prostate is very highly associated with TCC elsewhere in the urinary system and is rarely primary. It may be an incidental finding in suspected adenocarcinoma and tends to be poorly differentiated. The diagnostic method of choice is deep transurethral resections through the prostatic urethra. The incidental finding or suspicion of solitary prostatic TCC necessitates evaluation of the entire urothelium, including mapping biopsy of the urinary bladder under anesthesia and upper tract evaluation with intravenous pyelogram (IVP), retrograde pyelograms, or spiral CT scan. If associated TCC of the bladder is discovered associated with prostatic urethral mucosal or stromal invasion, the treatment is largely driven by the bladder cancer. We would generally proceed with cystoprostatectomy, with urethrectomy reserved for patients with stromal involvement, with a positive apical margin at surgery, or for those for whom a cutaneous diversion is planned (as opposed to a neobladder reconstruction). Patients not undergoing primary urethrectomy must be followed up by life-long periodic voided or urethral washing cytology to detect recurrence in the remnant urethra. A full discussion of the management of this more common disease is outside the scope of this chapter. As this is a rare disease, consideration may be given for referral to an experienced center and enrollment on a clinical trial, if available. For the patient with primary TCC of the prostate in the absence of bladder involvement, treatment
UNCOMMON CANCERS OF THE PROSTATE
should take into consideration the stage of the disease and the patient’s general condition. Superficial mucosal involvement or CIS can be effectively treated with TURP and intravesical BCG. Patients with stromal invasion or recurrence after BCG require surgical excision, with radical prostatectomy or cystoprostatectomy depending on the extent of disease, and full bilateral pelvic node dissection. We would consider adjuvant radiotherapy (either 3-D CRT or IMRT) to a dose of at least 66 Gy at 2 Gy per fraction for patients with positive surgical margins. For patients with locally advanced disease, systemic chemotherapy may be employed in the hopes of downstaging the patient and allowing surgical excision. Patients with metastatic disease who are candidates for therapy should receive systemic chemotherapy. The combination of gemcitabine and cisplatin is a reasonable alternative to MVAC for patients with metastatic disease, although the preferred option is participating in a clinical trial.
PRIMARY LYMPHOMA OF THE PROSTATE Hematologic malignancies rarely involve the prostate, and only a small number of cases of primary prostatic hematologic tumors have been reported. This section will review this uncommon disease. In 2005, there will be an expected 63 740 new cases of lymphoma (56 290 non-Hodgkin’ lymphoma (NHL), 7350 Hodgkin’ lymphoma) and 34 810 new cases of leukemia in the United States.3 Worldwide, the WHO reports via Globocan 2002, that there are approximately 362 900 (300 571 NHL, 62 329 Hodgkin’s lymphoma) and 300 522 new cases of leukemia per year.207 Only a small fraction of these will involve the prostate, and even fewer will be of primary prostatic origin.
Prostatic Involvement with Systemic Hematologic Malignancy Although tumors secondarily involving the prostate are not the subject of this chapter, we will briefly review prostatic involvement by hematologic malignancies (see also Chapter 49, Rare Lymphomas). Cases of both systemic leukemia and lymphoma have been reported to involve the prostate. In 1937, Jacobi et al. reported leukemic involvement of the prostate.303 Several other case reports and reviews of the literature have been published since then.304 – 308 Butler et al. reported on a large retrospective review of 4862 prostatectomies with six cases of leukemic infiltration identified.309 It appears that the predominant type of leukemia involving the prostate is chronic lymphocytic leukemia (CLL), with only a few reported cases of acute myeloid leukemia and even fewer with chronic myeloid leukemia. One interesting case of acute myelogenous leukemia affecting the prostate was reported by Thalhammer et al., the case of a 68-year-old man who developed granulocytic sarcoma (extramedullary leukemia) of the prostate as the initial site of relapse 9 years after complete remission.310 The predominance of CLL is not entirely unexpected because of the prevalence of this chronic disease compared to other leukemic subtypes. The exact subtypes of leukemia (and lymphoma) are difficult to attribute
57
because of different pathologists and different classification schemes over the years. Systemic lymphoma has also been described to involve the prostate. The largest series of prostatic lymphoma by Bostwick et al.311 describes 62 cases. In that series, secondary prostatic involvement was more common than primary prostatic lymphoma. Weimar et al. described a large retrospective series of 1068 patients with lymphoma and described genitourinary involvement.312 The genitourinary tract was involved in 6.7% of cases (49 cases of NHL, 23 cases of Hodgkin’s lymphoma), but prostatic involvement was not explicitly described in any case. A review of 6000 male autopsies revealed an incidence of 0.82% (49 patients) of prostatic lymphomatous involvement.313 In a review of extranodal lymphomas, only 3 of 1467 (0.2%) cases had documented prostatic involvement.314 Rosenberg reviewed 1269 cases of lymphosarcoma and only identified 2 (0.16%) with prostatic involvement.315 A series from MD Anderson Cancer Center reviewed 2928 untreated patients with lymphoma presenting to their center from 1980 through 1991 and found three cases (0.1%) of primary lymphoma of the prostate.316 Of 3446 cases of prostatic malignancies treated during that time, these 3 cases represented 0.09% of prostatic malignancies. Nearly all reported cases of prostatic lymphoma are of non-Hodgkin’s type, with only a few reported cases of Hodgkin’s disease involving the prostate. One such interesting case described by Klotz et al. reported a case of stage IV nodular sclerosing Hodgkin’s lymphoma with an initial response to surgery, chemotherapy, and radiation.317 A subsequent presentation with acute urinary retention revealed recurrence in the prostate, with subsequent long-term response to radiotherapy. The clinical presentation of secondary involvement of the prostate by hematologic malignancies is difficult to interpret. In a large surgical series all patients were symptomatic, although it is interesting to note that while there were six cases of leukemic infiltration into the prostate, eight other cases with systemic leukemia and prostatic obstructive symptoms were found to have leukemia-free prostates.309 Since the common types of leukemia with prostatic involvement are rare in people younger that 50 years old, it stands to reason that many patients with leukemia will have prostatic symptoms, even though on the basis of reported data only a minority of these will have leukemic involvement. Primary presentation of leukemia with prostatic symptoms has been reported.308 However, in modern times with the majority of patients with CLL (the most common type of leukemia with prostatic involvement by far) presenting with either peripheral lymphadenopathy and/or laboratory abnormalities, we expect this to be an even more uncommon phenomenon.
Primary Lymphoma of the Prostate In 1877, Coupland reported the first case of prostatic lymphoma.318 Since then, more cases and small series have been reported. In Bostwick’s original series319 of prostatic lymphoma, a modification of criteria described by King and Cox320 was used to define primary lymphoma of the prostate. The criteria are simply: (i) presenting symptoms attributable to prostatic enlargement, (ii) involvement of the
58
GENITOURINARY CANCER
prostate predominantly, with or without adjacent tissue, and (iii) absence of involvement of liver, spleen, or lymph nodes within 1 month of diagnosis of prostatic involvement.311,319 As these criteria have been used in the literature since then (sometimes referred to as the Bostwick and Mann criteria), we will follow the classification. It should be noted, however, that these criteria would exclude patients who on autopsy or biopsy are found to have lymphoma present only in their prostate if they did not present with symptoms of prostatic enlargement. Bostwick’s 1998 series of 62 patients includes 22 patients with primary lymphoma of the prostate.311 The most common histology amongst patients with primary prostatic lymphoma was diffuse large B cell lymphoma in 55%, followed by small lymphocytic lymphoma (CLL) in 18%. There were two cases of high-grade, Burkitt-like lymphoma. Several other cases or smaller series have also been reported in the last decade, fulfilling the same criteria as that for primary lymphoma of the prostate.316,321 – 331 The majority of these cases were also large cell lymphoma, with some small cell or mixed types with two additional cases of small noncleaved NHL, although the specific subtypes of NHL are not well described in all cases. There is emerging data regarding viruses and NHL, specifically human immunodeficiency virus (HIV), Epstein-Barr virus (EBV), and hepatitis C, but there are no studies in this regard for primary lymphoma of the prostate.
Clinical Presentation In the more recent Bostwick series, 91% (43 of 47) of all patients (including secondary prostatic lymphomatous involvement and unknown cases) in whom symptoms were known, presented with symptoms of prostatic enlargement.311 Nine patients (19%) had hematuria at presentation. Only three patients (6%) had B symptoms. Cystoscopically, there were no reported pathognomonic findings, with urethral luminal narrowing and bladder trabeculation as in nodular hyperplasia. PSA was known in 10 patients, and elevated above 4 ng mL−1 in two. In the 22 patients meeting the criteria for primary prostatic lymphoma, 16 (73%) developed extraprostatic sites of disease more than 1 month after diagnosis. Since symptoms need to be present to fulfill the criteria put forth for primary lymphoma of the prostate, all of the other recent case reports have presented with symptoms, again the most common being obstructive symptoms, with hematuria described not uncommonly. Most of the other cases have not reported PSA, although one case, from Australia, of a small cell lymphoma reported a PSA of 4.8 ng mL−1 with a prostatic nodule and a well-circumscribed hypoechoic lesion on ultrasound.328 While several of the case reports have described the ultrasonic appearance of the involved prostate, one case from the Mayo Clinic describes the ultrasound characteristics in detail.332 The prostate in a 39-year-old man appeared enlarged, but normal landmarks could be identified. Several areas of hypoechogenicity were identified, including the area of extraprostatic extension felt on digital rectal examination.
Prognosis In Bostwick’s original series, the median survival was 8 months overall (mean survival 2 months) and only 4 months (mean 8 months) for the primary prostatic lymphoma patients.319 In the follow-up series, 13 of 22 patients with primary prostatic lymphoma were known to have died, with 9 of them having lymphoma as their cause of death (median survival in these patients was 23 months).311 Most of the other case reports and series have quoted relatively short survival statistics, with several notable exceptions. There have been a few cases with long-term survival after local therapy (surgery and/or radiotherapy) as well as with chemotherapy and/or multimodality therapy, which will be discussed below.
Treatment Primary lymphoma of the prostate is, by definition, (at least at the time of clinical presentation) a local or regional disease and therefore, attempts at local therapy alone have been reported historically. Many of the cases and series reported are surgical series, with the diagnosis of lymphoma only becoming known after prostatectomy. Several case reports after prostatectomy have reported poor outcomes, however, at least one case of a large cell lymphoma removed by radical retropubic prostatectomy and no further treatment resulted in a disease-free survival of 13 years at the time of reporting.322 Likewise, there have been a few cases of patients treated with radiotherapy alone. Two cases resulted in rapid progression and death (5 and 11-month survival from diagnosis,322,329 but another case treated with radiotherapy alone resulted in 24-month disease-free survival at the time of reporting.321 Care should be taken to examine each case on an individual basis. Several important points should be noted when considering local therapy for primary lymphoma of the prostate. The pathologic subtype of lymphoma has often been overlooked in the historical literature when outcome has been reported. This may be illustrated by a case reported by Braslis et al. of a 65-year-old man with recurrent urinary tract infections found to have an abnormal result on digital rectal examination and a mildly elevated PSA level.328 A transrectal biopsy revealed small cell lymphoma. The patient received no treatment for this usually indolent lymphoma and was alive and well with no change on rectal examination, in the PSA level or the clinical symptoms at the time of reporting. As reported historically in the literature, indolent lymphomas such as this may have a 70% 10-year survival without impact by early aggressive therapy.333 Another important point to note is the natural history of lymphoma in general. While several cases of subtypes of true stage IE lymphoma may result in long-term survival with localized therapy alone, it is noteworthy that in the largest series, 16 of 22 patients (73%) developed extraprostatic sites of disease between 1 and 59 months after diagnosis.311 Only 7 of 22 patients (32%) received systemic therapy. In the series from MD Anderson Cancer Center, all three patients were treated with systemic chemotherapy, two with specific therapies for aggressive small noncleaved lymphoma
UNCOMMON CANCERS OF THE PROSTATE
and one with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) combination chemotherapy.316 All patients have done clinically well.
Recommendations For patients discovered to have a hematologic malignancy affecting their prostate, we recommend evaluation at a center with expertise in several areas, including hematology/oncology, urology, radiation oncology, and pathology. If available, enrollment in a clinical trial should be considered. After thorough review of the pathology, extensive workup for other sites of disease should be performed, including complete history and physical, laboratory evaluation, CT of the body, functional imaging with PET and/or gallium scans, and a bone marrow biopsy. If indicated, tests for infectious etiologies, such as HIV and/or hepatitis C should be considered. After workup, multidisciplinary evaluation should be performed. Taking into consideration the pathology (type and grade), stage, clinical status and symptoms, comorbidities, and patient preferences, a treatment plan should be rendered. High consideration should be given for systemic combination chemotherapy for intermediate to high-grade tumors, regardless of a negative staging workup. For patients with bulky local disease, a discussion about the risks and benefits of radiotherapy in addition to chemotherapy is reasonable on a case-by-case basis, extrapolating from results in bulky lymphomas of other sites. If given, we suggest either 3-D CRT or IMRT at 30–36 Gy with a complete response to chemotherapy, or 45–55 Gy with less than a complete response to chemotherapy. In any patient presenting in an unusual fashion, in particular, younger men with obstructive prostatic symptoms and/or hematuria, the diagnosis of lymphoma of the prostate should be considered. A transrectal ultrasound with biopsy may be diagnostic, prompting the workup as stated above. For the small subset of patients discovered to have lymphoma of the prostate only after prostatectomy, we recommend extensive workup (as above) to evaluate other sites of disease. It is not known whether “adjuvant” therapy for this subset of patients will affect outcome. However, given the historical poor prognosis for this subset of patients, especially those men with intermediate to high-grade lesions, referral to an experienced center is warranted and combination chemotherapy should be considered.
REFERENCES 1. Partin AW, Rodriguez R. The molecular biology, endocrinology, and physiology of the prostate and seminal vesicles. In Walsh PC (ed-in-chief), Retik AB, et al. (eds) Campbell’s Urology, 8th ed. Philadelphia, Pennsylvania: Saunders, 2002: 1237 – 1296. 2. Scher HI, et al. Cancer of the prostate. In Devita VT Jr, Hellman S, Rosenberg SA (eds) Cancer Principles and Practice of Oncology, 7th ed. Philadelphia, Pennsylvania: Lippincott Williams & Wilkins, 2005: 1192 – 1259. 3. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55: 10 – 30. 4. Mostofi FK, Price EB Jr. Atlas of Tumor Pathology:Tumors of the Male Genital System. Second Series. Fascicle 8. Washington, District of Columbia: Armed Forces Institute of Pathology, 1973: 177 – 259.
59
5. Sommers SC. Endocrine changes with prostatic carcinoma. Cancer 1957; 10: 345 – 58. 6. Pretl K. Zur Frage der Endokrinie der menschlichen Vorsfeherdruse. Virchows Arch A Pathol Anat 1944; 312: 392 – 404. 7. Feyrter F. Uber das urogenitale Helle-Zelien System des Menschen. Z Mikrosk Anat Forsch 1951; 57: 324 – 44. 8. diSant’Agnese PA, et al. Human prostatic endocrine-paracrine (APUD) cells. Arch Pathol Lab Med 1985; 109: 607 – 12. 9. Bonikos DS, Bensch KG. Endocrine cells of bronchial and bronchiolar epithelium. Am J Med 1977; 63: 765 – 71. 10. Azzopardi JG, Evans DJ. Argentaffin cells in carcinoma: differentiation from lipofuscin and melanin in prostatic epithelium. J Pathol 1971; 104: 247 – 51. 11. Berger CL, Goodwin G, Mendelsohn G, et al. Endocrine related biochemistry in the spectrum of human lung carcinoma. J Clin Endocrinol Metab 1981; 53: 422 – 9. 12. Schron DS, Gipson T, Mendelsohn G. The histogenesis of small cell carcinoma of the prostate. An immunohistochemical study. Cancer 1984; 53: 2478 – 80. 13. Spieth ME, Lin G, Nguyen TT. Diagnosis and treating small-cell carcinomas of prostatic origin. Clin Nucl Med 2002; 27: 11 – 7. 14. Wenk RE, et al. Ectopic ACTH, prostatic oat cell carcinoma and marked hypernatremia. Cancer 1977; 40: 773 – 8. 15. Ro JY, et al. Small cell carcinoma of the prostate. II. Immunohistochemical and electron microscopic studies of 18 cases. Cancer 1987; 59: 977 – 82. 16. Wise HM Jr, et al. Hyperadrenocorticism associated with ‘reactivated’ prostatic carcinoma. Surgery 1965; 57: 655 – 64. 17. Hall TC. Symptomatic hypokalaemic alkalosis in hyperadrenocorticism secondary to carcinoma of the prostate. Cancer 1968; 21: 190 – 2. 18. Newmark SR, Dlhuh RG, Bennett AH. Ectopic adrenocorticotropin syndrome with prostatic carcinoma. Urology 1973; 2: 666 – 8. 19. Lovern WJ, et al. Ectopic ACTH production in disseminated prostatic adenocarcinoma. Urology 1975; 5: 817 – 20. 20. Holland EA. Prostatic adenocarcinoma with ectopic ACTH production. Br J Urol 1978; 50: 538 – 41. 21. Hindson DA, Knight LL, Ocker JM. Small-cell carcinoma of prostate. Transient complete remission with chemotherapy. Urology 1985; 26: 182 – 4. 22. Ghandur-Mnaymneh L, Satterfield S, Block NL. Small cell carcinoma of the prostate gland with inappropriate antidiuretic hormone secretion: morphological, immunohistochemical and clinical expressions. J Urol 1986; 135: 1263 – 6. 23. Bleichner JC, Chun B, Klappenbach RS. Pure small cell carcinoma of the prostate with fatal liver metastasis. Arch Pathol Lab Med 1986; 110: 1041 – 4. 24. Van Haaften-Day C, et al. Xenografted small cell undifferentiated cancer of prostate: possible common origin with prostatic adenocarcinoma. Prostate 1988; 11: 271 – 9. 25. Tetu B, et al. Small cell carcinoma of the prostate, Part 1: A clinicopathologic study of 20 cases. Cancer 1987; 59: 1803 – 9. 26. Capella C, et al. The endocrine component of prostatic carcinomas, mixed adenocarcinoma-carcinoid tumors, and non-tumor prostate: histochemical and ultrastructural identification of the endocrine cells. Histopathology 1981; 5: 175 – 92. 27. Bono AV, Pozzi E. Endocrine-paracrine cells in prostatic carcinoma and clinical course of the disease. Eur Urol 1985; 11: 185 – 8. 28. Smedley HM, Brown C, Turner J. Ectopic ACTH-producing lung cancer presenting with prostatic metastases. Postgrad Med J 1983; 59: 371 – 2. 29. Hodge GB, Carson CC. Oat cell carcinoma of lung masquerading as prostatic carcinoma. Urology 1985; 25: 69 – 70. 30. Barkin J, et al. Hypercalcemia associated with cancer of prostate without bony metastases. Urology 1984; 24: 368 – 71. 31. Pruneri G, et al. Chromogranin A and B and secretogranin II in prostatic adenocarcinomas: neuroendocrine expression in patients untreated and treated with androgen deprivation therapy. Prostate 1998; 34: 113 – 20. 32. Vuitch MF, Mendelsohn G. Relationship of ectopic ACTH production to tumor differentiation. A morphologic and immunohistochemical study of prostatic carcinoma with Cushing’s syndrome. Cancer 1982; 47: 296 – 9.
60
GENITOURINARY CANCER
33. Slater D. Carcinoid tumour of the prostate associated with inappropriate ACTH secretion. Br J Urol 1985; 57: 591 – 2. 34. Sellwood RA, et al. Inappropriate secretion of antidiuretic hormone by carcinoma of the prostate. Br J Surg 1969; 56: 933 – 5. 35. Carey RM, et al. Ectopic secretion of corticotropin-releasing factor as a cause of Cushing’s syndrome. A clinical, morphologic and biochemical study. N Engl J Med 1984; 311: 13 – 20. 36. Jelbart ME, et al. Ectopic hormone production by a prostatic small cell carcinoma xenograft line. Mol Cell Endocrinol 1988; 55: 167 – 72. 37. Reeve JG, et al. Neuron specific enolase expression in carcinoma of the lung. Br J Cancer 1986; 53: 519 – 28. 38. Ellis DW, et al. Multiple immunoperoxidase markers in benign hyperplasia and adenocarcinoma of the prostate. Am J Clin Pathol 1984; 81: 279 – 84. 39. Heyderman E, Brown BM, Richardson TC. Epithelial markers in prostatic, bladder and colorectal cancer. An immunoperoxidase study of epithelial membrane antigen, carcino-embryonic antigen, and prostatic acid phosphatase. J Clin Pathol 1984; 37: 1363 – 9. 40. Ghazizadeh M, et al. Immunohistochemical detection of CEA in benign hyperplasia and adenocarcinoma of the prostate with monoclonal antibody. J Urol 1984; 131: 501 – 4. 41. Jelbart ME, et al. Site-specific growth of the prostate xenograft line UCRU-PR-2. Prostate 1989; 14: 163 – 75. 42. Sim SJ, et al. Serum calcitonin in small cell carcinoma of the prostate. Ann Clin Lab Sci 1996; 26: 487 – 95. 43. Sloane JP, Ormerod MG. Distribution of epithelial membrane antigen in normal and neoplastic tissues and its value in diagnostic tumor pathology. Cancer 1981; 47: 1786 – 95. 44. Ansari MA, et al. Diagnosis of carcinoid-line metastatic prostatic carcinoma by an immunoperoxidase method. Am J Clin Pathol 1981; 76: 94 – 8. 45. Ghali VS, Garcia RL. Prostatic adenocarcinoma with carcinoidal features producing adrenocorticotropic syndrome. Cancer 1984; 54: 1043 – 8. 46. Almagro UA. Argyrophilic prostatic carcinoma. Case report with literature review on prostatic carcinoid and ‘carcinoid-like’ prostatic carcinoma. Cancer 1985; 55: 608 – 14. 47. Smith CS. Small cell carcinoma of the prostate. J Urol 1985; 133: 371A, (abstract 1029). 48. Angelsen A, et al. Neuroendocrine differentiation in carcinomas of the prostate: do neuroendocrine serum markers reflect immunohistochemical findings? Prostate 1997; 30: 1 – 6. 49. Wu JT, et al. Serum chromogranin A: early detection of hormonal resistance in prostate cancer patients. J Clin Lab Anal 1998; 12: 20 – 5. 50. Berruti A, et al. Potential clinical value of circulating chromogranin A in patients with prostate carcinoma. Ann Oncol 2001; 12(Suppl 2): S153 – 7. 51. Isshiki S, et al. Chromogranin a concentration as a serum marker to predict prognosis after endocrine therapy for prostate cancer. J Urol 2002; 167: 512 – 5. 52. Kamiya N, et al. Pretreatment serum level of neuron specific enolase (NSE) as a prognostic factor in metastatic prostate cancer patients treated with endocrine therapy. Eur Urol 2003; 44: 309 – 14. 53. Osterling JE, Hauzeur CG, Farrow GM. Small cell anaplastic carcinoma of the prostate: a clinical, pathologic and immunohistochemical study of 27 patients. J Urol 1992; 147: 804 – 7. 54. Oshimura M, Sandberg AA. Isochromosome 17 in prostatic cancer. J Urol 1975; 114: 249 – 50. 55. Atkin NB, Baker MC. Chromosome 10 deletion in carcinoma of the prostate. N Engl J Med 1985; 312: 315. 56. Pittman S, et al. Flow cytometric and karyotypic analysis of a primary small cell carcinoma of the prostate: a xenografted cell line. Cancer Genet Cytogenet 1987; 26: 165 – 9. 57. Whang-Peng J, et al. A non-random chromosomal abnormality, del 3p (14 – 23), in human small cell lung cancer (SCCL). Cancer Genet Cytogenet 1982; 6: 119 – 34. 58. Wurster-Hill DH, et al. Cytogenetics of small cell carcinoma of the lung. Cancer Genet Cytogenet 1984; 13: 303 – 30. 59. Viola MV, et al. Expression of ras oncogene p2l in prostate cancer. N Engl J Med 1986; 314: 133 – 7.
60. Rijnders AWM, et al. Expression of cellular oncogenes in human prostatic carcinoma cell lines. Biochem Biophys Res Commun 1985; 132: 548 – 54. 61. Plummer H, et al. c-myc Expression correlates with suppression of c-kit protooncogene expression in small cell lung cancer cell lines. Cancer Res 1993; 53: 4337 – 42. 62. Raghavan D, et al. The interpretation of marker protein assays: a critical appraisal in clinical studies and a xenograft model. Br J Cancer 1980; 41(suppl IV): 191 – 4. 63. Noordzij MS, et al. Neuroendocrine differentiation in human prostatic tumor models. Am J Pathol 1991; 149: 859 – 71. 64. Perez-Stable C, et al. Prostate cancer progression, metastasis, and gene expression in transgenic mice. Cancer Res 1997; 57: 900 – 6. 65. Cohen R, et al. The neuroendocrine cell population of the human prostate gland. J Urol 1993; 150: 365 – 8. 66. Pearse AGE, Pollak JM. Neural crest origin of the endocrine polypeptide (APUD) cell of the gastrointestinal cell and pancreas. Gut 1971; 12: 783 – 8. 67. Pearse AG, Takor T. Embryology of the diffuse neuroendocrine system and its relationship to the common peptides. Fed Proc 1979; 38: 2288 – 94. 68. LeDouarin N, Tellet MA. The migration of neural crest cells to the wall of the digestive tract in the avian embryo. J Embryol Exp Morphol 1972; 30: 31 – 48. 69. Schron DS, Gipson T, Mendelsohn g. The histogenesis of small cell carcinoma of the pancreas: an immunohistological study. Cancer 1984; 53: 2478 – 80. 70. Helpap BB, Kollerman J, Ochler U. Neuroendocrine differentiation in prostatic carcinoma : histogenesis, biology, clinical relevance, and future therapeutic perspectives. Urol Int 1999; 62: 133 – 8. 71. Duguid JB, Kennedy AM. Oat cell tumors of mediastinal glands. J Pathol Bacteriol 1930; 33: 93 – 9. 72. Brown JR, et al. Small-cell cancers, an unusual reaction to chemotherapy. J Clin Oncol 2003; 21: 2437 – 43. 73. Dauge MC, Delmas V. APUD type endocrine tumor of the prostate. Incidence and prognosis in association with adenocarcinoma. Prog Clin Biol Res 1987; 243A: 529. 74. Amato RJ, et al. Chemotherapy for small cell carcinoma of prostatic origin. J Urol 1992; 147: 935 – 7. 75. Montasser AY, Ong MG, Mehta VT. Carcinoid tumor of the prostate associated with adenocarcinoma. Cancer 1979; 44: 307 – 10. 76. Wasserstein PW, Goldman RL. Diffuse carcinoid of prostate. Urology 1981; 18: 407 – 9. 77. McCutcheon IE, Eng DY, Logothetis CJ. Brain metastases from prostate carcinoma. Antemortem recognition and outcome after treatment. Cancer 1999; 86: 2301 – 11. 78. Oesterling JE, Hauzeur CG, Farrow GM. Small cell anaplastic carcinoma of the prostate: a clinical pathological and immunohistochemical study of 27 patients. J Urol 1992; 147: 804 – 7. 79. Bradley JD, et al. Positron emission tomography in limited-stage small-cell lung cancer: A prospective study. J Clin Oncol 2004; 22: 3248 – 54. 80. Brink I, et al. Impact of [18 F]FDG-PET on the primary staging of small-cell lung cancer. Eur J Nucl Med Mol Imaging 2004; 31: 1614 – 20. 81. Torii K, et al. A case of small cell carcinoma of the esophagus detected incidentally by FDG-PET. Ann Nucl Med 2004; 18: 699 – 702. 82. Chander S, et al. Small cell carcinoma of the parotid gland: evaluation with FDG PET imaging. Clin Nucl Med 2004; 29: 502 – 3. 83. Bishop JF, Raghavan D, Stuart-Harris R, et al. Carboplatin (CBDCA, JM-8) and VP-16-213 in previously untreated patients with small cell lung cancer. J Clin Oncol 1987; 5: 1574 – 9. 84. Seifker-Radtke AO, et al. Evidence supporting preoperative chemotherapy for small cell carcinoma of the bladder: a retrospective review of the MD Anderson cancer experience. J Urol 2004; 172: 481 – 4. 85. Quek ML, et al. Radical cystectomy for primary neuroendocrine tumors of the bladder: The University of Southern California experience. J Urol 2005; 174: 93 – 6. 86. Murren JR, Turrisi AT, Pass HI. Small cell lung cancer. In Devita VT Jr, Hellman S, Rosenberg SA (eds) Cancer Principles and Practice of Oncology, 7th ed. Philadelphia, Pennsylvania: Lippincott Williams & Wilkins, 2005: 810 – 843.
UNCOMMON CANCERS OF THE PROSTATE 87. Creaven PJ, et al. Paclitaxel and carboplatin in early phase studies: Roswell Park Cancer Institute experience in the subset of patients with lung cancer. Semin Oncol 1997; 24(suppl 12): 138 – 43. 88. Galanis E, Frytak S, Lloyd RV. Extrapulmonary small cell carcinoma. Cancer 1997; 79: 1729 – 36. 89. Papandreou CN, et al. Results of a phase II study with doxorubicin, etoposide, and cisplatin in patients with small-cell carcinoma of the prostate. J Clin Oncol 2002; 20: 3072 – 80. 90. Brevini TA, Bianchi R, Motta M. Direct inhibitory effect of Somatostatin on the growth of the human prostate cancer cell line LNCaP: possible mechanisms of action. J Clin Endocrinol Metab 1993; 77: 626. 91. Spieth ME, Lin G, Nguyen TT. Diagnosis and treating small-cell carcinomas of prostatic origin. Clin Nucl Med 2002; 27: 11 – 7. 92. Mackey JR, et al. Genitourinary small cell carcinoma: determination of clinical and therapeutic factors associated with survival. J Urol 1998; 159: 1624 – 9. 93. Weinstein MH, et al. Neuroendocrine differentiation in prostate cancer: enhanced prediction of progression after radical prostatectomy. Hum Pathol 1996; 27: 683 – 7. 94. Stirling WC, Ash JE. Sarcoma of the prostate. J Urol 1939; 41: 515 – 33. 95. Schmidt JD, Welch MJ. Sarcoma of the prostate. Cancer 1976; 37: 1908 – 12. 96. Russo P, et al. Adult urological sarcoma. J Urol 1992; 147: 1032 – 6. 97. Smith BH, Dehner JP. Sarcoma of the prostate gland. Am J Clin Pathol 1972; 58: 43 – 50. 98. Sexton WJ, et al. Adult prostate sarcoma: The MD Anderson cancer center experience. J Urol 2001; 166: 521 – 5. 99. Coindre JM, et al. Reproducibility of a histopathologic grading system for adult soft tissue sarcoma. Cancer 1986; 58: 306 – 9. 100. Shiraki M, et al. Pathologic analysis of advanced adult soft tissue sarcomas, bone sarcomas, and mesotheliomas. The Eastern Cooperative Oncology Group (ECOG) experience. Cancer 1989; 64: 484 – 90. 101. Alvegard TA, Berg NO. Histopathology peer review of high-grade soft tissue sarcoma: the Scandinavian Sarcoma Group experience. J Clin Oncol 1989; 7: 1845 – 51. 102. Fletcher CDM, et al. Clinicopathologic re-evaluation of 100 malignant fibrous histiocytomas: Prognostic relevance of subclassification. J Clin Oncol 2001; 19: 3045 – 50. 103. Enzinger FM, Weiss SW. General considerations. In Enzinger FM, Weiss SW (eds) Soft Tissue Sarcomas. St. Louis, Missouri: CV Mosby, 1988: 1 – 17. 104. Reddick RL, Michellitch H, Triche TJ. Malignant soft tissue tumors (malignant fibrous histiocytoma, pleomorphic liposarcoma, and pleomorphic rhabdomyosarcoma): an electron microscope study. Hum Pathol 1979; 10: 327 – 43. 105. Enzinger FM, Weiss SW. Immunohistochemistry of soft tissue lesions. In Enzinger FM, Weiss SW (eds) Soft Tissue Sarcomas. St. Louis, Missouri: CV Mosby, 1988: 93 – 101. 106. Sandberg AA, Turc-Carel C, Gemmill RM. Chromosomes in solid tumors and beyond. Cancer Res 1988; 48: 1049 – 59. 107. Mitelman F, Kaneko Y, Trent JM. Report of the committee on chromosome changes in neoplasia. Cytogenet Cell Genet 1990; 55: 358 – 86. 108. Koufos A, et al. Loss of heterozygosity in three embryonal tumours suggests a common pathogenetic mechanism. Nature 1985; 316: 330 – 4. 109. Khan J, et al. Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks. Nat Med 2001; 7: 673 – 9. 110. Nielsen TO, et al. Molecular characterization of soft tissue tumors: a gene expression study. Lancet 2002; 359: 1301 – 7. 111. Khan J, et al. CDNA microarrays detect activation of a myogenic transcription program by the PAC3-FKHR fusion oncogene. Proc Natl Acad Sci U S A 1999; 96: 13264 – 9. 112. Lessnick SL, Dacwag CS, Golub TR. The Ewing’s sarcoma oncoprotein EWS/FLI induces a p53-dependent growth arrest in primary human fibroblasts. Cancer Cell 2002; 1: 393 – 401.
61
113. Allander SV, et al. Gastrointestinal stromal tumors with KIT mutations exhibit a remarkably homogeneous gene expression profile. Cancer Res 2001; 61: 8624 – 8. 114. Allander SV, et al. Expression of synovial sarcoma by cDNA microarrays: Association of ERBB2, IGHBP2, and ELF3 with epithelial differentiation. Am J Pathol 2002; 161: 1587 – 95. 115. Bain GO, et al. Malignant fibrous histiocytoma of prostate gland. Urology 1985; 26: 89 – 91. 116. Hulbert JC, Rodriguez PN, Cummings KB. Perineal liposarcoma: diagnosis and management. J Urol 1984; 131: 1185 – 7. 117. Smith DM, et al. Angiosarcoma of the prostate: report of 2 cases and review of the literature. J Urol 1986; 135: 382 – 4. 118. Cea PC, Ward JN. Sarcoma of the prostate 13 years after suprapubic prostatectomy. J Urol 1977; 117: 129 – 30. 119. Mottola A, et al. Leiomyosarcoma of the prostate. Eur Urol 1985; 11: 131 – 3. 120. Cheville JC, et al. Leiomyosarcoma of the prostate. Report of 23 cases. Cancer 1995; 76: 1422 – 7. 121. Iwasaki H, et al. Synovial sarcoma of the prostate with t(x;18) (p11.2;q11.2). Am J Surg Pathol 1999; 23: 220 – 6. 122. Williams D, et al. Synovial sarcoma of the prostate. J Urol 2004; 171: 2376. 123. Peterson LJ, Paulson DF. Rhabdomyosarcoma in adult prostate. Urology 1974; 3: 689 – 92. 124. Dupree WB, Fisher C. Rhabdomyosarcoma of prostate in adult. Longterm survival and problem of histologic diagnosis. Urology 1982; 19: 80 – 2. 125. Keenan DJM, Graham WH. Embryonal rhabdomyosarcoma of the prostatic-urethral region in an adult. Br J Urol 1985; 57: 241. 126. Palmer MA, Viswanath S, Desmond AD. Adult prostatic rhabdomyosarcoma. Br J Urol 1993; 71: 489 – 90. 127. Miettinen M. Rhabdomyosarcoma in patients older than 40 years of age. Cancer 1988; 62: 2060 – 5. 128. Waring PM, Newland RC. Prostatic embryonal rhabdomyosarcoma in adults. A clinicopathologic review. Cancer 1992; 69: 755 – 62. 129. Lopez-Beltran A, et al. Malignant phyllodes tumor of prostate. Urology 1990; 35: 164 – 7. 130. Manivel C, et al. Cystosarcoma phyllodes of the prostate. Arch Pathol Lab Med 1986; 110: 534 – 8. 131. Reese JH, et al. Phyllodes type of atypical prostatic hyperplasia: a report of 3 new cases. J Urol 1987; 138: 623 – 6. 132. Yokota T, et al. Malignant cystosarcoma phyllodes of prostate. Acta Pathol Jpn 1984; 34: 663 – 8. 133. Gueft B, Walsh MA. Malignant prostatic cystosarcoma phyllodes. N Y State J Med 1975; 75: 2226 – 8. 134. Yum M, Miller JC, Agrawal BL. Leiomyosarcoma arising in atypical fibromuscular hyperplasia (phyllodes tumor) of the prostate with distant metastasis. Cancer 1991; 68: 910 – 5. 135. Tetu B, et al. Atypical spindle cell lesions of the prostate. Semin Diagn Pathol 1988; 5: 284 – 93. 136. Gaudin PB, Rosai J, Epstein JI. Sarcomas and related proliferative lesions of specialized prostatic stroma: A clinicopathologic study of 22 cases. Am J Surg Pathol 1998; 22: 148 – 62. 137. Watanabe M, et al. Malignant phyllodes tumor of the prostate: retrospective review of specimens obtained by sequential transurethral resection. Pathol Int 2002; 52: 777 – 83. 138. Kim HS, et al. Malignant phyllodes tumor of prostate. Pathol Int 1999; 49: 1105 – 8. 139. Schapmans S, et al. Phyllodes tumor of the prostate. A case report and review of the literature. Eur Urol 2000; 38: 649 – 53. 140. Lam KC, Yeo W. Chemotherapy induced complete remission in malignant phyllodes tumor of the prostate metastasizing to the lung. J Urol 2002; 168: 1104 – 5. 141. Tijare JR, Shrikhande AV, Shrikhande VV. Phyllodes type of atypical prostatic hyperplasia. J Urol 1999; 162: 803 – 4. 142. Yamamoto S, et al. Malignant phyllodes tumor of the prostate. Int J Urol 2000; 7: 378 – 81. 143. Young JF, Jensen PE, Wiley CA. Malignant phyllodes tumor of prostate. A case report with immunohistochemical and ultrastructural studies. Arch Pathol Lab Med 1992; 116: 296 – 9. 144. Bostwick DG, et al. Phyllodes tumor of the prostate: Long-term followup study of 23 cases. J Urol 2004; 172: 894 – 9.
62
GENITOURINARY CANCER
145. Zenklusen HR, et al. Carcinosarcoma of the prostate in combination with adenocarcinoma of the prostate and adenocarcinoma of the seminal vesicles. A case report with immunocytochemical analysis and review of the literature. Cancer 1990; 66: 998 – 1001. 146. Wick MR, et al. Prostatic carcinosarcomas. Clinical, histologic, and immunohistochemical data on two cases, with a review of the literature. Am J Clin Pathol 1989; 92: 131 – 9. 147. Ginesin Y, et al. Carcinosarcoma of the prostate. Eur Urol 1986; 12: 441 – 2. 148. Krastanova LJ, Addonizio JC. Carcinosarcoma of prostate. Urology 1981; 18: 85 – 8. 149. Tannenbaum M. Carcinoma with sarcomatoid changes or carcinosarcoma of prostate. Urology 1975; 6: 91 – 3. 150. Quay SC, Proppe KH. Carcinosarcoma of the prostate: case report and review of the literature. J Urol 1981; 125: 436 – 8. 151. Martin SA, et al. Carcinosarcoma of the prostate: report of a case with ultrastructural observations. J Urol 1979; 122: 709 – 11. 152. Hokamura K, et al. Carcinosarcoma of the prostate. Acta Pathol Jpn 1985; 35: 481 – 7. 153. Lindboe CF, Mjones J. Carcinosarcoma of prostate. Immunohistochemical and ultrastructural observations. Urology 1992; 40: 376 – 80. 154. Nazeer T, et al. Prostatic carcinosarcoma: Case report and review of literature. J Urol 1991; 146: 1370 – 3. 155. Kubosawa H, et al. Carcinosarcoma of the prostate. Acta Pathol Jpn 1993; 43: 209 – 14. 156. Shannon RL, et al. Sarcomatoid carcinoma of the prostate. A clinicopathologic study of 12 patients. Cancer 1992; 69: 2676 – 82. 157. Fukawa T, et al. Prostatic carcinosarcoma: A case report and review of literature. Int J Urol 2003; 10: 108 – 13. 158. Dundore PA, et al. Carcinosarcoma of the prostate: Report of 21 cases. Cancer 1995; 76: 1035 – 42. 159. Mostofi FK, Price EB. Malignant tumors of the prostate. Atlas of Tumor Pathology: Tumors of Male Genital System. series 2, part 8. Washington, District of Columbia: Armed Forces Institute of Pathology, 1973: 257 – 258. 160. Chung LW, et al. Co-inoculation of tumorigenic rat prostate mesenchymal cells with non-tumorigenic epithelial cells results in the development of carcinosarcoma in syngeneic and athymic animals. Int J Cancer 1989; 43: 1179 – 87. 161. Pierce SM, et al. Long-term radiation complications following conservative surgery (CS) and radiation therapy (RT) in patients with early stage breast cancer. Int J Radiat Oncol Biol Phys 1992; 23: 915 – 23. 162. Mark RJ, et al. Postirradiation sarcomas. A single-institution study and review of the literature. Cancer 1994; 73: 2653 – 62. 163. Amendola BE, et al. Radiation-associated sarcoma: A review of 23 patients with postradiation sarcoma over a 50-year period. Am J Clin Oncol 1989; 12: 411 – 5. 164. Cha C, et al. Long-term results with resection of radiation-induced soft tissue sarcomas. Ann Surg 2004; 239: 903 – 10. 165. Laskin WB, Silverman TA, Enzinger FM. Postradiation soft tissue sarcomas. An analysis of 53 cases. Cancer 1988; 62: 2330 – 40. 166. Huvos AG, et al. Postradiation osteogenic sarcoma of bone and soft tissues. A clinicopathologic study of 66 patients. Cancer 1985; 55: 1244 – 55. 167. Scully JM, et al. Radiation-induced prostatic sarcoma: A case report. J Urol 19909; 144: 746 – 8. 168. McKenzie M, et al. Postirradiation sarcoma after external beam radiation for localized adenocarcinoma of the prostate: Report of three cases. Urology 1999; 53: 1228. 169. Prevost JB, et al. Post irradiation sarcoma after external bean radiation therapy for localized adenocarcinoma of the prostate. Tumori 2004; 90: 618 – 21. 170. Terris MK. Transrectal ultrasound appearance of radiation-induced prostatic sarcoma. Prostate 1998; 37: 182 – 6. 171. Joerger M, et al. Postradiation high-grade myofibroblastic sarcoma of the prostate – a rare entity of prostatic tumors – responding to liposomal doxorubicin. Onkologie 2002; 25: 558 – 61. 172. Nishiyama T, et al. Osteogenic sarcoma of the prostate. Int J Urol 2001; 8: 199 – 201.
173. Canfield SE, et al. Postradiation prostatic sarcoma: de novo carcinogenesis or dedifferentiation of prostatic adenocarcinoma? Tech Urol 2001; 7: 294 – 5. 174. Varghese SL, Grossfeld GD. The prostate gland: malignancies other than adenocarcinomas. Radiol Clin North Am 2000; 38: 179 – 202. 175. Geoffray A, et al. Ultrasonography and computed tomography for diagnosis and follow-up of pelvic rhabdomyosarcomas in children. Pediatr Radiol 1987; 17: 132 – 6. 176. Bahnson RR, et al. Ultrasonography and diagnosis of pediatric genitourinary rhabdomyosarcoma. Urology 1989; 33: 64 – 8. 177. Terris MK, et al. Transrectal ultrasound in the evaluation of rhabdomyosarcoma involving the prostate. Br J Urol 1994; 74: 341 – 4. 178. Tehranzadeh J, et al. Comparison of CT and MR imaging in musculoskeletal neoplasms. J Comput Assist Tomogr 1989; 13: 466 – 72. 179. Sundaram M, McLeod RA. MR imaging of tumor and tumorlike lesions of bone and soft tissue. AJR Am J Roentgenol 1990; 155: 817 – 24. 180. Bartolozzi C, et al. Rhabdomyosarcoma of the prostate: MR findings. AJR Am J Roentgenol 1988; 150: 1333 – 4. 181. Kattan MW, Leung DHY, Brennan MF. Postoperative nomogram for 12-year sarcoma-specific death. J Clin Oncol 2002; 20: 791 – 6. 182. Borden EC, et al. Soft tissue sarcomas of adults: State of the translational science. Clin Cancer Res 2003; 9: 1941 – 56. 183. Markhede G, Angervall L, Stener B. A multivariate analysis of the prognosis after surgical treatment of malignant soft-tissue sarcomas. Cancer 1982; 49: 1721 – 33. 184. Tsujimoto M, et al. Multivariate analysis for histologic prognostic factors in soft tissue sarcomas. Cancer 1988; 62: 994 – 8. 185. Rooser B, et al. Prognostication in soft tissue sarcoma. A model with four risk factors. Cancer 1988; 61: 817 – 23. 186. Heslin MJ, et al. Prognostic factors associated with long-term survival for retroperitoneal sarcoma: implications for management. J Clin Oncol 1997; 15: 2832 – 9. 187. Pisters PWT, et al. Analysis of prognostic factors in 1,041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol 1996; 14: 1679 – 89. 188. American Joint Committee on Cancer. Soft tissue sarcoma. In AJCC Cancer Staging Manual, 6th ed. Philadelphia, Pennsylvania, Lippincott-Raven, 2002: 221 – 228. 189. Levine EA, et al. Evaluation of newer prognostic markers for adult soft tissue sarcomas. J Clin Oncol 1997; 15: 3249 – 57. 190. Yang JC, et al. Randomized prospective study of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas of the extremity. J Clin Oncol 1998; 16: 197 – 203. 191. Zagars GK, et al. Preoperative vs. postoperative radiation therapy for soft tissue sarcoma: a retrospective comparative evaluation of disease outcome. Int J Radiat Oncol Biol Phys 2003; 56: 482 – 8. 192. Potter DA, et al. Patterns of recurrence in patients with high-grade soft-tissue sarcomas. J Clin Oncol 1985; 3: 353 – 66. 193. Jaques DP, et al. Management of primary and recurrent soft-tissue sarcoma of the retroperitoneum. Ann Surg 1990; 212: 51 – 9. 194. Glenn J, et al. Results of multimodality therapy of resectable softtissue sarcomas of the retroperitoneum. Surgery 1985; 97: 316 – 25. 195. Farhood AI, et al. Soft tissue sarcomas of the head and neck in adults. Am J Surg 1990; 160: 365 – 9. 196. Meijer S, et al. Primary colorectal sarcoma. A retrospective review and prognostic factor study of 50 consecutive patients. Arch Surg 1990; 125: 1163 – 8. 197. Maurer HM, et al. The intergroup rhabdomyosarcoma study-I. A final report. Cancer 1988; 61: 209 – 20. 198. Maurer HM, et al. The intergroup rhabdomyosarcoma study-II. Cancer 1993; 71: 1904 – 22. 199. Crist W, et al. The third intergroup rhabdomyosarcoma study. J Clin Oncol 1995; 13: 610 – 30. 200. Crist WM, et al. Intergroup rhabdomyosarcoma study-IV: results for patients with nonmetastatic disease. J Clin Oncol 2001; 19: 3091 – 102. 201. Arndt C, et al. Does bladder preservation (as a surgical principle) lead to retaining bladder function in bladder/prostate rhabdomyosarcoma? Results from Intergroup Rhabdomyosarcoma study IV. J Urol 2004; 171: 2396 – 403.
UNCOMMON CANCERS OF THE PROSTATE 202. Brennan MF, et al. Soft tissue sarcomas. In Devita VT Jr, Hellman S, Rosenberg SA (eds). Cancer Principles and Practice of Oncology, 7th edition. Philadelphia, Pennsylvania: Lippincott Williams & Wilkins, 2005: 1581 – 1637. 203. Sarcoma Meta-analysis Collaboration (SMAC). Adjuvant chemotherapy for localized respectable soft tissue sarcoma in adults. Cochrane Database Syst Rev 2000; 2: CD001419. 204. Frustaci S, et al. Adjuvant chemotherapy for adult soft tissue sarcomas of the extremities and girdles: results of the Italian randomized cooperative trial. J Clin Oncol 2001; 19: 1238 – 47. 205. Petrioli R, et al. Adjuvant epirubicin with or without ifosfamide for adult soft-tissue sarcoma. Am J Clin Oncol 2002; 25: 468 – 73. 206. Glenn J, et al. A randomized, prospective trial of adjuvant chemotherapy in adults with soft tissue sarcomas of the head and neck, breast, and trunk. Cancer 1985; 55: 1206 – 14. 207. Parkin DM, Bray F, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin 2005; 55: 74 – 108. 208. Melicow MM, Hollowell JW. Intra-urothelial cancer: carcinoma in situ, Bowen’s disease of the urinary system: discussion of thirty cases. J Urol 1952; 68: 763 – 72. 209. Kahler JE. Carcinoma of the prostate gland: a pathologic study. J Urol 1939; 41: 557. 210. Thompson GJ. Transurethral resection of malignant lesions of the prostate gland. JAMA 1942; 120: 1105. 211. Ortega LG, Whitmore WF, Murphy AI. In situ carcinoma of the prostate with intraepithelial extension into the urethra and bladder. Cancer 1953; 6: 898. 212. Salazar G, Frable WJ. Extramammary Paget’s disease: a case involving the prostatic urethra. Am J Clin Pathol 1969; 52: 607 – 12. 213. Chesterman FC, Franks LM. Intra-epithelial carcinoma of prostatic urethra, peri-urethral glands, and prostatic ducts (Bowen’s disease of urinary epithelium). Br J Cancer 1956; 10: 223 – 5. 214. Ende N, Woods LP, Shelley HS. Carcinoma originating in ducts surrounding the prostatic urethra. Am J Clin Pathol 1963; 40: 183 – 9. 215. Karpas CM, Moumgis B. Primary transitional cell carcinoma of prostate gland: possible pathogenesis and relationship to reserve cell hyperplasia of prostatic periurethral ducts. J Urol 1969; 101: 201 – 5. 216. Bates HR. Transitional cell carcinoma of the prostate. J Urol 1969; 101: 206 – 7. 217. Johnson DE, Hogan JM, Ayala AG. Transitional cell carcinoma of the prostate. A clinical morphological study. Cancer 1972; 29: 287 – 93. 218. Seemayer TA, et al. Further observations on carcinoma in situ of the urinary bladder: silent but extensive intraprostatic involvement. Cancer 1975; 36: 514 – 20. 219. Grabstald H. Prostatic biopsy in selected patients with carcinoma in situ of the bladder: preliminary report. J Urol 1984; 132: 1117 – 8. 220. Rhamy RK, Buchanan RD, Spalding MJ. Intraductal carcinoma of the prostate gland. J Urol 1973; 109: 457 – 60. 221. Montie JE, et al. The significance and management of transitional cell carcinoma of the prostate. Semin Urol 1990; 8: 262 – 8. 222. Rubenstein AB, Rubnitz ME. Transitional cell carcinoma of the prostate. Cancer 1969; 24: 543 – 6. 223. Montie JE, Mirsky H, Levin H. Transitional cell carcinoma of the prostate in a series of cystectomies: incidence and staging problems. J Urol 1986; 135: 243A, (abstract 557). 224. Matzkin H, Soloway MS, Hardeman S. Transitional cell carcinoma of the prostate. J Urol 1991; 146: 1207 – 12. 225. Albert PS, Mallouh C, Nagamatsu GR. Transitional-cell carcinoma of prostate: an enigma. Urology 1973; 2: 128 – 30. 226. Goebbels R, et al. Urothelial carcinoma of the prostate. Appl Pathol 1985; 3: 242 – 54. 227. Zincke H, Utz DC, Farrow GM. Review of the Mayo clinic experience with carcinoma in situ. Urology 1985; 26(4 Suppl): 39 – 46. 228. Tannenbaum M. Transitional cell carcinoma of prostate. Urology 1975; 5: 674 – 8. 229. Dhom G, Mohr G. Urothel-Carcinome in der Prostata. Urologe A 1977; 16: 70 – 2. 230. Pagano F, et al. Is stage pT4a (D1) reliable in assessing transitional cell carcinoma involvement of the prostate in patients with a concurrent bladder cancer? A necessary distinction for contiguous or noncontiguous involvement. J Urol 1996; 155: 244 – 7.
63
231. Schellhammer PF, Bean MA, Whitmore WF. Prostatic involvement by transitional cell carcinoma: pathogenesis, patterns and prognosis. J Urol 1977; 118: 399 – 403. 232. Wood DP, et al. Transitional cell carcinoma of the prostate in cystoprostatectomy specimens removed for bladder cancer. J Urol 1989; 141: 346 – 9. 233. Honda N, et al. Clinical study of transitional cell carcinoma of the prostate associated with bladder transitional cell carcinoma. Int J Urol 2001; 8: 662 – 8. 234. Esrig D, et al. Transitional cell carcinoma involving the prostate with a proposed staging classification for stromal invasion. J Urol 1996; 156: 1071 – 6. 235. Kirk D, et al. Transitional cell carcinoma involving the prostate – an unfavorable prognostic sign in the management of bladder cancer?. Br J Urol 1981; 53: 610. 236. Hudson MA, Catalona WJ. Urothelial tumors of the bladder, upper tracts, and prostate. In Gillenwater JY et al. (eds) Adult and Pediatric Urology. St Louis, Missouri: Mosby-Year Book, 1996. 237. Mevorach RA, et al. Human papillomavirus type 6 in grade I transitional cell carcinoma of the urethra. J Urol 1990; 143: 126. 238. Monseur J, et al. Asbestose du col v´esical et de la prostate. J Urol 1986; 92: 17. 239. Ullmann AS, Ross OA. Hyperplasia, atypism, and carcinoma in situ in prostatic periurethral glands. Am J Clin Pathol 1967; 47: 497. 240. Mahadevia PS, Koss LG, Tar IJ. Prostatic involvement in bladder cancer: prostate mapping in 20 cystoprostatectomy specimens. Cancer 1986; 58: 2096. 241. Montie JE, et al. Transitional cell carcinoma in situ of the seminal vesicles: 8 cases with discussion of pathogenesis, and clinical and biological implications. J Urol 1997; 158: 1895. 242. Dube VE, Farrow GM, Greene LF. Prostatic adenocarcinoma of ductal origin. Cancer 1973; 32: 402. 243. Aydin F. Endometrioid adenocarcinoma of prostatic urethra presenting with anterior urethral implantation. Urology 1993; 41: 91. 244. Bostwick DG. Neoplasms of the prostate. In Bostwick DG, Eble JN. (eds) Urologic Surgical Pathology. St. Louis, Missouri: Mosby-Year Book, 1997. 245. Sawczuk I, et al. Primary transitional cell carcinoma of prostatic periurethral ducts. Urology 1985; 25: 339. 246. Greene LF, O’Dea MJ, Dockerty MB. Primary transitional cell carcinoma of the prostate. J Urol 1976; 116: 761. 247. Murphy WM, Gaeta JF. Diseases of the prostate gland and seminal vesicals. In Murphy WM (ed) Urological Pathology. Philadelphia, Pennsylvania: WB Saunders, 1989. 248. Bryan RL, et al. The significance of prostatic urothelial dysplasia. Histopathology 1993; 22: 501. 249. Wishnow KI, Ro JY. Importance of early treatment of transitional cell carcinoma of prostatic ducts. Urology 1988; 32: 11. 250. Prout GR, et al. Carcinoma in situ of the urinary bladder with and without associated vesical neoplasms. Cancer 1983; 52: 524. 251. Cheville JC, et al. Transitional cell carcinoma of the prostate: clinicopathologic study of 50 cases. Cancer 1998; 82: 703. 252. Freeman JA, et al. Management of the patient with bladder cancer: urethral recurrence. Urol Clin North Am 1994; 21: 645. 253. Freeman JA, et al. Urethral recurrence in patients with orthotopic ileal neobladders. J Urol 1996; 156: 1615. 254. Algaba F, et al. Transitional cell carcinoma of the prostate. Eur Urol 1985; 11: 87 – 90. 255. Winfield HN, Reddy PK, Lange PH. Coexisting adenocarcinoma of prostate in patients undergoing cystoprostatectomy for bladder cancer. Urology 1987; 30: 100. 256. Pritchett TR, et al. Unsuspected prostatic adenocarcinoma in patients who have undergone radical cystoprostatectomy for transitional cell carcinoma of the bladder. J Urol 1988; 139: 1214. 257. Dhom G. Histopathology of prostate carcinoma: diagnosis and differential diagnosis. Pathol Res Pract 1985; 179: 277. 258. Marshall VF. The relation of the preoperative estimate to the pathologic demonstration of the extent of vesical neoplasms. J Urol 1952; 68: 714. 259. Samsonov VA, Kolomoitsev SV. Dimorphic (transitional-anaplastic) ductal cancer of the prostate. Arkh Patol 1984; 46: 71.
64
GENITOURINARY CANCER
260. Lemberger RJ, et al. Carcinoma of the prostate of ductal origin. Br J Urol 1984; 56: 706. 261. Zincke H, Utz DC, Farrow GM. Review of the Mayo clinic experience with carcinoma in situ. Urology 1985; 26: 39. 262. Nicolaisen GS, Williams RD. Primary transitional cell carcinoma of the prostate. Urology 1984; 24: 544. 263. Wolfe JHN, Lloyd-Davies RW. The management of transitional cell carcinoma of the prostate. Br J Urol 1981; 53: 253. 264. Razvi M, Firfer R, Berkson B. Occult transitional cell carcinoma of the prostate presenting as skin metastasis. J Urol 1975; 113: 734. 265. Wendelken JR, et al. Transitional cell carcinoma: cause of refractory cancer of the prostate. Urology 1979; 13: 557. 266. Epstein NA. The cytologic appearance of metastatic transitional cell carcinoma. Acta Cytol 1977; 21: 723. 267. Griffiths GJ, et al. The ultrasound appearances of prostatic cancer with histological correlation. Clin Radiol 1987; 38: 219. 268. Terris MK, Villers A, Freiha FS. Transrectal ultrasound appearance of transitional cell carcinoma involving the prostate. J Urol 1990; 143: 953. 269. Beer M, Schmidt H, Riedl R. Klinische Wertigkeit des pr¨aoperativen Stagings von Blasen- und Prostatakarzinomen mit NMR und Computertomographie. Urologe A 1989; 28: 65. 270. Wood DP, et al. Identification of transitional cell carcinoma of the prostate in bladder cancer patients: a prospective study. J Urol 1989; 141: 83 – 5. 271. Bates HR, Thornton JL. Carcinoma of prostatic ducts. Am J Clin Pathol 1966; 45: 96. 272. Laplante M, Brice M. The upper limits of hopeful application of radical cystectomy for vesical carcinoma: does nodal metastasis always indicate incurability? J Urol 1973; 109: 261. 273. Thelmo WL, et al. Carcinoma in situ of the bladder with associated prostatic involvement. J Urol 1974; 111: 491. 274. Greene LF, et al. Primary transitional cell carcinoma of the prostate. J Urol 1973; 110: 235. 275. Rikken CHM, Van Helsdingen PJRO, Kazzaz BA. Are biopsies from the prostatic urethra useful in patients with superficial bladder carcinoma. Br J Urol 1987; 59: 145. 276. Sakamoto N, et al. An adequate sampling of the prostate to identify prostatic involvement by urothelial carcinoma in bladder cancer patients. J Urol 1993; 149: 318. 277. Laor E, Grabstald H, Whitmore WF. The influence of simultaneous resection of bladder tumors and prostate on the occurrence of prostatic urethral tumors. J Urol 1981; 126: 171. 278. Chibber PJ, et al. Transitional cell carcinoma involving the prostate. Br J Urol 1981; 53: 605 – 9. 279. Kirk D, Hinton CE, Shaldon C. Transitional cell carcinoma of the prostate. Br J Urol 1979; 51: 575 – 8. 280. Raz S, et al. Management of the urethra in patients undergoing radical cystectomy for bladder carcinoma. J Urol 1978; 120: 298. 281. Solsona E, et al. The prostate involvement as prognostic factor in patients with superficial bladder tumors. J Urol 1995; 154: 1710. 282. Reese JH, et al. Transitional cell carcinoma of the prostate in patients undergoing radical cystoprostatectomy. J Urol 1992; 147: 92. 283. Palou J, et al. In situ transitional cell carcinoma involvement of prostatic urethra: bacillus Calmette-Guerin therapy without previous transurethral resection of the prostate. Urology 1996; 47: 482. 284. Bretton PR, et al. Intravesical bacillus Calmette-Guerin therapy for in situ transitional cell carcinoma involving the prostatic urethra. J Urol 1989; 141: 853. 285. Hillyard RW, Ladaga L, Schellhammer PF. Superficial transitional cell carcinoma of the bladder associated with mucosal involvement of the prostatic urethra: results of treatment with intravesical bacillus Calmette-Guerin. J Urol 1988; 139: 290. 286. Orihuela E, Herr HW, Whitmore WF. Conservative treatment of superficial transitional cell carcinoma of prostatic urethra with intravesical BCG. Urology 1989; 34: 231. 287. Schellhammer PF, Ladaga LE, Moriarty RP. Intravesical bacillus Calmette-Guerin for treatment of superficial transitional cell carcinoma of the prostatic urethra in association with carcinoma of the bladder. J Urol 1995; 153: 53. 288. Lamm D. Re: Transitional cell carcinoma of the prostate, editorial. J Urol 1991; 146: 1630.
289. Droller MJ. Bacillus Calmette-Guerin in the management of bladder cancer. J Urol 1986; 135: 331. 290. Siami P, et al. BCG in management of transitional cell carcinoma invasive to prostate. Urology 1989; 34: 381. 291. Lockhart JL, et al. Prostatic recurrences in the management of superficial bladder tumors. J Urol 1983; 130: 256. 292. Droller MJ, Walsh PC. Intensive intravesical chemotherapy in the treatment of flat carcinoma in situ: is it safe? J Urol 1985; 134: 1115. 293. Shenasky JH, Gillenwater JY. Management of transitional cell carcinoma of the prostate. J Urol 1972; 108: 462. 294. Kopelson G, et al. Periurethral prostatic duct carcinoma: clinical features and treatment results. Cancer 1978; 42: 2894. 295. Frazier HA, et al. The value of pathologic factors in predicting cancerspecific survival among patients treated with radical cystectomy for transitional cell carcinoma of the bladder and prostate. Cancer 1993; 71: 3993. 296. Taylor HG, Blom J. Transitional cell carcinoma of the prostate: response to treatment with adriamycin and cis-platinum. Cancer 1983; 51: 1800. 297. Alexander SJ, Lee SS, Bekhrad A. Transitional cell carcinoma of the prostate: response to treatment with cisplatinum and cyclophosphamide. J Urol 1984; 131: 975. 298. Dexeus FH, et al. Complete responses in metastatic transitional cell carcinoma of the prostate with cisplatin regimens. J Urol 1987; 137: 122. 299. Takashi M, et al. Primary transitional cell carcinoma of prostate: case with lymph node metastasis eradicated by neoadjuvant methotrexate, vinblastine, doxorubicin, and cisplatin (M-VAC) therapy. Urology 1990; 36: 96. 300. von der Maase H, et al. Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic bladder cancer: results of a large, randomized, multinational, multicenter, phase III study. J Clin Oncol 2000; 18: 3068 – 77. 301. Bajorin DF, et al. Treatment of patients with transitional-cell carcinoma of the urothelial tract with ifosfamide, paclitaxel, and cisplatin: A phase II trial. J Clin Oncol 1998; 16: 2722 – 7. 302. Hussain M, et al. Combination paclitaxel, carboplatin, and gemcitabine is an active treatment for advanced urothelial cancer. J Clin Oncol 2001; 19: 2527 – 33. 303. Jacobi M, Panoff CE, Herzlich J. Leukemic infiltration of the prostate. J Urol 1937; 38: 494 – 9. 304. Johnson MA, Gundersen A. Infiltration of the prostate bland by chronic lymphatic leukemia. J Urol 1953; 69: 681 – 5. 305. Tighe JR. Leukaemic infiltration of the prostate. Br J Surg 1960; 47: 658 – 60. 306. Vinnicombe J. Leukaemic infiltration of the prostate. Br J Urol 1963; 35: 297 – 8. 307. Waddington RT. Leukaemic infiltration of the prostate in a patient with chronic lymphatic leukaemia – a case report. Br J Urol 1973; 45: 184 – 6. 308. Dajani YF, Burke M. Leukemic infiltration of the prostate: A case study and clinicopathological review. Cancer 1976; 38: 2442 – 6. 309. Butler MR, O’Flynn JD. Prostatic disease in the leukaemic patient – with particular reference to leukaemic infiltration of the prostate – a retrospective clinical study. Br J Urol 1973; 45: 179 – 83. 310. Thalhammer F, et al. Granulocytic sarcoma of the prostate as the first manifestation of a late relapse of acute myelogenous leukaemia. Ann Hematol 1994; 68: 97 – 9. 311. Bostwick DG, et al. Malignant lymphoma involving the prostate: Report of 62 cases. Cancer 1998; 83: 732 – 8. 312. Weimar G, et al. Urogenital involvement by malignant lymphomas. J Urol 1980; 125: 230 – 1. 313. Zein TA, et al. Secondary tumors of the prostate. J Urol 1985; 133: 615 – 6. 314. Freeman C, Berg JW, Cutler SJ. Occurrence and prognosis of extranodal lymphomas. Cancer 1972; 29: 252 – 60. 315. Rosenberg SA, et al. Lymphosarcoma: A review of 1269 cases. Medicine 1961; 40: 31 – 84. 316. Sarris A, et al. Primary lymphoma of the prostate: good outcome with doxorubicin-based combination chemotherapy. J Urol 1995; 153: 1852 – 4.
UNCOMMON CANCERS OF THE PROSTATE 317. Klotz LH, Herr HW. Hodgkin’s disease of the prostate: A detailed case report. J Urol 1986; 135: 1261 – 2. 318. Coupland S. Lymphoma of the prostate. Trans Pathol Soc Lond 1877; 28: 179. 319. Bostwick DG, Mann RB. Malignant lymphomas involving the prostate. Cancer 1985; 56: 2932 – 8. 320. King LS, Cox TR. Lymphosarcoma of the prostate. Am J Pathol 1951; 27: 801 – 23. 321. Fell P, O’Connor M, Smith JM. Primary lymphoma of prostate presenting as bladder outflow obstruction. Urology 1987; 29: 555 – 6. 322. Patel DR, et al. Primary prostatic involvement in non-Hodgkin lymphoma. Urology 1988; 32: 96 – 8. 323. Kerbl K, Pauer W. Primary non-Hodgkin lymphoma of prostate. Urology 1988; 32: 347 – 9. 324. Suzuki H, et al. Malignant lymphoma of the prostate. Report of a case. Urol Int 1991; 47: 172 – 5. 325. Sarlis NJ, et al. Primary non-Hodgkin lymphoma of the prostate gland. Int Urol Nephrol 1993; 25: 163 – 8.
65
326. Morozumi M, et al. Primary non-Hodgkin lymphoma of the prostate: a case report. Nippon Hinyokika Gakkai Zasshi 1993; 84: 2023 – 63. 327. Mounedji-Boudiaf L, et al. Primary, highly malignant B-cell lymphoma of the prostate. Apropos of a case and review of the literature. Bull Cancer 1994; 81: 334 – 7. 328. Braslis KG, et al. Primary prostatic lymphoma: a rare prostatic malignancy. Aust N Z J Surg 1994; 64: 58 – 9. 329. Parks RW, et al. Primary non-Hodgkin’s lymphoma of the prostate mimicking acute prostatitis. Br J Urol 1995; 76: 409. 330. Ghose A, et al. Lymphoma of the prostate treated with radiotherapy. Clin Oncol (R Coll Radiol) 1995; 7: 134. 331. Bell CR, et al. Primary non-Hodgkin’s lymphoma of the prostate gland: case report and review of the literature. Clin Oncol (R Coll Radiol) 1995; 7: 409 – 10. 332. Rainwater LM, Barrett DM. Primary lymphoma of the prostate: Transrectal ultrasonic appearance. Urology 1990; 36: 522 – 5. 333. Horning SJ, Rosenberg SA. The natural history of initially untreated low-grade non-Hodgkin’s lymphomas. N Engl J Med 1984; 311: 1471 – 5.
Section 1 : Genitourinary Cancer
5
Rare Tumors of the Testis and Paratesticular Tissues Vedang Murthy, Cyril Fisher and Alan Horwich
INTRODUCTION The testes originate embryologically from the retroperitoneum. They have both endocrine and reproductive functions, and tumors of the testes and their adnexae (see Figure 1) reflect this diversity of both origin and function. The majority of testicular cancers arise from germinal epithelium.1 Rarely tumors of lymphoid, interstitial, or ductal origin arise in the testes and this chapter describes these along with tumors of paratesticular, mesothelial, or connective tissues. The main sections of this chapter cover the following areas: • • • • •
Non-Hodgkin’s lymphoma (NHL) of the testis Gonadal (sex cord) stromal tumors (GSTs) Malignant mesothelioma of the tunica vaginalis (MMTV) Adenocarcinoma of the rete testis Paratesticular rhabdomyosarcoma (PTR)
NON-HODGKIN’S LYMPHOMA OF THE TESTIS History The first description of malignant lymphoma of the testis in 1877 is accredited to the French worker Malassez.2 In a case report published by Hutchinson3 many of the notable features of this disease were illustrated: patient was elderly, had bilateral tumors, and the disease disseminated widely to involve bone and subcutaneous tissues (a feature also of Malassez’s original case). Early reports documented the age distribution of patients, the pattern of dissemination, and the tendency to bilateral involvement of these rare tumors. Ficari4 reviewed the literature from 1877 to 1947; 10 of 18 cases were in patients over 50 years of age, the skin was involved in 10 cases and the disease was bilateral in 8. Gowing5 reported on 128 cases. Seventy-eight percent of patients were over 50 years of age, 20% had bilateral testicular involvement, and 62% had died of disseminated disease within 2 years of onset. However, 15 of 124 patients treated by orchidectomy survived disease-free for 5 or more
years, supporting the conclusion that testicular lymphoma occurs as a primary manifestation of NHL as well as in the context of disseminated disease. The prognosis has been considered poor, in keeping with the aggressive histology. However, the importance of both the stage, emphasized by Kiely et al.,6 and the exact histological type7 is now recognized, as is the high relapse rate even in patients with early-stage disease, leading to the investigation of systemic therapy in patients with stage I presentation. An excellent retrospective outcomes study on behalf of the International Extranodal Lymphoma Study Group (IELSG) has been performed by Zucca et al., based on 373 patients from 23 institutions worldwide.8 This forms the largest body of systematically analyzed evidence to date and provides a tool for defining the clinical features and outcome of this subset of lymphoma, which differs from its nodal counterpart in several respects.
Epidemiology and Etiology Testicular tumors are rare, affecting from <1 to 4.5 patients per 1 000 000 men per year, and there are marked racial and geographic variations.9 The reported incidence is increasing10 and the vast majority of these tumors arise from the germinal epithelium. NHL accounts for approximately 5% of all testicular neoplasms (see Table 1). The relative incidence increases with age (see Figure 2). In patients over 50 years, lymphomas account for 25–50% of all testicular neoplasms.11 Nearly 80% of the cases reported by the British Testicular Tumour Panel (BTTP) were over 50 years, contrasting with just three children reported in the same series.5 The testis is an uncommon site of presentation for NHL.17,18 The Princess Margaret Hospital, Toronto, serving a population of four million, saw 1934 new patients with NHL between 1967 and 1978 of whom 16 (0.83%) had a primary testicular presentation.11 Wahal et al.19 reported on NHL in north India and of the 1283 cases 264 (21%) were extranodal, of which 20 (1.5%) presented with a testicular lesion.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES
67
10
Spermatic cord (paratesticular rhabdomyosarcoma)
Lobules (germ cell tumors, gonadal stromal tumors and lymphomas)
Rete testis (adenocarcinoma)
Tunica albuginea Tunica vaginalis (mesothelioma)
Age specific incidence per 100 000
Cremasteric fascia
1.0
0.1
Cremaster muscle Figure 1 Diagram of testis and spermatic cord indicating sites of tumor origin. PTR may arise from connective tissue of the cord or of adjacent structures. Table 1 Incidence and age distribution of testicular lymphomas.
Overall number of testicular tumors
Number with lymphoma
Eckert12 Gowing5
665 2106
35 (5%) 140 (6.5%)
Mehrotra13 Ciatto14 Read15
292 280 1307
22 (7.5%) 12 (4%) 51 (4%)
200
17 (8%)
Series
Hayes16
Age distribution – 78% over 50 years – – 50% over 50 years –
The etiology of testicular lymphoma is unknown. Association with maldescent, trauma, and familial occurrence, although described, are probably not of etiological significance.5
Pathology Testicular lymphoma may present with a very large testicular mass. Kiely et al.6 recorded one case measuring 16 cm × 9 cm and weighing 750 g. In the BTTP series,5 the largest tumor measured 13 cm × 10 cm × 8 cm and weighed 380 g. The cut surface may appear homogeneous or have yellowish areas of necrosis present and the color has been variously described as cream, pink, gray, or buff. No capsule is present and the lesions have ill-defined edges. Local extension is common. Gowing5 gave a full histological description of these tumors and the morphological features that help distinguish them from seminoma. In this series, all the tumors were composed of either poorly differentiated cells of the lymphocytic series or of lymphomas of larger undifferentiated cell type. The classification most widely applied to testicular lymphoma in the past has been that of Rappaport,20 and using this virtually all reported cases are of diffuse histiocytic or diffuse poorly differentiated lymphocytic type. Turner et al.,7 reported on 35 cases of malignant lymphoma of the testis
0.02 0.01
0
10
20
30
40 50 Age
60
70
80
90
Figure 2 Incidence of germ cell tumors and of lymphomas of the testis by age of presentation. (Data from Thames Cancer Registry). Solid line represents germ cell tumors. Dashed line represents Lymphomas.
classified by both the Rappaport criteria and the International Working Formulation of NHL.21 All the 35 tumors showed a diffuse pattern of growth and 29 were diffuse histiocytic lymphomas. In the Working Formulation study 22 of these were of intermediate grade (large cell cleaved, large cell noncleaved) and seven were high grade (immunoblastic lymphoma). In subsequent studies, the majority of testicular lymphomas have been found to be of intermediate grade.22,23 Immunohistological studies have suggested that the majority are tumors of B lymphocytes.24 The majority of the tumors are classified in the recent World Health Organization (WHO) classification as mature B-cell neoplasms (diffuse large B-cell lymphoma).25 It has been suggested that rapid dissemination associated with this tumor may be because of the lack of expression of adhesion molecules.26 The tumors usually have BCL-2 protein, but not the t(14;18) translocation typical of follicular lymphomas.27
Prognosis and Patterns of Relapse The prognosis of testicular NHL is poor (see Table 2) with early reports showing disappointing results. In the BTTP series, 62% had died of disseminated lymphoma within 2 years of presentation and only 15 of the 124 patients treated by orchidectomy survived 5 or more years postoperatively. Jackson and Montessori22 confirmed this low 5-year survival rate by tabulating reported series up to 1977 and found only 28 out of 194 (14.5%) patients surviving diseasefree for more than 5 years. However more recent reports of aggressive multimodality treatment using a combination
68
GENITOURINARY CANCER
better outcome, with a 5-year OS of 81% and progressionfree survival (PFS) of 68%. This small subset of patients has a prognosis approaching that of their nodal lymphoma counterparts.8 The Royal Marsden Hospital series34 demonstrated a significantly longer lymphoma-free survival in 18 patients with stage I –II disease compared with six patients with stage III/IV disease, but there was no difference in the probability of lymphoma-free survival between the 10 stage I and 8 stage II patients. Visco et al. from MD Anderson Hospital reported 43 patients treated with multimodality therapy, with a 10 year PFS for stages I/II and III/IV being 36% versus 0% respectively.28 Tepperman et al.11 analyzed 15 patients and found that para-aortic nodal involvement was the strongest adverse prognostic factor. Median survival without this was 57 months, but if the nodes were involved it was 6 months (p = 0.002). No patient with para-aortic node involvement survived beyond 19 months. Also described were two exceedingly rare cases, one of a primary nodular lymphoma of the testis and one of a mixed seminoma/lymphoma. These tumors often disseminate widely. In the BTTP series,5 13 patients (10%) had clinical evidence of disseminated disease at the time of orchidectomy. Among series reported since 1980, the incidence of dissemination at the time of orchidectomy varies from 35% to 70%11,15,23,24,34 (see Table 3); this apparent increase probably reflects more sophisticated and sensitive staging procedures. The sites of possible metastatic involvement are numerous and reflect both lymphatic and hematogenous spread. In the more recent series, para-aortic node involvement as the sole extragonadal site of disease occurred in just over one-third of cases. In the IELSG study, the sites of extranodal involvement at presentation were bone marrow (5%), central nervous system (CNS) (3%), adrenal glands and skin (2.5% each), bone and kidney (2% each) and soft tissue gastrointestinal tract and liver (1% each).8 Sites of involvement at relapse are varied (see Table 4) but the high incidence of relapse in the upper airways (15%), CNS (10%), and bone involvement (10%) is noteworthy. There appears to be a particular risk of relapse in the CNS.35 – 39 In a Multicentre Rare Cancer Network study of 36 patients, CNS was found to be the principal site of relapse (8/14 relapses).40 This has led to intrathecal chemotherapy prophylaxis as part of routine management of these patients.
Table 2 Prognosis of testicular lymphoma.
Author
Stagea
Number
% Disease-free survival at 5 years
Zucca8 – – – Gowing5 Sussman30 Jackson22 Read15b – – Crellin31 Zietman32
I II III – IV I – IV I – IV I – IV I – IV I/IIA IIB/IV I – IV I – IV I
214 80 79 373 128 37 194 24 27 51 34 26
54 48 30 48 12 20 14.5 40 0 20 33 61
a b
Staging according to the Ann Arbor classification.33 I/IIA, no palpable metastases; IIB/IV, clinically overt metastases.
of systemic and intrathecal chemotherapy and scrotal and lymphatic radiotherapy, have demonstrated 5 year overall survivals (OS) of over 80% depending upon the prognostic factors present.28,29 The IELSG multicenter review8 revealed a median survival of 4.8 years and a 5-year and 10-year OS of 48% and 27%, respectively for the whole group. However, a majority of these tumors were stage I. In keeping with the aggressive nature of the disease most patients who succumb to lymphoma do so within the first 2 to 3 years, but there are a proportion of deaths that occur later. Sussman et al.30 noted that of 37 patients, 24 (65%) died of disseminated disease; 71% of these died within the first year of postorchidectomy, 17% within the second, 4% within the third, none in the fourth, and 8% within the fifth year of follow-up. In the IELSG report, lymphoma-related deaths were seen up to 14 years from diagnosis.8 In the IELSG report, the factors significantly associated with a favorable prognosis on univariate analysis were good performance status, limited stage, absence of a bulky mass, normal serum lactate dehydrogenase (LDH) and β-2 microglobulin, and absence of additional extranodal involvement, absence of B symptoms, favorable International Prognostic Index, anthracycline-based chemotherapy, and prophylactic scrotal irradiation. However, on multivariate analysis only the latter four remained significant. Among stage I patients, those with normal LDH levels, age less than 60 years, and a performance status 1 had a significantly Table 3 Testicular lymphoma: stage distribution.
Number of cases Author Zucca
8
IE
IIE
214
80
Duncan34 Read15
10
Baldetorp23 Crellin31
8 13
IIIE
IV 79
8
2 6
4 17
8 10
3 3
5 8
28
Stage IV major sites involved (n) Bone marrow (18), CNS (11), Adrenal glands (9), Skin (9) Bone (8), Kidney (7) Skin (3), Lung (1) Bone (4), Skin (4), Lung (3), CNS (3), Upper airways (4), Bone marrow (2), Liver (1) Bone (2), Liver (1), Skin (1) Bone (2), Marrow (3), Pelvis (3), Kidney (3)
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES Table 4 Testicular lymphoma: sites of relapse.
Author
Number of cases
Number relapsing
Zucca8
Stage I (214)
102
–
Stage II (80)
44
–
Stage III – IV (79)
49
Total (373) 24
195 13
Read15
51
32
Baldetorp23
24
13
Crellin31
34
23
– Duncan34
Major sites of relapse Contralateral testis plus other nodal and extranodal sites (stage I: 19, II: 5, III/IV: 7, total : 31) All CNS relapse (stage I: 27, II: 12, III/IV: 17, total: 56) Isolated CNS relapse (stage I: 17, II: 6, III/IV: 11, total: 34) Upper airways (5), bone (5), nodal (4), CNS (2) Liver (11), skin (9), CNS (5), lung (5), bone (2) Nodal (6), upper airways (2), other testes (3) CNS (6), testis (2), lung (3), heart (2), marrow (2), thyroid (1), and nodal sites
69
The following investigations are recommended: full blood count and differential count, erythrocyte sedimentation rate, bone marrow aspirate and trephine biopsy, renal and liver function tests, α-fetoprotein and β-human chorionic gonadotrophin (BHCG) estimation (to exclude a mis-diagnosed germ cell tumor), chest x-ray, computed tomography (CT) scan of the chest and abdomen, and cerebrospinal fluid cytology. Staging laparotomy has been infrequently reported and is not recommended in view of the age and stage distribution of patients and the emphasis on systemic treatment. In some centers, gallium or positronemission tomographic scanning are included as routine staging tools to define the distribution of disease more clearly.
Management and Outcome Important considerations for appropriate management include the patient’s age and general condition, the histological subtype, and the stage of the disease.
Clinical Features
Surgery
Enlargement of the testicle is the usual presenting feature. The duration of symptoms in Talerman’s series41 varied from a few weeks to 3 years but was usually less than 6 months. Pain is present in 8–25% of cases and in the 402 cases reported up to 1985 there was a right-sided preponderance of 1.3 : 1. The clinical features that may help differentiate lymphoma from the more common germ cell tumors and seminoma are the patient’s greater age, the tendency to bilateral involvement, lack of association with maldescent, the different pattern of metastatic spread, and a lack of gynecomastia. Many other authors have commented on this striking difference in age distribution between lymphoma and the more common testicular tumors. In the series of Eckert and Smith,12 the mean age for lymphoma was 59.8 years compared with 33 years for teratoma and 42.3 years for seminoma. In the IELSG report, the median age at presentation was 66 years (range 19–91 years). Overall, 79% had stage I/II disease, 90% had no B symptoms, 5% had bulky disease (defined as >10 cm diameter), and 47% had a raised LDH level.8 Bilateral synchronous primaries occur in about 2–5% of cases.5,42,43 Duncan et al.34 commented that patients presenting with bilateral testicular lymphoma usually have advanced disease, with only 1 of 11 patients reported in the medical literature surviving more than 2 years following locoregional therapy. In childhood, testicular lymphoma is rarely localized,44,45 and Burkitt’s lymphoma is a more common subtype,46 although it has also been reported in the elderly.47
A radical inguinal orchidectomy is the recommended procedure. Historically, this was the only method of treatment and can result in cure of a minority of patients.5 In the IELSG study,8 41 patients had orchidectomy as the sole treatment. Their PFS was significantly shorter than those of patients receiving additional chemotherapy and/or radiotherapy (median PFS, 1.0 vs 5.4 years). The OS was also shorter, but the difference did not reach statistical significance [median OS, 2.1 vs 6.4 years (p = 0.07)]. [Editorial Note: Although it seems likely that this reflected the relatively small number of cases.]
Investigation and Staging Patients should be staged according to the Ann Arbor classification.48 Important areas of clinical assessment include the contralateral testis, regional and other lymph node areas, the skin, the neurological system, and the upper airways.
Radiotherapy Abdominal and pelvic lymph node irradiation, e.g. by an inverted-Y field, has been employed either as an adjuvant following orchidectomy or to treat overt disease, and either alone or in combination with chemotherapy. The radiotherapy prescription as shown in Table 5 varies in different institutions, with a midplane dose recommendation of between 30 and 40 Gy in 20 fractions over 4 weeks using megavoltage equipment. With orchidectomy and adjuvant radiotherapy, approximately 50% of stage I patients and 20% of stage II patients can be expected to be long-term survivors. Data on abdominal control of disease are scanty, but following orchidectomy and radiotherapy, abdominal recurrence was seen as a solitary event in only two cases and in conjunction with more widespread disease in three cases in the collected series of Duncan et al.,34 Tepperman et al.,11 and Buskirk et al.49 In view of the high systemic relapse rate, consideration should be given to the use of adjuvant chemotherapy in earlystage disease, an approach showing promise in early-stage lymphoma at other sites46 and which appears to improve the prognosis for testicular presentations.6,50 For patients with bulky localized NHL, there may be an advantage to the addition of involved field radiation therapy (RT) to
70
GENITOURINARY CANCER
Table 5 Testicular lymphoma: radiotherapy results.
Stage
Number of patients
Radiotherapy prescription
IE
9
IIE I/IIA
7 24
– Tepperman11
IIB IE
4 4
– Buskirk49
IIE IE
6 8
IIE
3
30 – 40 Gy in 20 – 24# in 34 days Daily fractionation 30 Gy in 20# over 4 weeks – 25 Gy in 20# over 4 weeks – 25 – 40 Gy in 20# over 4 weeks –
Author Duncan34 – Read15a
–
Disease-free survival (%) 45
–
30 40
At 5 years –
0 75
At 5 years –
17 50
At 2 years –
33
At 2 years
a
Manchester staging system. Gy, Gray; # fraction.
chemotherapy.51 This would be a consideration in those presenting with abdominal node metastases or in those who have an involved unresected contralateral testis. However, Zietman et al.32 did not find adjuvant RT to improve results in stage I disease. Adjuvant irradiation of the scrotal sac along with the contralateral remaining testicle has been shown to improve outcome. In the IELSG report, a continuous risk of recurrence in the contralateral testis (15% at 3 years, 42% at 15 years) was present in patients not receiving radiotherapy to the contralateral testis (p = 0.003). Prophylactic radiotherapy to the contralateral testis was also associated with better PFS (5-year PFS, 36 vs 70%; p < 0.001) and OS (5-year OS, 38 vs 66%; p < 0.001) rates. Among patients receiving radiotherapy to the primary testicular site of involvement, the OS was longer for those receiving an irradiation dose of at least 30 Gy (p = 0.02).8
Chemotherapy The poor results, especially in patients with extensive disease, may be improved by the use of modern combination chemotherapy regimens. In the IELSG study,8 combination chemotherapy with anthracycline-containing regimens significantly improved the outcome in all patients (5-year PFS, 35 vs 55%; p < 0.001; and 5-year OS, 39 vs 52%; p = 0.02). Even when the analysis was limited to stage I/II patients, the benefit was statistically significant for PFS and OS. Patients receiving six or more cycles of chemotherapy had a better long-term outcome than those treated for a shorter period (10-year OS, 44 vs 19%; p = 0.03). Complete response rates of upto 75% have been reported especially by the use of alternating combinations such as M-BACOP,52,53 PROMACE-MOPP,51 or MACOP-B.54,55 However, these responses are often short lived and aggressive regimens are toxic in elderly patients. Cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) is a less-intensive alternative.56 Prospective trials in advanced NHL have not demonstrated an advantage for more intensive schedules compared with CHOP.57 The recent report of improved survival with the addition of rituximab to the CHOP chemotherapy for elderly patients with diffuse large B-cell lymphoma
supports the view that this combination may also be optimal in patients with primary testicular large B-cell NHL.58
Recommendations In view of the high systemic relapse rate even for patients with early-stage disease we would recommend systemic chemotherapy in all patients. The management of patients with advanced disease should be as for advanced lymphoma at other sites,59 (see Chapter 49, Rare Lymphomas). The choice of chemotherapy would depend on the age and health of the patient and preference of the treating institution. At least six cycles of CHOP with rituximab, should be considered appropriate for advanced disease although fewer courses may be effective in stage I/II disease. Adult patients with stage II disease should receive adjuvant involved field nodal RT to a dose of 30–35 Gy. Irradiation of the contralateral scrotal sac is also recommended to a dose of 30 Gy. All patients should be considered for CNS staging and prophylaxis with intrathecal chemotherapy.
GONADAL (SEX CORD) STROMAL TUMORS Introduction 60
Mostofi et al. referred to Sertoli and Leydig cell tumors (LCTs) as “tumors of the specialized gonadal stroma”, and suggested a common origin for these cell types. The term sex cord stromal tumors, which is used for analogous ovarian tumors, may be more appropriate.61 They are uncommon, accounting for approximately 4% of all testicular neoplasms, have a broad age distribution at presentation, and they may be hormonally active. Expression of the S-100 protein has been reported.62 Testicular stromal cells include Leydig and Sertoli cells. Leydig cells are named after Franz von Leydig (1821–1908), the German anatomist who first described them. They arise embryologically from the posterior urogenital ridge, secreting primarily testosterone, but also lesser amounts of estrogens. They are capable of producing testosterone from cholesterol and can secrete just under 10 mg of testosterone per day.63 Sertoli cells are named after the Milan physiologist Enrico Sertoli (1842–1910). They form the supporting cells of
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES
71
the seminiferous tubules lying between the germinal cells, often with clumps of spermatids buried in their cytoplasm. Sertoli cells have certain features characteristic of hormoneproducing cells, and although apparently unable to generate steroids they may be able to promote their interconversion. The majority of GSTs are LCTs, with smaller numbers of Sertoli cell tumors (SCTs) and occasional rarities such as granulosa cell tumors. Mixed GSTs have been described, and one example revealed the germ cell cytogenetic marker isochromosome i(12p) in an inguinal node metastasis.64 Most GSTs are benign but about 10% metastasize. It is difficult to predict the malignant phenotype from the presenting characteristics. Surgery remains the mainstay of treatment.
Leydig Cell Tumors Epidemiology and Etiology
Figure 3 LCT composed of relatively uniform cells with rounded nuclei and abundant cytoplasm (H&E ×310).
In the BTTP series of 2739 testicular tumors, 43 (1.6%) were LCTs.65 Mostofi66 recorded a 3% and Ward et al.67 a 2% incidence. The age of the patients in the BTTP series ranged from 21 to 81 years. Most published cases of LCT have occurred in adult life; childhood cases are uncommon.68 Leydig cell hyperplasia may occur with testicular atrophy, and although LCTs are reported with maldescent and atrophy, this association is of uncertain significance.65 Constant luteinizing hormone stimulation of the testis in experimental animals can induce both Leydig cell hyperplasia and LCT. By 1985, Kim et al.69 reported that only 37 cases of malignant LCT had been described; the average age was nearly 63 years. Pathology and Biology
These tumors, which are rarely bilateral, arise within the substance of the testis and are usually well demarcated with a striking yellowish brown color. Necrosis and hemorrhage may be seen. They can vary widely in size, the range in the BTTP series being 0.7 to 10 cm,65 and sometimes involve the rete testis and spermatic cord. The tumor cells may be polygonal or fusiform in shape with prominent nucleoli and abundant eosinophilic or granular cytoplasm. Characteristic crystalloids of Reinke are seen in about one-third of cases, and these have been used to aid in cytologic diagnosis.70 Mitotic figures can be frequent and do not necessarily imply malignancy.65 The absence of seminiferous tubules within the tumor helps differentiate LCT from Leydig cell hyperplasia (see Figures 3 and 4). Features of the primary lesion that should arouse suspicion of malignancy include large size (>5 cm), lack of encapsulation, presence of satellite nodules, areas of necrosis, and on microscopy, a high mitotic rate (>3 mitoses per 10 high-power fields) and blood vessel or lymphatic invasion.65 In the review by Kim et al.,69 84% of the malignant LCTs were 5 cm or greater in diameter, 74% had an infiltrative margin, and 72% had lymphatic or vascular invasion. However, the occurrence of metastasis is the only definite proof of malignancy.71 Cheville et al.72 reported thirty cases of LCT of the
Figure 4 Diffuse B-cell lymphoma involving testis. The infiltrate is interstitial and widely separates seminiferous tubules that contain only Sertoli cells (H&E ×62).
testis (23 tumors were localized and 7 had metastasized). Patients with LCTs that metastasized were diagnosed at a mean age of 62 years (range, 39–70 years) compared with 48 years (range, 9–79 years) in patients with nonmetastasizing tumors (p = 0.25). LCTs that metastasized were significantly larger than nonmetastasizing tumors (mean, 4.7 vs 2.6 cm, respectively; p = 0.008). In this study, the presence of cytologic atypia, necrosis, angiolymphatic invasion, increased mitotic activity, atypical mitotic figures, infiltrative margins, extension beyond the testicular parenchyma, DNA aneuploidy, and increased MIB-1 activity were significantly associated with metastatic behavior in LCTs. Inhibin was reported to be the most sensitive marker, with positive staining in 91% of SCTs and 100% of LCTs and very useful to differentiate testicular sex cord stromal tumors from germ cell tumors.73,74 Metastases are present at diagnosis in about 25% of cases and develop in the remainder usually within 3 years. Recurrence after 5 years can occur but is uncommon.75 The
72
GENITOURINARY CANCER
most common sites of metastases include the regional lymph nodes (68%), liver (45%), lung (45%), and bone (27%). Clinical Features
Although stromal tumors may present in childhood, these usually do not metastasize.75 – 77 Malignant LCT tumors usually present in adult life with a testicular mass in the majority.72 In the review by Grem et al.,78 the age range at presentation was 20–82 years with a median age of 58 years. Painless testicular enlargement was present in 81%. All patients in this series by definition had metastases, 72% with regional node involvement, 43% with pulmonary metastases, 38% with hepatic metastases, and 28% with lung involvement. Gynecomastia was present in 19%. Metastases were present in 22% at presentation and developed within 1 year in a further 19%. Investigations
A full endocrinological assessment of the pituitary/gonadal axis is advised with particular attention to androgen, estrogen, and progesterone production.66,76,79 The value of these data is that if elevated titers are noted they can act as a tumor marker. A case report has indicated that spermatic venous blood from the tumor-bearing testis had high concentrations of testosterone and androstenedione.80 Elevations of urinary ketosteroids were seen in 14 of 22 patients (64%), of serum and/or urinary androgens in 12 of 22 patients (54%) and of serum and/or urinary estrogens in 11 of 22 patients (50%) with metastatic LCT.80 Increased 17hydroxyprogesterone (17-OHP) level in a child may suggest congenital adrenal hyperplasia; however, in LCTs plasma cortisol levels are normal and the 17-OHP does not suppress with dexamethasone.81 CT scan of the chest and abdomen should also be performed.
Management and Prognosis Surgery
Inguinal orchidectomy is the treatment of choice for LCT and will cure the majority of patients. Malignancy is often defined in terms of the presence of metastatic disease and surgery has been used to resect established regional nodal disease in the retroperitoneum82,83 and solitary pulmonary metastases with long-term benefit.84 Laparoscopic retroperitoneal lymph node dissection (RPLND) may be a less-invasive approach.85 Most patients have multiple sites of metastases and surgery can offer useful palliation in some circumstances.86 Radiotherapy
This was prescribed for 11 of 32 patients with malignant LCT reviewed by Grem et al.78 No objective responses were seen but two patients (one with bone metastases and one with a retroperitoneal mass) noted reduction in pain. Total radiation dose varied but no response was recorded in patients receiving 48 Gy87 or 50 Gy88 as well as in those receiving lower doses. Chemotherapy
This has been presented only in the setting of advanced metastatic disease. Two of seven patients treated with the
agent o,p -DDD responded, one with a reduction in liver size and falling urinary 17-ketosteroid levels and another with resolution of pulmonary nodules.80,89 Standard chemotherapy regimens have had no major benefit90 and more experimental therapy with lonidamine, which can impair spermatogenesis probably via changes induced in Sertoli cells, although producing symptomatic improvement in one of two patients, produced no objective responses.80 Prognosis of those with metastasis is variable, with a median survival of 2 years (range 2 months to 17 years). Recommendations
Inguinal orchidectomy is recommended both to establish the diagnosis and to remove the primary lesion. In those patients whose tumors appear benign and whose hormone profile returns to normal postorchidectomy, no further treatment is recommended beyond surveillance, reserving lymphadenectomy for those who relapse either solely in regional nodes or with limited resectable pulmonary disease. Those patients with histologic features suggestive of malignancy should be fully investigated and considered for template retroperitoneal node dissection. Radiotherapy and chemotherapy together appear to be ineffective but may be considered for palliation in patients with widespread or unresectable disease.
Sertoli Cell Tumor Epidemiology and Etiology
In the BTTP series, there were 32 SCTs (1.2%).65 At least seven exhibited malignancy. This figure is in keeping with other large reported series.69 The age of patients ranged from 2 months to 80 years with seven patients presenting in the first decade of life. Godec91 reported the 11th case of malignant SCT (excluding the BTTP series patients) and noted only two cases in children. SCT may be found either as pure tumors or in combination with germ cell tumors or other gonadal stromal tumors; the etiology remains obscure. The variable clinical course of these tumors has been reviewed by Giglio et al.92 SCT may occur in association with Peutz –Jeghers syndrome when feminization has been reported to be because of increased transcription of the aromatase P450 gene.93,94 Pathology and Biology
A full description of these tumors is provided by Symington and Cameron.65 SCTs arise within the testis and in the BTTP series they showed considerable variation in size (1–30 cm).65 They were well demarcated, and creamy white or tan in color, with occasional foci of hemorrhage or necrosis. Histologically, SCTs usually show solid or glandlike tubule formation, forming nests and cords separated by fibrous tissue (see Figure 5) with considerable variation in pattern both within individual tumors and between different tumors. The tumor cells may appear clear because of intracytoplasmic lipid, and Call–Exner-like bodies or focal calcifications are sometimes noted within larger aggregations of cells. Immunohistochemical evidence of vimentin and cytokeratin, and lack of AFP, HCG, PLAP, EMA, and CEA corresponds to the phenotype of the normal Sertoli cell.95,96
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES
73
and testosterone levels have been reported. Young et al.97 described 60 SCTs of the testis, with an age range from 15 to 80 years (mean, 45 years). In 14 cases a testicular mass had been enlarging slowly, the longest period being 14 years (mean 3.7 years). Only five patients had testicular pain. Four patients had metastatic disease at the time of presentation. All the tumors were unilateral and ranged from 0.3 to 15 cm (mean 3.6 cm) in size. Investigations
Figure 5 SCT with a pattern of variably sized tubules in a loose fibrous stroma (H&E ×155).
In the BTTP series, seven cases were malignant. The sites of metastatic involvement included regional nodes, liver, lung, bone, and brain. In four of seven cases the tumor invaded the rete testis, epididymis, or spermatic cord, and in all cases lymphatic and/or vascular invasion was seen. In the Boston series of 60 SCTs,97 the pathologic features that best correlated with a clinically malignant course were as follows: a tumor diameter of 5.0 cm or greater, necrosis, moderate to severe nuclear atypia, vascular invasion, and a mitotic rate of more than 5 mitoses per 10 high-power fields. Only one of nine benign tumors for which follow-up data of 5 years or more were available had more than one of these features, whereas five of seven malignant tumors had at least three. As with Leydig tumors, staining for inhibin is common, but lack of staining may be a marker of malignant potential.98 The distinction of SCTs from seminoma may be difficult. Henley et al. highlighted 13 malignant SCTs of the testis with light microscopic features that mimicked seminoma with a nested growth pattern, prominence of clear cells, lymphoid infiltrate cytoplasmic glycogen and prominent nucleoli.99 However, staining for inhibin-α and epithelial membrane antigen was positive in those tested. The time from presentation to detection of metastases in a review96 was generally short, with 8 of 11 malignant cases developing metastases within 1 year, although long diseasefree intervals are occasionally noted. Clinical Features
SCTs usually present with testicular swelling: 29 of 32 cases in the BTTP series did so in this manner.65 They may also be discovered on routine examinations or as an incidental finding either at autopsy or at surgery for maldescent.66 Gabrilove et al.100 reported on 72 cases of SCT of which 60 were benign. Seventeen cases presented at less than 1 year of age, and 28 between 20 and 45 years. Malignancy was noted in only one child less than 10 years and was predominantly seen in patients over 25 years. Gynecomastia was recorded in 17 of the 72 cases and was associated with malignancy in seven cases. Hormone studies have been infrequent but elevation of serum and urinary estrogens
A full endocrinological assessment of the pituitary/gonadal axis is advised with particular attention being paid to androgen, estrogen, and progesterone production.66 CT scan of the chest, abdomen, and pelvis should be performed particularly when the primary tumor exhibits features of malignancy. BHCG, AFP, and placental alkaline phosphatase should also be measured to exclude germ cell tumor.
Management and Prognosis Surgery
Inguinal orchidectomy is the treatment of choice for SCT and will cure the majority of patients. RPLND has been performed both as a staging/adjuvant101 procedure and to resect established metastatic disease. The value of the former procedure, although not tested in a trial setting, is apparent in case reports in which long-term survivors are reported following resection of both microscopic and macroscopic nodal disease.102,103 In Godec’s series reviewing 11 cases with malignant SCT, four had RPLND for nodal disease and three remained disease-free at 5 years, 7 months, and 6 months respectively.96 Mosharafa et al.83 reviewed 17 patients with malignant sex cord stromal tumors who underwent RPLND. All patients with stage I disease were alive and disease-free at last follow-up (mean 4.5 years). Six of eight patients with stage II –III tumors died of disease within 9 months to 6 years of surgery (median 1.2 years). Radiotherapy
Radiotherapy has also been prescribed as adjuvant treatment to the inguinal nodes and to treat recurrent disease. In a review of metastasizing SCTs, Madson and Hultberg reported that irradiation of para-aortic lymph nodes may provide effective local treatment. None of the 13 patients they reviewed developed para-aortic lymph node recurrences.104 In general, high radiation doses (40 Gy) have been recommended. Chemotherapy
Chemotherapy has been infrequently prescribed for this rare tumor, which does not appear to be chemosensitive, except possibly in childhood.102 Recommendations
Inguinal orchidectomy is recommended both to establish the diagnosis and to remove the primary lesion. In patients
74
GENITOURINARY CANCER
whose tumors appear benign and whose hormone profile returns to normal after orchidectomy, no further treatment is recommended. Patients with histologic features suspicious of malignancy should be fully investigated and if no evidence of spread of disease is found it is reasonable to consider either retroperitoneal node dissection or close surveillance. Patients who have disease limited to the regional nodes at presentation should initially have a retroperitoneal lymphadenectomy performed. Addition of radiotherapy is of uncertain benefit and radiotherapy should be reserved for palliation of patients with widespread or unresectable disease. The prognosis is variable. In the BTTP series65 six of seven patients with malignant SCT died 18 months after diagnosis, and one after 18 years (histologically confirmed). In the series of Godec,91 4 of 11 patients remain alive, deaths occurring within 2 years in five of seven patients who died. Large Cell Calcifying Sertoli Cell Tumor
A rare subtype of SCT is the large cell calcifying SCT.105 – 109 Only 48 cases have been reported in the literature as reviewed recently by Giglio et al.92 It is generally seen in childhood and adolescence and is sometimes associated with other abnormalities including Leydig cell and pituitary tumors110 and may form part of the autosomal-dominantly inherited syndrome, Carney’s complex.111,112 It presents as a slowly enlarging testicular mass, but may be bilateral or multifocal. Grossly, the tumor is well circumscribed, and microscopically it has fibrous septa with calcification, which separate sheets, cords, and solid nests of cells with abundant eosinophilic cytoplasm. Ultrastructurally,113,114 they resemble Sertoli cells, and characteristic Charcot–B¨ottcher crystalloids have been demonstrated in one case.110 Testosterone and estradiol have been demonstrated in the neoplastic cells by immunohistochemistry.111 Kratzer et al. reported six malignant and six benign large cell calcifying SCTs of the testis and reviewed the literature. Malignant tumors were unilateral and solitary and occurred at a mean age of 39 years (range 28–51 years), whereas the benign neoplasms were bilateral and multifocal in 28% of cases and occurred at a mean age of 17 years (range 2–38 years). Only one malignant tumor occurred in a patient with evidence of a genetic syndrome (Carney syndrome), whereas 36% of benign tumors had various genetic syndromes or endocrine abnormalities. Most of the tumors in the latter cases were bilateral and multifocal. There were strong associations of malignant behavior with size >4 cm, extratesticular growth, gross or microscopic necrosis, high-grade cytologic atypia, vascular space invasion, and a mitotic rate greater than 3 mitoses per 10 high-power fields.115 Management guidelines are as for other SCTs.
ADENOCARCINOMA OF THE RETE TESTIS History This tumor was first described by Curling.116 Feek and Hunter117 outlined the histological criteria for diagnosis and Schoen and Rush118 emphasized its aggressive nature. Sarma and Weillbaecher119 reviewed the literature indicating that long-term survival postorchidectomy was possible.
Epidemiology and Etiology This is a very rare tumor with only about 45 cases reported in the English medical literature.120 The age at presentation ranges from 30 to 91 years with a median of 47 years and a mean of 50 years.119,121,122 The etiology is unknown although some patients have had a history of undescended testis,123 chronic epididymitis,124 or trauma.
Pathology and Biology The rete testis forms part of the collecting system of the testis, and tumors arising in the region tend to be located at the hilum. A full description of these tumors is given by Mostofi and Price,125 who described the tendency for this tumor to form papillary structures, although solid and tubular patterns are also seen. There is usually moderate nuclear pleomorphism and mitotic activity. The differential diagnosis includes germ cell tumor invading the rete testis, adenocarcinoma arising from embryonic remnants, and particularly, mesothelioma. Also, in cryptorchidism the relative increase or even hyperplasia of the cells of the collecting system can be misinterpreted as Adenocarcinoma of the Rete Testis (ART). Criteria for diagnosis61 include a predominant location in the rete, a pattern consistent with origin from rete epithelium, continuity between tumor and normal rete epithelium, and absence of a primary carcinoma elsewhere. The differential diagnosis includes malignant mesothelioma, certain ovarian-type tumors, metastatic adenocarcinoma, epididymal carcinoma, and malignant SCT.126 ART is an aggressive tumor by virtue of both local extension and recurrence after excision, and wide dissemination. In 17 cases reviewed by Sarma with adequate follow-up data, 10 patients developed metastases within one year of presentation.119 Sites of metastatic involvement included regional nodes, lung, and liver. In a literature review by Sanchez-Chapado et al.,127 information about disease-free survival was collected in 38 patients. As many as 40% of them died within the first year of diagnosis. Three- and fiveyear disease-free survivals were 49% and 13%, respectively. Tumors that were organ-confined and small (testicular mass <5 cm in maximum diameter) behaved better than those disseminated at diagnosis or those of a bigger size.
Clinical Features The tumor may arise in either testis. The most common presenting feature is an enlarging scrotal mass, which may be tender and can have an associated hydrocele. The primary tumors can infiltrate widely, one case presenting with a scrotal sinus and urethral fistula with involvement of the prostate, trigone, and ureteric orifices noted at autopsy.128 ART can also present with symptoms because of either nodal or metastatic disease.
Investigation and Staging The authors recommend CT scan of the thorax and abdomen to assess extent of disease at diagnosis. Chest x-ray and routine hematologic and biochemical tests should be performed
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES
as well as serum estimations of AFP and BHCG, and placental alkaline phosphatase to help in the differential diagnosis.
Management and Prognosis Surgery
A radical inguinal orchidectomy is recommended because local recurrence has been recorded after scrotal interference;129 this may be considered as an indication for hemiscrotectomy. Only two cases are recorded in which a RPLND was performed and where there is adequate follow-up information.130 In one case the nodes were negative for tumor and in the other a solitary nodal metastasis was removed; both patients remain disease-free at 7 and 3.5 years, respectively. Radiotherapy
This has been prescribed to treat extensive local disease at the primary site, following orchidectomy as an adjuvant to the regional lymph nodes, and to treat metastatic disease. One patient presenting with extensive local disease and treated with radiotherapy alone (dose not stated) died 40 days from presentation, having had no apparent response.128 In the five patients receiving adjuvant radiotherapy, one died at 2 months with metastases involving the skin and inguinal nodes. Follow-up times were less than 1 year in the remainder, the impact of radiotherapy in an adjuvant setting remaining unclear.119 Of 14 patients presenting with or developing metastases, radiotherapy was prescribed in 7. In only one case was comment made about a beneficial response to radiotherapy, the patient remaining symptom-free at 5 months from presentation. Dose –response data and information on local control with radiotherapy could not be ascertained,119 although in the report by Whitehead et al.131 no regression was seen despite an applied dose of 76 Gy using cobalt-60. Chemotherapy
This has been used infrequently and only in the setting of overt metastatic disease. No responses have been seen to combinations of cyclophosphamide, 5-fluorouracil, and actinomycin D132 or to single-agent methotrexate.119 Recommendations
Patients should have a radical inguinal orchidectomy and a hemiscrotectomy if scrotal violation has occurred. This establishes the diagnosis and should control the primary site. If no evidence of metastases is present on staging then consideration should be given to RPLND, based on isolated reports of benefit. An alternative policy is one of surveillance with particular attention being paid to the regional lymph nodes, reserving surgery for patients relapsing in the site alone. Patients with metastatic disease limited to the regional nodes should be managed by surgery. Patients with distant metastases should be treated with palliative intent, reserving radiotherapy and chemotherapy for specific symptoms. More
75
information is required on the chemotherapy responsiveness of these tumors, especially using combinations that are active in germ cell tumors.133 The overall prognosis is poor, with a 5-year survival of only 13%. Prompt diagnosis and resection of a small-sized tumor appears to offer the best chance for cure.
MALIGNANT MESOTHELIOMA OF THE TUNICA VAGINALIS Introduction The tunica vaginalis surrounds the testis, being formed from the processus vaginalis, which is part of the peritoneum that is carried down to the scrotum during the descent of the testis. MMTV is extremely rare and was first described in 1957.134 Its association with prior asbestos exposure has been emphasized.135 Surgery represents the only potentially curative treatment and must be meticulously performed in view of the tendency for local recurrence. Careful appraisal of regional nodes is important as this represents a major pathway of dissemination. In the presence of metastatic disease, radiotherapy and chemotherapy are used palliatively with optimal treatment yet to be defined.
Epidemiology and Etiology Malignant mesotheliomas can arise from any part of the body where a mesothelial membrane exists;136 tumors of the tunica vaginalis, however, are very rare accounting for less than 5% of all mesotheliomas.137 Less than 80 cases have been reported in the literature in the last 30 years as reviewed recently by Plas et al.138 The age range of these patients was 7–87 years; with more than two-thirds of the cases being reported in patients older than 45 years with a median age of 60 years. Although there are several mechanisms hypothesized for the development of tunical mesothelioma, the exposure to asbestos or asbestos-containing materials remains the only established risk factor. In the literature review by Plas,138 asbestos exposure was documented in 34.2% of the patients. Even this may be an underestimate because detailed case notes were available in only half of the patients. Even a family history of asbestos exposure has been associated with the development of MMTV.139 Other suggested associations include chronic immunosuppression,140 chromosomal abnormalities such as losses on 1p, 3p, 6q, 9q, and monosomy 22,141 radiation exposure,142 Simian virus exposure,143 trauma, and previous hernia repair.135,144
Pathology and Biology The tumor usually presents as a hydrocele associated with a scrotal mass. The hydrocele fluid is often clear but can be bloodstained and the tumor is seen as a papillary structure or as nodules145 standing out from the clear smooth lining of the sac. Great variation in histologic appearance occurs, but like pleural mesothelioma, solid, papillary glandular, and occasionally biphasic patterns occur. It may be possible to trace continuity between the cells lining the
76
GENITOURINARY CANCER
hydrocele and the tumor itself. When this is not demonstrable, distinguishing MMTV from metastatic adenocarcinoma can present difficulty. Much of the literature in this respect is concerned with pleural or peritoneal mesotheliomas, the rarity of intra-scrotal tumors preluding investigation of large series. However, similar considerations probably apply to the histological diagnosis of MMTV. In general, mesothelial cells may produce hyaluronic acid but, unlike many adenocarcinomas, do not produce periodic acid-Schiff (PAS)-positive mucin.146 Immunohistochemically, detection of carcino-embryonic antigen (CEA) or LeuM1 is much more likely in adenocarcinoma than mesothelioma.147 Mesothelial cells also have distinct ultrastructural features, including abundant, elongated microvilli, and electron microscopy can be helpful in diagnosis.148 Regional lymph node involvement has been reported in three of eight cases undergoing laparotomy, and diffuse visceral nodules were seen in a further two cases.135 Distant metastases at presentation are rare but are seen on relapse, with pulmonary involvement being most readily apparent. MMTV also tends to recur locally and in the regional lymph nodes.135 Time from initial therapy to relapse can vary considerably, from months to many years. In a series of 24 patients there were nine recurrences, three by 1 year, five by 2 years and seven by 3 years from presentation.135
Clinical Features and Prognosis MMTV presents with a scrotal mass in association with a hydrocele and there may be a history of prior asbestos exposure. The tumor can produce multiple nodules145,149 and occasionally malignant mesothelioma of the peritoneum can spread to involve the tunica vaginalis and vice versa. The presence of a bilateral MMTV was reported in only two patients (3.8%) in the review by Plas.138 Primary metastatic presentation was seen in 15% of the cases. Lymphatic spread was most common followed by spread to lung, liver, and pleura. Rare cases of cerebral metastasis have also been reported.150 Interestingly, the review also highlighted the difficulty of diagnosing the disease preoperatively. In only 2 of the 74 published cases (2.7%) was an accurate preoperative diagnosis made of the malignant disease. The clinical course is variable although prognostic information may be derived from details of presentation and histology.151 Plas et al.138 reported that age over 60 years and presence of metastatic disease were significant factors for a poorer outcome on univariate analysis. Patients with a positive history of asbestos exposure also had a poor outcome with a shorter disease-free interval.
Investigations and Staging Ultrasonography of the testis is a useful initial diagnostic test and the features recorded on ultrasonography have been summarized in a recent report.152 In view of the propensity to spread to involve abdominal lymph nodes and lung, lymphangiography and CT scan of the thorax and abdomen are recommended along with routine hematologic and biochemical investigations. The role of staging lymphadenectomy is controversial.132
Management Surgery
Surgery is the mainstay of treatment for MMTV and a radical inguinal orchidectomy is the optimal surgical procedure for localized tumor. Trans-scrotal procedures are associated with local recurrence and where scrotal violation has occurred hemiscrotectomy should be considered. After primary surgery, complete remission was reported in 47.5% patients for a median follow-up of 12 months. Most patients with recurrent tumor went on to develop disseminated disease (83.9%). The median time to tumor recurrence was 10.5 months (range, 2–180 months). More than 60% of recurrences developed within the first 2 postoperative years.138 Patients who have radiological evidence of regional lymph node involvement with no evidence of distant metastases require lymph node dissection, which can be curative – one such patient with histologically confirmed regional node involvement survived 15 years disease-free following this procedure.153 RPLND as an adjuvant therapy is more controversial. As no other therapy except surgery is curative in MMTV an argument for a retroperitoneal lymph node dissection at presentation can be made in a patient with good general health; however, benefit from this treatment depends on the abdominal nodes being the only site of metastases. Of eight patients undergoing laparotomy three were found to have lymph node involvement, of whom only one survived disease-free. Three patients had negative laparotomies, of whom one relapsed with groin nodes, one at the primary site, and one remained disease-free. The two remaining patients were discovered to have multiple visceral nodules of disease at laparotomy. Sixteen patients did not undergo laparotomy and of these, only two relapsed with abdominal disease at 5 and 10 years respectively.135 A reasonable alternative approach would therefore be to monitor patients radiologically, reserving lymphadenectomy for those who relapse only in nodes. Radiotherapy
Radiotherapy alone has been prescribed as part of the primary management of MMTV in 10 of 74 cases reported in the literature.138,154 – 156 The total doses prescribed varied from 25 to 60 Gy, and fractionation and overall time were not specified. A complete remission was reported in 5 (50%) of 10 cases for a maximum follow-up of 12 months. Followup times ranged from 1.5 to 36 months and the impact of radiotherapy on survival is not clear. Radiotherapy has also been used for those patients who develop local recurrence following surgery.135 Of three patients with such a complication one received radiotherapy (45 Gy total dose over approximately 1 month) following resection of the recurrence and, although he subsequently died from pulmonary metastases, he suffered no further local recurrences. Radiotherapy can also be prescribed palliatively and occasional short partial responses are recorded.157 Chemotherapy
A number of agents have been used to treat recurrent or metastatic MMTV. In the review by Plas, partial remission,
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES
defined as stable disease or reduction of tumor volume, was reported in 2 of 10 cases receiving chemotherapy (20%). No improvement in symptoms or tumor size was seen in six patients (60%), and none of the patients experienced complete remission. Combined chemotherapy and radiotherapy were given to six patients with disseminated disease and partial remission was seen in three patients, including one who had had stable disease for 16 years.138 Doxorubicin-containing chemotherapy probably is the most active.135 The similarity between MMTV and malignant mesothelium at other sites would suggest that chemotherapy is largely ineffective.149,158 New agents such as pemetrexed should be investigated.159 Recommendations
Radical inguinal orchidectomy and hemiscrotectomy in those patients with previous scrotal violation lead to a high local control rate. Patients with no evidence of metastatic disease should be followed up clinically with special attention being paid to the primary site and the regional lymph nodes. Patients relapsing in these sites with no evidence of other disease should undergo further surgery, and if resection margins are in doubt, consideration may be given to highdose local radiotherapy. Similarly, patients presenting with metastatic disease solely involving the regional nodes should undergo radical surgery in an attempt to cure. Patients relapsing with disseminated disease or those who have unresectable disease should be managed palliatively. The efficacy of current chemotherapy is unproven although transient responses have occurred either to doxorubicin alone or in combination with cyclophosphamide and 5-fluorouracil. It is also likely that high doses of radiation would be required to cause tumor regression. Pemetrexed has been shown to have activity against mesothelioma at other sites and should be investigated in this context.
PARATESTICULAR RHABDOMYOSARCOMA History Rokitansky is credited with the first case that reported rhabdomyosarcoma affecting the spermatic cord,160 and in 1934 a review by Hirsch161 pointed out that the majority of cases occurred in childhood. Tanimura and Furata162 reviewed the literature and noted that the majority of patients died with disseminated disease within 1 year of onset and that the overall prognosis was poor. Alexander,163 reported two long-term survivors who presented in early childhood at the age of 2.5 years and 3 months respectively, and thought that the prognosis of this disease in young children was not necessarily as bad as it had been claimed previously. The view was upheld in Gowing’s series160 and by Green164 when reviewing available data from older surgical patients. The value of chemotherapy in rhabdomyosarcoma was clearly demonstrated in the 1970s, and with the advent of effective chemotherapy Olive et al.165 were able to demonstrate that regional surgery or radiotherapy was unnecessary in lymphogram-negative patients managed by orchidectomy and adjuvant chemotherapy, thus diminishing potential toxicity from these procedures. At present, attempts are being
77
made to reduce toxicity from chemotherapy in early-stage disease and to improve results for poor-risk patients, and the overall cure rates are over 80%.166
Epidemiology and Etiology Rhabdomyosarcoma is the most frequently encountered malignant tumor affecting the soft tissues in childhood. Between 1954 and 1973, 2048 cases of malignancy in childhood were recorded by the Manchester Children’s Tumour Registry (CTR), and of these, 85 cases (4%) were rhabdomyosarcoma. Of 150 consecutive cases of soft tissue sarcoma seen by that group, 94 (63%) were rhabdomyosarcomas.167 The overall incidence in the United Kingdom is approximately four cases per year per million of the population under 15 years168 and this figure is in keeping with the United States experience reported by Young and Miller.169 Geographic variations in incidence are noted, with Sweden having a particularly low incidence of approximately one case per year per million in children.170 A slight male to female preponderance is noted from collected series.14 In the CTR it was 1.5:1.167 PTR represents 4 to 8% of rhabdomyosarcoma at all sites. The etiology of rhabdomyosarcoma is unknown. The fact that the tumor tends to occur at sites of fusion between the three layers of embryonic tissues and is more frequent in children with other anomalies171 implicates problems of organ development in the etiology. Cytogenetic analyses have demonstrated a consistent t(2;13) translocation in the alveolar subtype, and loss of heterozygosity on chromosome 11 occurs frequently.172 An association with genetically transmitted disease exists, rhabdomyosarcoma occurring more frequently with von Recklinghausen’s disease.163 Interestingly, in von Recklinghausen’s disease, compound tumors consisting of schwannian elements and rhabdomyosarcoma are seen.173 These are sometimes called “Triton” tumors after the experiments of Locatelli,174 in which implantation of the cut end of the sciatic nerve into Tritons (a small salamander) induced the growth of supernumerary limbs, leading to the supposition that endoneural cells may be able to differentiate into muscle. There may also be a higher incidence of breast cancer among relatives,163 a pattern consistent with the Li–Fraumeni cancer syndrome.175 The age incidence of rhabdomyosarcoma shows two peaks, an early one around 5 years of age and one later on during adolescence.161 This bimodal distribution in the age of presentation is also seen in PTR. In the 22 cases referred to the BTTP series,160 a better prognosis was associated with younger age.
Pathology and Biology PTR usually arises in the spermatic cord but it can compress or invade neighboring structures such as the epididymis or testes, and may be very extensive. The largest lesion in the BTTP series measured 14 cm × 12 cm × 7 cm and weighed 980 g.160 The histologic subtype of PTR is most commonly embryonal rather than pleomorphic or alveolar and, as indicated
78
GENITOURINARY CANCER
by Willis,176 resembles primitive embryonic tissue. In a recent Italian and German Cooperative study, 84% of the 198 patients with PTR had embryonal histology while only 8% of the patients had alveolar histology, a proportion significantly smaller than in the rhabdomyosarcoma population as a whole (20% to 30%).177 The embryonal and alveolar histologies have distinctive molecular characteristics that assist diagnostic confirmation. Unique translocations between the FKHR gene on chromosome 13 and either the PAX3 gene on chromosome 2 or the PAX7 gene on chromosome 1 are characteristic of alveolar rhabdomyosarcoma.178,179 Genomic amplification is rare in embryonal rhabdomyosarcoma, although gains of whole chromosomes occur commonly while gene amplification occurs commonly in alveolar rhabdomyosarcoma.180 Research on myocyte development shows that crossstriations do not become apparent until the 14th week of embryonic life and the absence of such striations does not therefore preclude the diagnosis of rhabdomyosarcoma.181,182 The usual histological appearance is of a myxoid stroma in association with small dark ovoid or spindle cells with varying degrees of myoblastic differentiation.167 These ovoid cells may show enlargement around an eccentrically placed nucleus, producing the so-called “tadpole” or “tennis racket” cells. Two histopathologic classifications are shown in Table 6.183,184 The prognostic significance of such groupings is not yet clear. Although electron microscopy can help establish the diagnosis by showing actin and myosin filaments and Z bands, immunohistochemistry is much more sensitive.185 Antibodies to the muscle intermediate filament desmin186 and (to a lesser extent) the skeletal muscle protein myoglobin,187 are particularly useful for immunohistological diagnosis.188,189 These may help differentiate rhabdomyosarcoma from other small round cell tumors of childhood such as neuroblastoma, Ewing’s sarcoma, or lymphoma, and from other paratesticular sarcomas in adults.190
Clinical Features
Table 6 Classification of rhabdomyosarcoma.
The rhabdomyosarcoma group of the International Society of Paediatric Oncology (SIOP) classification183 1. Embryonal type (293 cases) (a) Dense (i) Poorly differentiated (ii) Well differentiated (b) Loose (i) Botryoid (ii) Nonbotryoid (c) Alveolar 2. Adult type (1 case)
37% 14% 11% 15%
Intergroup rhabdomyosarcoma classification (IRS) system (581 cases): Gaiger et al., 1981184 1. 2. 3. 4. 5. 6. 7.
Embryonal Alveolar Botryoid Pleomorphic Special undifferentiated type 1 Special undifferentiated type 2 Undifferentiated mesenchymal sarcoma
In rhabdomyosarcoma, site, stage, and histologic type are important prognostic factors,191 – 195 paratesticular and orbital primaries being favorable. Both the Intergroup Rhabdomyosarcoma Study (IRS) Group196 and an Italian Cooperative Study,197 have confirmed the better prognosis of embryonal compared to alveolar or pleomorphic histology. A spindle cell subtype of embryonal histology is identified having a storiform growth pattern with abundant collagen and highly differentiated by immunohistochemistry and electron microscopy. Clinical data from 173 patients with spindle cell subtypes of the paratesticular lesions revealed that they almost always had an association with clinical groups of limited disease (74.4%, with group I and 23.3%, with group II disease) and a significantly better prognosis (95.5% survival at 5 years) when compared with patients with the classic embryonal histology.198 LaQuaglia et al.199 identified the importance of completeness of surgical resection in a multivariate analysis of mortality in 28 patients with PTR. Other important and statistically significant factors from large studies are tumor resectability, local invasiveness, nodal involvement, patient’s age at diagnosis (>10 years) and tumor size (>5 cm).177,200 Patients with either local invasiveness or with tumor size more than 5 cm represented the worst group. A majority of the patients with PTR have localized disease. In a recently published European study,177 retroperitoneal lymph node assessment was done with CT scan in all 216 patients and retroperitoneal lymph node involvement was detected in 21 patients (10%). The disease was localized in 92% of the cases as compared with 25% for rhabdomyosarcomas at other sites. The incidence of node involvement does not vary with age. Hematogenous metastases are uncommon at presentation, occurring in less than 5% of patients, mainly in liver, bone marrow, and lung. The most common sites of recurrence after treatment for early-stage disease include the groin, retroperitoneal lymph nodes, lung, bone, and bone marrow.201
57% 19% 6% 1% 4% 3% 10%
The most common presenting feature is an enlarging painless scrotal mass. Duration of symptoms varied in a series202 of 18 patients between 24 hours and 4 months, and in seven cases the tumor was initially unrecognized (hydrocele in four, hernia in two, epididymitis in one). An associated hydrocele may be present.203 Less commonly, symptoms are due to metastatic disease, especially in the regional lymph nodes. The presence of a scrotal mass in a child, especially if separate from the testis, should suggest the possibility of rhabdomyosarcoma.
Investigation The following investigations are recommended: full blood count, differential and erythrocyte sedimentation rate, AFP and BHCG estimation, renal and liver function tests, chest x-ray, CT scan of the thorax and abdomen, isotope bone scan, and bone marrow aspirate and biopsy. Also the primary tumor should be imaged clearly in view of the adverse significance of local extension.204
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES
The International Union against Cancer (UICC) published a tumor-nodes-metastasis (TNM) classification in 1982205 for rhabdomyosarcoma which has been widely accepted. In Table 7, the IRS classification206 is also shown and has been compared with the UICC TNM classification.196
PT3c pN1a
7 years of age compared to those over 7 years of age. For example, of the seven cases of PTR reported by Malek et al.208 essentially treated by orchidectomy, three patients remained disease-free (follow-up of 1, 3, and 39 years) and four patients subsequently developed metastases (all older children). Arlen et al.153 also correlated the risk of recurrence with older age. Surgery has been used diagnostically to sample retroperitoneal nodes and also as a therapeutic measure. However, the use of RPLND in PTR is controversial with conflicting data emerging from the European and IRS studies. In the Italian and German Cooperative Group study,177 surgical assessment of the retroperitoneal lymph nodes was recommended as a staging procedure for all patients in the earlier studies. Thereafter, it was avoided in patients with clearly negative nodes on imaging and reserved for patients with doubtful retroperitoneal involvement. No significant difference was observed in the rate of positive node detection between the two periods. These data contrast with experience in the Intergroup Rhabdomyosarcoma Study Group (IRSG) studies209,210 which required ipsilateral RPLND for all patients with PTR treated in IRS III (1984–91), but changed to computerized tomography (CT) evaluation in IRS IV (1991 through 1997) which demonstrated a reduction in both 3-year failure-free survival and OS (86% and 92%) compared with 92% and 96% in IRS III, respectively. This was attributed to the failure of imaging to detect para-aortic lymph node disease at diagnosis and, hence, inappropriate downstaging and undertreatment. Surgery also has a role in the management of patients who relapse and, when combined with chemotherapy and radiotherapy, has led to prolonged survival.165
pN1b
Radiotherapy
Management and Prognosis Surgery
Radical inguinal orchidectomy both confirms the diagnosis and usually completely removes the primary tumor. Resection of the scrotal skin is usually performed when there is scrotal tissue involvement or for primary re-excision after a prior trans-scrotal approach. In patients with residual disease after primary surgery that is not amenable to re-excision, a second surgery after chemotherapy is recommended.207 With orchidectomy alone, information from collected surgical series suggests an approximately 50% 2-year relapsefree survival rate, this figure being higher in those under Table 7 T, N, and M staging of rhabdomyosarcoma.196
TNM T1
Summary
T2a
Confined to organ / tissue <5 cm 5 cm Involving other organs / tissues Effusion <5 cm
T2b
5 cm
T3/4 N1
(Not applicable) Regional involvement
T1a T1b T2
pTNM Limited to organ
pT1
Excision complete Invasion beyond organ Excision incomplete Microscopic residual tumor Macroscopic residual tumor Nonresectable tumor Nodes completely resected Nodes incompletely resected
pT2
pT3a pT3b
M0
No distant metastasis M1 Metastasis present Intergroup rhabdomyosarcoma study (IRS):206 Group I
Group II
Group III Group IV
79
Localized disease, completely removed, regional nodes not involved I.1 Confined to muscle or organ of origin I.2 Contiguous involvement with infiltration outside the muscle organs of origin, as through fascial planes Inclusion in this group includes both gross impression of complete removal and microscopic confirmation of complete removal II.1 Grossly removed tumor with microscopic residual disease; no evidence of gross residual tumor; no evidence of regional node involvement II.2 Regional disease, completely removed (no microscopic residual disease) II.3 Regional disease with involved nodes, grossly removed, but with evidence of microscopic residual disease Incomplete resection or biopsy with gross residual disease Distant metastatic disease present at onset
TNM, tumor-nodes-metastasis; pTNM, pathological tumor-node-metastasis.
The IRS series examined the role of local radiotherapy in 13 patients with completely excised primaries and no evidence of nodal metastases, randomizing patients to receive or not to receive local radiotherapy, in addition to adjuvant chemotherapy. No patient in either arm developed local recurrence, suggesting that radiotherapy to the primary site is not usually indicated but could be considered if scrotal contamination has occurred.201 Radiotherapy to the regional nodes has also been recommended as an adjuvant in stage I PTR and also to treat known abdominal or pelvic nodal disease often in conjunction with chemotherapy and surgery. However, radiotherapy to the para-aortic and pelvic lymph nodes in young children will produce growth impairment. In the IRS series201 described earlier, the good abdominal control of disease may reflect adequacy of node removal rather than control of disease by radiotherapy or chemotherapy. Tefft et al.211 addressed the question of regional lymph node irradiation in rhabdomyosarcoma of the genitourinary tract in childhood, reporting on 58 cases. Thirty-eight patients had lymph node sampling of which 15 were positive; 11 out of 15 patients with positive lymph nodes and 6 out of 23 with negative lymph nodes received radiotherapy, as did 10 patients with no node sampling. The radiation prescription varied up to a
80
GENITOURINARY CANCER
maximum of 45 Gy in 5–6 weeks. In the entire series, only one patient (who received 45 Gy to the regional nodes) failed in regional nodes. Despite the lack of demonstrable benefit either in terms of local control or survival, the authors suggested that regional node irradiation was a wise precaution in node positive disease, but that doses should be limited to 35 Gy in 4 weeks. Radiotherapy in combination with surgery and chemotherapy can result in prolonged survival in patients relapsing in regional lymph nodes.165 Sequential hemibody radiotherapy (5 Gy to each half with a 6–8 week gap between treatments) in conjunction with chemotherapy and autologous marrow grafting if required was used for six patients with disseminated rhabdomyosarcoma (two cases were PTR). Five out of six patients died of disease with a median duration of survival of 30 months, and one patient remained disease-free at 15 months from diagnosis.212 Chemotherapy
Rhabdomyosarcoma in childhood responds to single agents but it was the success of combinations that led to chemotherapy having a major role in the management of PTR. The adjuvant value of chemotherapy in rhabdomyosarcoma in which all disease has been apparently surgically resected was demonstrated by Heyn et al.213 where 8 of 15 patients (53%) treated by surgery and local radiotherapy developed recurrence of metastases versus 3 of 17 (17.6%) receiving adjuvant chemotherapy (actinomycin D and vincristine). This achieved statistical significance (p = 0.03). Of interest is that, of the 11 relapsing patients, 7 died from disease, 6 of whom died in the nonchemotherapy arm despite chemotherapy for relapse. The impact of adjuvant chemotherapy on microscopic nodal disease in PTR was often obscured in early studies201 by the routine use of either para-aortic node dissection or abdominal radiotherapy. The study by Olive et al.165 on 19 children with complete tumor removal and negative lymphangiograms confirms the effectiveness of combination chemotherapy as an adjuvant in this group. All but 1 of the 19 children received adjuvant chemotherapy alone with combinations of vincristine, doxorubicin, actinomycin D, and cyclophosphamide, and of these 18 children only one relapsed with spermatic cord and iliac lymph node involvement, and he was successfully salvaged with combined modality treatment and was considered cured 56 months later. Multiagent chemotherapy is an important component of the multidisciplinary approach in the treatment of these tumors. Various combinations and dose schedules have been used in different protocols in Europe and North America. Overall, regimens have been progressively reduced in intensity and duration over the years. The active agents include vincristine, dactinomycin, cyclophosphamide, ifosfamide, and doxorubicin. In the Italian and German study, 106 patients with lowrisk disease i.e., favorable histology, T1N0M0, and complete resection at diagnosis (IRS group I), received chemotherapy alone. The 5-year survival of this subset was 99.1%.177 The International Society of Pediatric Oncology (SIOP) studies (MMT 84 and MMT 89) systematically studied the
role of combination chemotherapy in PTR.200 The standard first-line chemotherapy for all patients in MMT 84 was combination therapy with ifosfamide, vincristine, and dactinomycin. In the MMT 89 study, an attempt was made to avoid alkylating agents for those patients in whom tumors had been completely resected at primary surgery (stage I pT1). These patients received only vincristine and dactinomycin. Important prognostic factors influencing survival were large tumors (>5 cm) and males aged >10 years. Because of the low frequency of metastatic PTR, inferences regarding the role of chemotherapy in metastatic PTR may be drawn from studies of rhabdomyosarcoma at other sites. Using vincristine and actinomycin D in conjunction with appropriate surgery and radiotherapy, Heyn et al.214 achieved a 20% 5-year survival in 14 patients with metastatic disease at presentation. In the Italian and German Cooperative study, the 5-year survival rate for the PTR patients with metastases was 22.2%.177 The IRS206,215 tested vincristine and actinomycin D against those two agents plus cyclophosphamide (vincristine, doxorubicin, cyclophosphamide [VAC]) and found no survival difference between the two regimens in patients with microscopic residual disease and/or nodal involvement with approximately a 70% relapse-free survival at 5 years. They also compared VAC and VAC plus doxorubicin in patients with more advanced disease (gross residual disease/systemic metastases), achieving a response rate of over 80% in each arm with no differences in duration of response or survival being noted. In the report of Maurer et al.206 423 children had been entered on the IRS series, 85 of whom had microscopic or regional nodal disease and 151 had gross residual or metastatic disease. Fewer than 10% of patients with metastatic disease at presentation were alive at 2 years compared to over 60% with gross residual disease postsurgery at presentation. Of 14 patients with PTR treated with VAC ± doxorubicin and ± radiotherapy, there were 12 complete remissions of whom 3 relapsed, leading to a 3-year survival estimate of 64%.191 In the Children’s Solid Tumor Group study,154 the 5year predicted actuarial survival rate for children with rhabdomyosarcoma confined to the tissue of origin and with no evidence of nodal or metastatic spread was 86% compared with 21% for those with extension outside the tissue of origin. Of the 73 children reported, 14 had distant metastases at presentation and were treated with VAC chemotherapy, surgery being usually confined to biopsy only and radiotherapy being prescribed to bulky disease. Approximately 15% survived for 2 years. The results of treatment are shown in Table 8. There is a suggestion that ifosfamide be used in place of cyclophosphamide.216 Between 1974 and 1986, 17 cases of PTR have been managed at the Royal Marsden Hospital;185 12 patients had no evidence of dissemination at presentation and 10 of these received adjuvant chemotherapy; four of this group relapsed in regional nodes at 14–35 months and three were salvaged with further chemotherapy, radiotherapy, and surgery, and remained disease-free at 2–10 years from presentation. Three children had regional nodal involvement at presentation and with a combined approach all were disease-free at 9 months to 4 years from presentation. Of the two children
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES
81
Table 8 Rhabdomyosarcoma – results of treatment.
Series
Stage
Number of patients
Treatment
% Disease-free survival
Median duration of observation
Ferrari177a
Gp I Gp II Gp III I – III I – III I II – IV All I All All Gp I Gp II Gp III Gp IV All
164 21 13 27 69 19 13 17 12 12 18 57 20 4 14 28
C S, R, C S, R, C C C C – S, R, C R, C S, C S, R, C S, R, C S, R, C R, C R, C S, R, C
91 95 76 93 78 89 46 88 90 74 89 93 90 67 67 57
9 years – – 7 years 7 years >3 years – 5 years 5 years – 4 years – 3 years – – –
SIOP MMT 84200 SIOP MMT 89101 Olive165b Hamilton193c Loughlin195 Blyth217 Raney201a
LaQuaglia199 a
IRS staging system (see Table 7). SIOP staging system. c RMH/Barts staging system. Note: S, Abdominal surgery; R, radiotherapy; C, chemotherapy; RMH, Royal Marsden Hospital. b
with widespread metastases one died from PTR at 29 months from presentation and the other was disease-free at 7 years after presenting with hypercalcemia, bone marrow infiltration, and bone destruction from PTR, which was managed with orchidectomy and 2 years of chemotherapy with combinations of vincristine, doxorubicin, actinomycin, and oral cyclophosphamide. Recommendations
Primary surgery consists of a high inguinal orchidectomy. RPLND in not routinely recommended for staging. Adjuvant treatment is indicated following primary surgery for clinical stage I PTR and the most extensive experience has been with VAC chemotherapy. Cure rates are high (approximately 85%) and it may be possible that less-toxic chemotherapy, such as deletion of the alkylating agent could be employed, particularly in those with negative surgical staging. The role of postchemotherapy irradiation is not established and we would only recommend it in postpubertal patients, in whom the prognosis is not good and there would be a less significant effect on bone growth. An inverted-Y field is usually treated to 40 Gy in 20–25 fractions over 4–5 weeks. Patients presenting with abdominal lymph node involvement should be managed by combined modalities. It is beneficial to use chemotherapy postorchidectomy and use nodal disease as a marker of response. In patients with a complete radiological response, an argument for continued chemotherapy alone can be made but we recommend surgical confirmation of response. In those presenting prior to their growth spurt with residual disease postchemotherapy, the authors would recommend a surgical approach to control intra-abdominal disease. In those postpuberty, an argument for irradiation can be made and if this were to be combined with excision of previously-abnormal nodes, our preference would be the sequence of radiotherapy after surgery. Disseminated PTR should be managed as rhabdomyosarcoma at other sites, and although high-response rates using
combinations of vincristine, cyclophosphamide, and doxorubicin are seen208 the overall survival is poor.
ACKNOWLEDGMENTS This work was undertaken in The Royal Marsden NHS Trust who received a proportion of its funding from the NHS Executive; the views expressed in this publication are those of the authors and not necessarily those of the NHS Executive. This work was supported by the Institute of Cancer Research, the Bob Champion Cancer Trust and Cancer Research UK Section of Radiotherapy [CUK] grant number C46/A2131.
REFERENCES 1. Woodward P, et al. Germ cell tumours. In Eble JN, et al. (eds) Pathology and Genetics of Tumours of the Urinary System and Male Genital Organs. Lyon, France: IARC Press, 2004: 221 – 249. 2. Malassez M. Lymphadenome du testicule. Bull Soc Anat Paris 1877; 52: 176. 3. Hutchinson J. Lymphosarcoma of both testes with considerable interval of time. Br Med J 1889; 1: 413. 4. Ficari A. A case of lymphosarcoma with metastases in unusual situations. J Pathol Bacteriol 1950; 62: 103 – 11. 5. Gowing NFC. Malignant lymphoma of the testis. In Pugh RCB (ed) Pathology of the Testis. London, England: Blackwell Scientific Publications, 1976: 334 – 355. 6. Kiely JM, et al. Lymphoma of the testis. Cancer 1970; 26: 847 – 52. 7. Turner RR, Colby TV, MacKintosh FR. Testicular lymphomas: a clinicopathologic study of 35 cases. Cancer 1981; 48: 2095 – 102. 8. Zucca E, et al. Patterns of outcome and prognostic factors in primary large-cell lymphoma of the testis in a survey by the International Extranodal Lymphoma Study Group. J Clin Oncol 2003; 21: 20 – 7. 9. Whelan SL, et al. Patterns of Cancer in Five Continents, IARC Scientific Publications No 102. Lyon, France: International Agency for Research on Cancer, 1990. 10. Waterhouse JAH. Epidemiology of testicular tumours. J Soc Med 1985; 78: 3 – 7. 11. Tepperman BS, et al. Non-Hodgkin’s lymphoma of the testis. Radiology 1982; 142: 203 – 8.
82
GENITOURINARY CANCER
12. Eckert H, Smith JP. Malignant lymphoma of the testis. Br Med J 1963; ii: 891 – 4. 13. Mehrotra RML, Wahal KM, Agarwal PK. Testicular lymphoma: a clinicopathologic study of 22 cases. Indian J Pathol Microbiol 1978; 21: 91 – 6. 14. Ciatto S, Cionini L. Malignant lymphoma of the testis. Acta Radiol Oncol 1979; 18: 572 – 6. 15. Read G. Lymphomas of the testis – results of treatment 1960 – 1977. Clin Radiol 1981; 32: 687 – 92. 16. Hayes MMM, Sacks MI, King HS. Testicular lymphoma: a retrospective review of 17 cases. S Afr Med J 1983; 64: 1014 – 6. 17. Freilone R, et al. Combined modality treatment with a weekly brief chemotherapy (ACOP-B) followed by locoregional radiotherapy in localized-stage intermediate-to high-grade non-Hodgkin’s lymphoma. Ann Oncol 1996; 7: 919 – 24. 18. Economopoulos T, et al. Primary extranodal non-Hodgkin’s lymphoma in adults: clinicopathological and survival characteristics. Leuk Lymphoma 1996; 21: 131 – 6. 19. Wahal KM, Mehrotra R, Agarwal PK. Extranodal lymphomas in North India. J Indian Med Assoc 1983; 80: 130 – 2. 20. Rappaport H. Tumors of the Hematopoietic System, Atlas of Tumor Pathology. Washington, District of Columbia: Armed Forces Institute of Pathology, 1966: 91 – 161. Section 3, fasc 8. 21. The Non Hodgkin’s Lymphoma Pathologic Classification Project. National Cancer Institute Sponsored study of classification of non Hodgkin’s lymphomas. Cancer 1982; 49: 2112 – 35. 22. Jackson SM, Montessori GA. Malignant lymphoma of the testis: review of 17 cases in British Columbia with survival related to pathological subclassification. J Urol 1980; 123: 881 – 3. 23. Baldetorp LA, et al. Malignant lymphoma of the testis. Br J Urol 1984; 56: 525 – 30. 24. Nonomura N, et al. Malignant lymphoma of the testis: histological and immunohistological study of 28 cases. J Urol 1989; 141: 1368 – 71. 25. WHO. Pathology and Genetics of Tumours of Hematopoietic and Lymphoid Tissues. Lyon, France: IARC Press, 2001. 26. Horstmann WG, Timens W. Lack of adhesion molecules in testicular diffuse centroblastic and immunoblastic B cell lymphomas as a contributory factor in malignant behaviour. Virchows Arch 1996; 429: 83 – 90. 27. Lambrechts AC, et al. Lymphomas with testicular localisation show a consistent BCL-2 expression without a translocation (14;18): a molecular and immunohistochemical study. Br J Cancer 1995; 71: 73 – 7. 28. Visco C, et al. Non-Hodgkin’s lymphoma affecting the testis: is it curable with doxorubicin-based therapy? Clin Lymphoma 2001; 2: 40 – 6. 29. Linassier C, et al. Stage I-IIE primary non-Hodgkin’s lymphoma of the testis: results of a prospective trial by the GOELAMS Study Group. Clin Lymphoma 2002; 3: 167 – 72. 30. Sussman EB, et al. Malignant lymphoma of the testis. A clinicopathologic study of 37 cases. J Urol 1977; 118: 1004 – 7. 31. Crellin AM, et al. Non-Hodgkin’s lymphoma of the testis. Radiother Oncol 1993; 27: 99 – 106. 32. Zietman AL, et al. The management and outcome of stage IA(E) nonHodgkin’s lymphoma of the testis. J Urol 1996; 155: 943 – 6. 33. Touroutoglou N, et al. Testicular lymphoma: late relapses and poor outcome despite doxorubicin-based therapy. J Clin Oncol 1995; 13: 1361 – 7. 34. Duncan PR, et al. Extranodal non-Hodgkin’s lymphoma presenting in the testicle: a clinical and pathologic study of 24 cases. Cancer 1980; 45: 1578 – 84. 35. Keldsen N, et al. Risk factors for central nervous system involvement in non-Hodgkins-lymphoma – a multivariate analysis. Acta Oncol 1996; 35: 703 – 8. 36. Sasai K, et al. Primary testicular non-Hodgkin’s lymphoma: a clinical study and review of the literature. Am J Clin Oncol 1997; 20: 59 – 62. 37. Ostronoff M, et al. Localized stage non-Hodgkin’s lymphoma of the testis: a retrospective study of 16 cases. Nouv Rev Fr Hematol 1995; 37: 267 – 72. 38. Hasselblom S, et al. Testicular lymphoma – a retrospective, population -based, clinical and immunohistochemical study. Acta Oncol 2004; 43: 758 – 65.
39. Fonseca R, et al. Testicular lymphoma is associated with a high incidence of extranodal recurrence. Cancer 2000; 88: 154 – 61. 40. Zouhair A, et al. Outcome and patterns of failure in testicular lymphoma: a multicenter Rare Cancer Network Study. Int J Radiat Oncol Biol Phys 2002; 52: 652 – 6. 41. Talerman A. A Primary malignant lymphoma of the testis. J Urol 1977; 118: 783 – 6. 42. Hurley LJ, et al. Bilateral primary non-Hodgkin’s lymphoma of the testis. Urology 1996; 47: 596 – 8. 43. Kupfer H, von der Beek K. Recurrence in the other testicle of a nonHodgkin’s lymphoma with no other manifestations. Br J Urol 1995; 76: 516. 44. Kellie SJ, Pui CH, Murphy SB. Childhood non Hodgkin’s lymphoma involving the testis: clinical features and treatment outcome. J Clin Oncol 1989; 7: 1066 – 70. 45. Haddy TB, Sandlund JT, Magrath IT. Testicular involvement in young patients with non-Hodgkin’s lymphoma. Am J Pediatr Hematol Oncol 1988; 10: 224 – 9. 46. Miller TP, Jones SE. Chemotherapy of localised histiocytic lymphoma. Lancet 1979; 1: 358 – 60. 47. Leonard MP, et al. Burkitt’s lymphoma of the testis: an unusual scrotal mass in childhood. J Urol 1990; 143: 104 – 6. 48. Carbone PP, et al. Report of the committee on Hodgkin’s disease staging classification. Cancer Res 1971; 31: 1860 – 1. 49. Buskirk SJ, et al. Primary lymphoma of the testis. Int J Radiat Oncol Biol Phys 1982; 8: 1699 – 703. 50. Roche H, et al. Stage IE non Hodgkin’s lymphoma of the testis: a need for a brief aggressive chemotherapy. J Urol 1989; 141: 554 – 6. 51. Miller TP, et al. Three cycles of CHOP (3) plus radiotherapy (RT) is superior to eight cycles of CHOP (8) alone for localised intermediate and high grade non-Hodgkin’s lymphoma (NHL): a Southwest Oncology Group Study (Abstract 1257). Proc Am Soc Clin Oncol 1996; 15: 411. 52. Skarin A, et al. Therapy of diffuse histiocytic (DH) and undifferentiated (DU) lymphoma with high dose methotrexate and citrovonum rescue (MTX/CF), Bleomycin (B), Adriamycin (A), Cyclophosphamide (C), Oncovin (O) and Decadron (D) (MBACOP). Am Soc Clin Oncol 1980; 21: 463. 53. Canellos GP, et al. The M-BACOP combination chemotherapy regimen in the treatment of diffuse large cell lymphoma. Semin Hematol 1987; 24: 2 – 7. 54. Klimo P, Connors JM. MACOP-B chemotherapy for the treatment of diffuse large cell lymphoma. Ann Intern Med 1985; 102: 596 – 602. 55. Hoskins PJ, et al. Prognostic variables in patients with diffuse largecell lymphoma treated with MACOP-B. J Clin Oncol 1991; 9(2): 220 – 6. 56. McKelvey EM, et al. Hydroxyldaunomycin (Adriamycin) combination chemotherapy in malignant lymphoma. Cancer 1976; 38: 1484 – 93. 57. Fisher RI, et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin’s lymphoma. N Engl J Med 1993; 328: 1002 – 6. 58. Coiffier B, et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 2002; 346: 235 – 42. 59. Horwich A, Peckham MJ. ‘Bad risk’ non Hodgkin’s lymphoma. Semin Oncol 1983; 20: 35 – 55. 60. Mostofi FK, Theiss EA, Ashley DJB. Tumors of specialized gonadal stroma in human male patients. Cancer 1959; 12: 944 – 57. 61. Young RH, Talerman A. Testicular tumors other than germ cell tumors. Semin Diagn Pathol 1987; 4: 342 – 60. 62. McLaren K, Thomson D. Localisation of S-100 protein in a Leydig and Sertoli cell tumour of testis. Histopathology 1989; 15: 649 – 52. 63. Campbell EJM, et al. Sex and reproduction. In Campbell EJM, et al. (eds) Clinical Physiology, 5th ed. London, England: Blackwell Scientific Publications, 1984: 651 – 708. 64. Oosterhuis JW, et al. A malignant mixed gonadal stromal tumor of the testis with heterologous components and I(12P) in one of its metastases. Cancer Genet Cytogenet 1989; 41: 105 – 14. 65. Symington T, Cameron KM. Testicular diseases. In Pugh RCB (ed) Pathology of the Testis. London, England: Blackwell Scientific Publications, 1976: 259 – 303.
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES 66. Mostofi FK. Testicular tumours. Cancer 1973; 32: 1186 – 201. 67. Ward JA, et al. Interstitial cell tumour of the testis: report of two cases. J Endocrinol 1960; 20: 1622 – 32. 68. Johnstone G. Pre-pubertal gynaecomastia in association with an interstitial cell tumour of the testis. Br J Urol 1967; 39: 211 – 20. 69. Kim I, Young RH, Scully RE. Leydig cell tumours of the testis. Am J Surg Pathol 1985; 19: 177 – 92. 70. Assi A, et al. Leydig cell tumor of the testis: a cytohistological, immunohistochemical, and ultrastructural case study. Diagn Cytopathol 1997; 16: 262 – 6. 71. Goswitz JJ, Pettinato G, Manivel JC. Testicular sex cord-stromal tumors in children: clinicopathologic study of sixteen children with review of the literature. Pediatr Pathol Lab Med 1996; 16: 451 – 70. 72. Cheville JC, et al. Leydig cell tumor of the testis: a clinicopathologic, DNA content, and MIB-1 comparison of nonmetastasizing and metastasizing tumors. Am J Surg Pathol 1998; 22: 1361 – 7. 73. Iczkowski KA, et al. Inhibin A is a sensitive and specific marker for testicular sex cord-stromal tumors. Mod Pathol 1998; 11: 774 – 9. 74. Young RH. Sex cord-stromal tumors of the ovary and testis: their similarities and differences with consideration of selected problems. Mod Pathol 2005; 18(Suppl 2): S81 – 98. 75. Harms D, Kock LR. Testicular juvenile granulosa cell and Sertoli cell tumours: a clinicopathological study of 29 cases from the Kiel paediatric tumour registry. Virchows Arch 1997; 430: 301 – 9. 76. Valensi P, et al. Endocrine investigations in two cases of feminizing Leydig cell tumour. Acta Endocrinol (Copenh) 1987; 115: 365 – 72. 77. Thomas JC, Ross JH, Kay R. Stromal testis tumors in children: a report from the prepubertal testis tumor registry. J Urol 2001; 166: 2338 – 40. 78. Grem JL, et al. Metastatic Leydig cell tumor of the testis. Cancer 1986; 58: 2116 – 9. 79. Mineur P, et al. Feminizing testicular Leydig cell tumor: hormonal profile before and after unilateral orchidectomy. J Clin Endocrinol Metab 1987; 64: 686 – 91. 80. Sasano H, et al. Leydig cell tumor of the testis: analysis of testosterone production and secretin by three-dimensional histoculture. Endocr J 1996; 43: 73 – 8. 81. Solish SB, et al. Molecular characterization of a Leydig cell tumor presenting as congenital adrenal hyperplasia. J Clin Endocrinol Metab 1989; 69: 1148 – 52. 82. Lockhart JL, et al. Nonfunctioning interstitial cell carcinoma of testes. Urology 1976; 8: 392 – 4. 83. Mosharafa AA, et al. Does retroperitoneal lymph node dissection have a curative role for patients with sex cord-stromal testicular tumors? Cancer 2003; 98: 753 – 7. 84. Parker RG. Treatment of apparent solitary pulmonary metastases. J Thorac Cardiovasc Surg 1958; 36: 81 – 7. 85. Peschel R, et al. Management of adult Leydig-cell testicular tumors: assessing the role of laparoscopic retroperitoneal lymph node dissection. J Endourol 2003; 17: 777 – 80. 86. Sawin PD, VanGilder JC. Spinal cord compression from metastatic Leydig’s cell tumor of the testis: case report. Neurosurgery 1996; 38: 407 – 11. 87. Feldman PS, et al. Malignant Leydig cell tumor: clinical, histologic and electron microscopic features. Cancer 1982; 49: 714. 88. Davies JM. Testicular cancer in England and Wales: some epidemiological aspects. Lancet 1981; i: 928 – 32. 89. Azer PC, Braunstein GD. Malignant Leydig cell tumour: objective tumor response to o, p - DDD. Cancer 1981; 47: 1251 – 5. 90. Farkas LM, et al. High frequency of metastatic Leydig cell testicular tumours. Oncology 2000; 59: 118 – 21. 91. Godec CJ. Malignant Sertoli cell tumour of the testicle. J Urol 1985; 16: 185 – 8. 92. Giglio M, et al. Testicular Sertoli cell tumours and relative subtypes. Analysis of clinical and prognostic features. Urol Int 2003; 70: 205 – 10. 93. Ceccamea A, et al. Feminizing Sertoli cell tumour associated with Peutz-Jeghers syndrome (histologic and ultrastructural study). Tumori 1985; 71: 379 – 85. 94. Young S, et al. Feminizing Sertoli cell tumors in boys with PeutzJeghers syndrome. Am J Surg Pathol 1995; 19: 50 – 8.
83
95. Ventura T, et al. Light microscopic, immunocytochemical and ultrastructural study of a case of Sertoli cell tumor of the testis. Tumori 1987; 73: 649 – 53. 96. Nielsen K, Jacobsen GK. Malignant Sertoli cell tumour of the testis, an immunohistochemical study and a review of the literature. APMIS 1988; 96: 755 – 60. 97. Young RH, Koelliker DD, Scully RE. Sertoli cell tumors of the testis, not otherwise specified: a clinicopathologic analysis of 60 cases. Am J Surg Pathol 1998; 22: 709 – 21. 98. Comperat E, et al. Non-Leydig sex-cord tumors of the testis. The place of immunohistochemistry in diagnosis and prognosis. A study of twenty cases. Virchows Arch 2004; 444: 567 – 71. 99. Henley JD, Young RH, Ulbright TM. Malignant Sertoli cell tumors of the testis: a study of 13 examples of a neoplasm frequently misinterpreted as seminoma. Am J Surg Pathol 2002; 26: 541 – 50. 100. Gabrilove JL, et al. Feminising and non-feminising Sertoli cell tumours. J Urol 1980; 124: 757 – 67. 101. Weitzner S, Addridge JE, Lamar Weems W. Stertoli cell tumour of the testis. J Urol 1979; 13: 87 – 9. 102. Sharma S, Seam RK, Kapoor HL. Malignant Sertoli cell tumour of the testis in a child. J Surg Oncol 1990; 44: 129 – 31. 103. Herera LO, et al. Malignant (androblastoma) Sertoli cell tumor of testis. J Urol 1981; 18: 287 – 90. 104. Madson E, Hultberg B. Metastasizing Sertoli cell tumors of the human testis-a report of two cases and a review of the literature. Acta Oncol 1990; 29: 946 – 9. 105. Proppe KH, Scully RE. Large cell calcifying Sertoli cell tumor of the testis. Am J Clin Pathol 1980; 74: 607 – 19. 106. Waxman M, et al. Large cell calcifying Sertoli tumor of the testis. Light microscopic and ultrastructural study. Cancer 1984; 54: 1574 – 81. 107. Perez-Atayde AR, et al. Large-cell calcifying Sertoli cell tumour of the testis. An ultrastructural immunocytochemical and biochemical study. Cancer 1983; 51: 2287 – 92. 108. Buchino JJ, Buchino JJ, Uhlenhuth EE. Large-cell calcifying Sertoli cell tumour. J Urol 1989; 141: 953 – 4. 109. Plata C, et al. Large cell calcifying Sertoli cell tumour of the testis. Histopathology 1995; 26: 255 – 9. 110. Carney JA, et al. The complex of myxomas, spotty pigmentation and endocrine overactivity. Medicine 1985; 64: 270 – 83. 111. Carney JA. The Carney complex (myxomas, spotty pigmentation, endocrine overactivity, and schwannomas). Dermatol Clin 1995; 13: 19 – 26. 112. Noszian IM, et al. Bilateral testicular large-cell calcifying Sertoli cell tumor and recurrent cardiac myxoma in a patient with Carney’s complex. Pediatr Radiol 1995; 25: S236 – 7. 113. Proppe KH, Dickersin GR. Large cell calcifying Sertoli cell tumour of the testis. Light microscopic and ultrastructural study. Hum Pathol 1982; 13: 1109 – 14. 114. Horn T, Jao W, Keh PC. Large-cell calcifying Sertoli cell tumor of the testis: a case report with ultrastructural study. Ultrastruct Pathol 1983; 4: 359 – 64. 115. Kratzer SS, Ulbright TM, Talerman A. et al. Large cell calcifying Sertoli cell tumor of the testis: contrasting features of six malignant and six benign tumors and a review of the literature. Am J Surg Pathol 1997; 21: 1271 – 80. 116. Curling TB. Observations on cystic disease of the testicle. Med Chir Trans 1853; 36: 449. 117. Feek JD, Hunter WC. Papillary carcinoma arising from rete testis. Arch Pathol 1945; 30: 399. 118. Schoen SS, Rush BF. Adenocarcinoma of the rete testis. J Urol 1959; 82: 356 – 63. 119. Sarma DP, Weillbaecher TG. Adenocarcinoma of the rete testis. J Surg Oncol 1985; 30: 67 – 71. 120. Menon PK, et al. A case of carcinoma rete testis: histomorphological, immunohistochemical and ultrastructural findings and review of literature. Indian J Cancer 2002; 39: 106 – 11. 121. Crisp-Lindgren N, et al. Papillary adenocarcinoma of rete testis. Autopsy findings, histochemistry, immunohistochemistry, ultrastructure and clinical correlations. Am J Surg Pathol 1988; 12: 492 – 501. 122. Mrak RE, Husain MM, Schaefer RF. Ultrastructure of metastatic rete testis adenocarcinoma. Arch Pathol Lab Med 1990; 114: 84 – 8.
84
GENITOURINARY CANCER
123. Dundon C. Carcinoma of the rete testis occurring ten years after orchidopexy. Br J Urol 1952; 24: 58 – 63. 124. Desberg T, Tanno V. Adenocarcinoma of the rete testis. J Urol 1964; 91: 87 – 9. 125. Mostofi FK, Price EB. Tumors of the Male Genital System. Washington, District of Columbia: Armed Forces Institute of Pathology, 1973: 170 – 173. 126. Amin MB. Selected other problematic testicular and paratesticular lesions: rete testis neoplasms and pseudotumors, mesothelial lesions and secondary tumors. Mod Pathol 2005; 18(Suppl 2): S131 – 45. 127. Sanchez-Chapado M, Angulo JC, Haas GP. Adenocarcinoma of the rete testis. Urology 1995; 46: 468 – 75. 128. Roy JB, et al. Adenocarcinoma of rete testis. J Urol 1979; 14: 270 – 2. 129. Schapira HE, Engel M. Adenocarcinoma of rete testis. N Y State J Med 1972; 72: 1283. 130. Orozco RE, Murphy WM. Carcinoma of the rete testis: case report and review of the literature. J Urol 1993; 150: 974 – 7. 131. Whitehead ED, Valensi QJ, Brown JS. Adenocarcinoma of the rete testis. J Urol 1972; 107: 992 – 9. 132. Smith JJ, et al. Retroperitoneal lymph node dissection in malignant mesothelioma of tunica vaginalis testis. J Urol 1990; 144: 1242 – 3. 133. Peckham MJ, et al. The treatment of metastatic germ-cell testicular tumours with bleomycin, etoposide and cis-platin (BEP). Br J Cancer 1983; 47: 613 – 9. 134. Barbera V, Rubino M. Papillary mesothelioma of the tunica vaginalis. Cancer 1957; 10: 183 – 9. 135. Antman K, et al. Malignant mesothelioma of the tunica vaginalis testis. J Clin Oncol 1984; 2: 447 – 51. 136. Antman KH, et al. Mesothelioma. Cancer: Princ Pract Oncol Updat 1989; 3: 1 – 16. 137. Serio G, et al. Malignant mesothelioma of the testicular tunica vaginalis. Eur Urol 1992; 21: 174 – 6. 138. Plas E, Riedl CR, Pfluger H. Malignant mesothelioma of the tunica vaginalis testis: review of the literature and assessment of prognostic parameters. Cancer 1998; 83: 2437 – 46. 139. Vianna NJ, Polan AK. Non-occupational exposure to asbestos and malignant mesothelioma in females. Lancet 1978; 1: 1061 – 3. 140. Lew F, et al. High frequency of immune dysfunctions in asbestos workers and in patients with malignant mesothelioma. J Clin Immunol 1986; 6: 225 – 33. 141. Taguchi T, et al. Recurrent deletions of specific chromosomal sites in 1p, 3p, 6q, and 9p in human malignant mesothelioma. Cancer Res 1993; 53: 4349 – 55. 142. Cavazza A, et al. Post-irradiation malignant mesothelioma. Cancer 1996; 77: 1379 – 85. 143. Stenton SC. Asbestos, Simian virus 40 and malignant mesothelioma. Thorax 1997; 52(Suppl 3): S52 – 7. 144. Amin R. Case report: malignant mesothelioma of the tunica vaginalis testis – an indolent course. Br J Radiol 1995; 68: 1025 – 7. 145. Tyagi G, et al. Malignant mesothelioma of tunica vaginalis testis. Urology 1989; 34: 102 – 4. 146. Kannerstien M, Churg J. Histochemistry in the diagnosis of malignant mesothelioma. Ann Clin Lab Sci 1973; 3: 207 – 11. 147. Khoury N, et al. A comparative immunohistochemical study of peritoneal and ovarian serous tumours, and mesotheliomas. Hum Pathol 1990; 21: 811 – 9. 148. Ehya H. Cytology of mesothelioma of the tunica vaginalis metastasis to the lung. Acta Cytol 1985; 29: 79 – 84. 149. Prescott S, et al. Malignant mesothelioma of the tunica vaginalis testis: a case report. J Urol 1988; 140: 623 – 4. 150. Mah E, Bittar RG, Davis GA. Cerebral metastases in malignant mesothelioma: case report and literature review. J Clin Neurosci 2004; 11: 917 – 8. 151. Grove A, Jensen ML, Donna A. Mesotheliomas of the tunica vaginalis testis and hernial scs. Virchows Arch 1989; 415: 283 – 92. 152. Mak CW, et al. Malignant mesothelioma of the tunica vaginalis testis. Br J Radiol 2004; 77: 780 – 1. 153. Arlen M, Grabstald H, Whitemore WF Jr. Malignant tumours of the spermatic cord. Cancer 1969; 23: 525 – 32.
154. Kingston JE, McElwain TJ, Malpas JS. Childhood rhabdomyosarcoma: experience of the Children’s Solid Tumour Group. Br J Cancer 1983; 48: 195 – 207. 155. Kasdon EJ. Malignant mesothelioma of the tunica vaginalis propria testis: report of two cases. Cancer 1969; 23: 1144 – 50. 156. Eimoto T, Inoue I. Malignant fibrous mesothelioma of the tunica vaginalis: a histologic and ultrastructural study. Cancer 1977; 39: 2059 – 66. 157. Fligiel Z, Kaneko M. Malignant mesothelioma of the tunica vaginalis propria testis in a patient with asbestos exposure: a case report. Cancer 1976; 37: 1478 – 84. 158. Aisner J, Wiernik PH. Chemotherapy in the treatment of malignant mesothelioma. Semin Oncol 1981; 8: 335 – 43. 159. Vogelzang NJ, Porta C, Mutti L. New agents in the management of advanced mesothelioma. Semin Oncol 2005; 32: 336 – 50. 160. Gowing NFC. Paratesticular Tumours of Connective Tissue and Muscle. London, England: Blackwell Scientific Publications, 1976. 161. Hirsch EF. Rhabdomyosarcoma of the spermatic cord. Am J Cancer 1934; 20: 398 – 403. 162. Tanimura H, Furata M. Rhabdomyosarcoma of the spermatic cord. Cancer 1968; 22: 1215 – 20. 163. Alexander F. Pure testicular rhabdomyosarcoma. Br J Cancer 1968; 22: 498 – 501. 164. Green DM. Diagnosis and Management of Malignant Solid Tumour in Infants and Children. Martinus Nijhoff Publications, 1985: 15 – 90. 165. Olive D, et al. Para-aortic lymphadenectomy is not necessary in the treatment of localised paratesticular rhabdomyosarcoma. Cancer 1984; 54: 1283 – 7. 166. Pappo AS, Shapiro DN, Crist WM. Rhabdomyosarcoma. Biology and treatment. Pediatr Clin North Am 1997; 44: 953 – 72. 167. Marsden HB. The pathology of soft-tissue sarcomas with emphasis on childhood tumours. In D’Angio GJ, Evans AE (eds) Bone Tumours and Soft Tissue Sarcomas. London, England: Edward Arnold, 1985: 14 – 25. 168. Draper GJ, et al. Childhood Cancer in Britain 1953 – 1975: Incidence, Mortality and Survival. London, England: HMSO, 1982. 169. Young JL, Miller RW. Incidence of malignant tumours in US children. J Pediatr 1975; 86: 254 – 8. 170. Ericsson JL-E, Karnstrom L, Mattsson B. Childhood cancer in Sweden 1958 – 1974. Acta Paediatr Scand 1978; 67: 425 – 32. 171. McKeen EA, et al. Rhabdomyosarcoma complicating multiple neurofibromatosis. J Pediatr 1978; 93: 992 – 3. 172. Scrable H, et al. A model for embryonal rhabdomyosarcoma tumorigenesis that involves genome imprinting. Proc Natl Acad Sci USA 1989; 86: 7480 – 4. 173. Woodruss JM, et al. Peripheral nerve tumors with rhabdomyosarcomatous differentiation (malignant Triton: tumors). Cancer 1973; 32: 426. 174. Locatelli P. Formation de Membres Surnum´eraires, 20e r´eunion. Turin, Italy: CR Association des Anatomistes, 1925: 279 – 82. 175. Birch JM, et al. Cancer in the families of children with soft tissue sarcoma. Cancer 1990; 66: 2239 – 48. 176. Willis RA. Pathology of Tumours. London, England: Butterworth, 1948. 177. Ferrari A, et al. Paratesticular rhabdomyosarcoma: report from the Italian and German Cooperative Group. J Clin Oncol 2002; 20: 449 – 55. 178. Barr FG. Molecular genetics and pathogenesis of rhabdomyosarcoma. J Pediatr Hematol Oncol 1997; 19: 483 – 91. 179. Merlino G, Helman LJ. Rhabdomyosarcoma – working out the pathways. Oncogene 1999; 18: 5340 – 8. 180. Weber-Hall S, et al. Gains, losses, and amplification of genomic material in rhabdomyosarcoma analyzed by comparative genomic hybridization. Cancer Res 1996; 56: 3220 – 4. 181. Patton RB, Horn RC Jr. Rhabdomyosarcoma: clinical and pathological features and comparison with human fetal and embryonal skeletal muscle. Surgery 1962; 52: 572 – 84. 182. Porterfield JF, Zimmermann LE. Rhabdomyosarcoma of the orbit. A clinico pathological study of 55 cases. Virchows Arch A 1962; 335: 329 – 44. 183. Flamant F, et al. Primary chemotherapy in the treatment of rhabdomyosarcoma in children: Trial of the International Society of
RARE TUMORS OF THE TESTIS AND PARATESTICULAR TISSUES
184.
185.
186.
187.
188. 189. 190.
191.
192.
193. 194.
195. 196.
197.
198.
199. 200.
Paediatric Oncology SIOP preliminary results. Radiother Oncol 1985; 3: 187 – 293. Gaiger AM, Soule EH, Newton WA. Pathology of rhabdomyosarcoma. Experience of the intergroup rhabdomyosarcoma study 1972 – 1978. NCI Monogr 1981; 56: 19 – 27. Fisher C. The value of electron microscopy and immunohistochemistry in the diagnosis of soft tissue sarcomas: a study of 200 cases. Histopathology 1990; 16: 441 – 55. Miettinen M, et al. Alveolar rhabdomyosarcoma. Demonstration of the muscle type of intermediate filament protein, desmin, as a diagnostic aid. Am J Pathol 1988; 108: 246 – 51. Mukai K, Rosai J, Hallway B. Localisation of myoglobin in normal and neoplastic human muscle cells using an immunoperoxidase method. Am J Surg Pathol 1979; 3: 373 – 6. Toskos M, Howard R, Costa J. Immunohistochemical study of alveolar and embryonal rhabdomyosarcoma. Lab Invest 1983; 48: 148 – 55. Eusebi V, et al. Immunocytochemistry of rhabdomyosarcoma. The use of four different markers. Am J Surg Pathol 1986; 10: 293 – 9. Soosay GN, et al. Paratesticular sarcomas revisited: a review of cases in the British Testicular Tumour Panel and Registry. Br J Urol 1996; 77: 143 – 6. Lawrence WJ, et al. Lymphatic metastases with childhood rhabdomyosarcoma. A report from the Intergroup Rhabdomyosarcoma Study. Cancer 1987; 60: 910 – 5. Rodary C, et al. Prognostic factors in 281 children with non metastatic rhabdomyosarcoma (RMS) at diagnosis. Med Pediatr Oncol 1988; 16: 71 – 7. Hamilton CR, Pinkerton R, Horwich A. The management of paratesticular rhabdomyosarcoma. Clin Radiol 1989; 40: 314 – 7. Ghavimi F, et al. Genitourinary rhabdomyosarcoma (RMS) in children: memorial Sloan-Kettering cancer experience (meeting abstract). Med Pediatr Oncol 1989; 17: 308. Loughlin KR, et al. Genitourinary rhabdomyosarcoma in children. Cancer 1989; 63: 1600 – 6. Lawrence W Jr et al., Children’s Cancer Study Group. Pediatric Oncology Group. Pretreatment TNM staging of childhood rhabdomyosarcoma: a report of the Intergroup Rhabdomyosarcoma Study Group. Cancer 1997; 80: 1165 – 70. Carli M, et al. Prognostic significance of histology in childhood rhabdomyosarcoma (RMS): improved survival with a new histologic ’leiomyomatous’ subtype (meeting abstract). Proc Am Soc Clin Oncol 1990; 9: 1153. Leuschner I, et al., Spindle cell variants of embryonal rhabdomyosarcoma in the paratesticular region. A report of the Intergroup Rhabdomyosarcoma Study. Am J Surg Pathol 1993; 17: 221 – 30. LaQuaglia MP, et al. Mortality in pediatric paratesticular rhabdomyosarcoma: a multivariate analysis. J Urol 1989; 142: 473 – 8. Stewart RJ, et al. Treatment of children with nonmetastatic paratesticular rhabdomyosarcoma: results of the Malignant Mesenchymal
201.
202. 203. 204.
205.
206. 207.
208. 209.
210.
211.
212.
213. 214. 215. 216.
217.
85
Tumors studies (MMT 84 and MMT 89) of the international society of pediatric oncology. J Clin Oncol 2003; 21: 793 – 8. Raney RBJ, et al. Paratesticular sarcoma in childhood and adolescence. A report from the intergroup rhabdomyosarcoma studies I and II, 1973 – 1983. Cancer 1987; 60: 2337 – 43. Johnson DE, McHugh TA, Jaffe N. Paratesticular rhabdomyosarcoma in childhood. J Urol 1982; 128: 1275 – 6. Zaslau S, et al. Rhabdomyosarcoma of tunica vaginalis masquerading as hydrocele. Urology 2005; 65: 1001. Rodary C, et al. Prognostic factors in 951 nonmetastatic rhabdomyosarcoma in children: a report from the international rhabdomyosarcoma workshop. Med Pediatr Oncol 1991; 19: 85 – 95. Spiesse B, et al. pTNM classification of malignant tumours. In Spiesse B, et al. (eds) Atlas: Illustrated Guide to the TNM, 2nd ed. Berlin, Germany; Heidelberg, Germany; New York, Tokyo, Japan: Springer, 1982. Maurer HM. The intergroup rhabdomyosarcoma study: update, November 1978. Natl Cancer Inst Monogr 1981; 56: 61 – 8. Dall’Igna P, et al. Primary transcrotal excision for paratesticular rhabdomyosarcoma: is hemiscrotectomy really mandatory? Cancer 2003; 97: 1981 – 4. Malek RS, Utz DC, Farrow GM. Malignant tumours of the spermatic cord. Cancer 1972; 29: 1108 – 13. Wiener ES, et al. Retroperitoneal node biopsy in paratesticular rhabdomyosarcoma. J Pediatr Surg 1994; 29: 171 – 7; discussion 178. Wiener ES, et al. Controversies in the management of paratesticular rhabdomyosarcoma: is staging retroperitoneal lymph node dissection necessary for adolescents with resected paratesticular rhabdomyosarcoma? Semin Pediatr Surg 2001; 10: 146 – 52. Tefft M, et al. Radiation to regional nodes for rhabdomyosarcoma of the genitourinary tract in children: is it necessary? Cancer 1980; 45: 3065 – 8. Munoz LL, et al. Magna-field irradiation and autologous marrow rescue in the treatment of pediatric solid tumours. Int J Radiat Oncol Biol Phys 1983; 9: 1951 – 4. Heyn RM, et al. The role of combined chemotherapy in the treatment of rhabdomyosarcoma in children. Cancer 1974; 34: 2128 – 42. Heyn RM. Late effects of therapy in rhabdomyosarcoma. Clin Oncol 1985; 4: 287 – 97. Maurer HM, Moon T, Donaldson M. The Intergroup rhabdomyosarcoma study: a preliminary report. Cancer 1977; 40: 2015 – 26. Crist WM, et al. Prognosis in children with rhabdomyosarcoma: a report of the intergroup rhabdomyosarcoma studies I and II. Intergroup Rhabdomyosarcoma Committee [see comments]. J Clin Oncol 1990; 8: 443 – 52. Blyth B, et al. Paratesticular rhabdomyosarcoma: results of therapy in 18 cases. J Urol 1990; 144: 1450 – 3.
Section 2 : Head and Neck Cancer
6
Uncommon Tumors of the Oral Cavity and Adjacent Structures
A. Robert Kagan, Stephen I. Shibata, Michael P. McNicoll and Najeeb S. Alshak
INTRODUCTION Ninety-five percent of the cancers of the oral cavity are squamous cell carcinomas. The remaining 5% are uncommon, by definition.1 Nine of ten patients with the classical squamous cell cancer are men, and most are elderly. The majority of these patients have sustained a lifetime of health neglect, smoking and drinking to excess, poor oral hygiene, and disregard of conspicuous ulcerated masses in their oral cavity. One uncommon variant of squamous cell carcinoma, verrucous carcinoma, may cause diagnostic difficulty. The fronds on this cauliflower-like lesion are so well differentiated that cancer is not suspected; even deep biopsies at the base of the epithelium will only show broad based invasion (see Figure 1). Frequently, patients with stage III or IV squamous cell carcinoma of the oral cavity will verbalize a symptomatic history of a few weeks, or blame the primary care physician for not recognizing the cancer 1 year ago when the tongue was painful. Since the doubling time of these squamous cancers often occurs with a time span measured in years, the patient’s neglect of long-standing symptoms is a more likely explanation for the frequent advanced stage at diagnosis. Patients with uncommon malignancies of the oral cavity, in contrast, may be of any age, are more often health compliant (as there is usually no background of tobacco and alcohol abuse), and the incidence in men and women is closer to 1 : 1. For the uncommon cancers, minor salivary gland carcinoma, bone and soft tissue sarcomas, and lymphomas, there are no precursor lesions such as leukoplakia and/or erythroplakia. The very rare prior history elements of “head and neck” radiation, Paget’s disease, or neurofibromatosis as antecedent factors prior to the development of sarcomas, do not have high incidence figures (as does erythroleukoplakia in some of the squamous malignancies), and thus have little practical value. In this chapter, we review the approach to management of uncommon tumors of the oral cavity and surrounding
tissues. By definition, these tumors are rare, and thus structured information on management paradigms and outcomes is scant. There are few specific surgical or radiotherapeutic principles that are distinct from those that apply to the more common epithelial tumors of the oral cavity, perhaps with the exception of management of small round cell tumors and malignant lymphomas. A detailed review of the management of squamous carcinoma of the oral cavity is beyond the scope of this review, and the reader is referred to standard works on this topic. With respect to uncommon malignancies, we have drawn upon the literature and on our own experience to attempt to provide the reader with a structured approach to optimal care. Where no clear guidelines exist, we have reported our own approach to management, at least as a guide.
TUMORS OF THE BONE Presentation The initial presentation of tumors of the bones surrounding the oral cavity is protean. These lesions are often insidious, with a relatively indolent initial course. The majority of tumors are found in the mandible, although occasional primary nonsquamous carcinomas will present in the maxilla. Tumors of the mandible are usually found by dental films, as a result of tooth movement engendered by a slow-growing intra-alveolar mass, or by a faster-growing mass manifesting as an enlarging, often painful jaw and/or loosening of teeth. The majority of presenting tumors are located posteriorly. As a diagnostic step, the removal of a symptomatic cyst detected on dental films associated with a tooth, by curettage, is acceptable, with the proviso that the tissue must be studied carefully to distinguish an ameloblastoma from a benign cyst. Tumors of the maxilla may present with local pain, syndromes resembling sinusitis, rhinorrhea, or with dental/gingival symptoms. Occasionally, an asymptomatic mass will be identified on a routine dental X ray.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
88
HEAD AND NECK CANCER
Figure 1 Verrucous Carcinoma – Biopsy at base showing broad based invasion.
Figure 2 Ameloblastoma – Islands of odontogenic epithelial proliferation within collagenous stroma.
Osteosarcomas, Ewing’s sarcoma/PNET (primitive neuroectodermal tumors), variants of odontogenic carcinomas, chondrosarcomas, and non-Hodgkin’s (large-cell) lymphomas should be suspected when there is bone destruction or soft tissue invasion. The bone sarcomas are often relatively large at diagnosis, and thus incisional biopsy is recommended. Pain in the upper jaw and teeth can be caused by neuropathy of the second branch of the fifth cranial nerve (V2) and similarly by V3 neuropathy in the lower jaw. Myeloma of the jaw differs from that of the mucous membranes, as it is usually a part of a disseminated disease. Myeloma cells found in a biopsy of the mandible in proximity to the gingiva have occasionally been misdiagnosed as malignant squamous cell carcinoma. Odontogenic Tumors
About 9% of tumors of the oral cavity are odontogenic.2 These may evolve histologically along different pathways and are classified accordingly as (i) tumors related to odontogenic epithelium; (ii) tumors related to odontogenic mesenchyme; (iii) mixed tumors. These tumors represent a spectrum, from relatively benign to quite malignant tumors, and their natural histories and responses to treatment reflect their cellular lineages. Ameloblastoma, one of the variants of odontogenic tumor, usually with predominantly mesenchymal features (see Figure 2), must be differentiated from a dentigerous cyst with extensive squamous metaplasia. It often presents radiographically as a “soap bubble” expansion of bone with a painless swelling causing malocclusion of teeth or dentures. Recurrences are more common after initial treatment with curettage than when treated by resection. The rare malignant ameloblastoma can metastasize to the lungs. Odontogenic carcinomas are very rare. The malignant squamous pathology may initially suggest the origin as a metastasis from a lung primary, or the clear cell pattern, if present, may be reminiscent of metastases from a renal cell carcinoma. The presence of facial anesthesia or paresthesia suggests the likelihood that the bone tumor seen on
Figure 3 Giant cell tumor of bone – Sheets of mononuclear cells with scattered multinucleated giant cells.
the image is most likely malignant. Differential diagnoses should include malignant ameloblastoma, carcinosarcoma, and fibrosarcoma. The “tip off” is rapid growth with bone destruction. There is considerable functional and morphological overlap between the variants of odontogenic tumor, which represent a continuum between poorly differentiated carcinoma and sarcoma. Mucoepidermoid and Giant Cell Tumors
Another rare entity is the intraosseous mucoepidermoid carcinoma. It is a low-grade carcinoma that presents asymptomatically with a multilobular radiolucency on dental radiographs, often near the third molar. Giant cell tumor of bone (see Figure 3) and giant cell granuloma of bone are often benign, but with substantial recurrences if not totally removed. The giant cell granuloma especially, can mimic a malignancy because of the presence of mitoses, osteoid, and polymorphism.3
UNCOMMON TUMORS OF THE ORAL CAVITY AND ADJACENT STRUCTURES
89
Round Cell Tumors
Langerhans Cell Histiocytosis
A series of small, round cell tumors may sometimes be found in the oral cavity. The controversy concerning the histopathological differences between Ewing’s sarcoma and primitive neuroectodermal tumor has finally been resolved. Both of these small round “blue” cell tumors have the same cytogenetic translocation t(11 : 22)(q24;q21), leading to the EWS/FLI1 fusion protein MIC2,4 which can be detected immunohistochemically. As in most “oral” bone tumors, they are more frequent in the mandible than in the maxilla. Pain (uncommon with benign tumors) with swelling is the major presentation. The pattern of presentation, including risk factors (and compounded by the histological pattern) will determine the prognosis.5 Because Ewing’s sarcoma may have a larger soft tissue component relative to bone, soft tissue sarcoma secondarily invading bone, although rare, ought to be included in the differential diagnosis. This would include synovial sarcoma, malignant fibrous histiocytoma (MFH), leiomyosarcoma, fibrosarcoma, and liposarcoma.
Langerhans cell histiocytosis of bone is commonly reported in children but also occurs in adults, where it may present as a solitary osteolytic involvement of the mandible.6 Like a carcinoma or sarcoma, it may present with an ulcer or loose teeth and pain. At least one-half of adult patients will have other manifestations: bone, diabetes insipidus, skin, lung, etc. The presence of giant and inflammatory cells can make the diagnosis difficult with a small biopsy. Osteomyelitis and lymphoma are the two most difficult differential diagnoses.
Sarcomas
Diagnostic Workup
Osteosarcoma of the mandible is more often osteolytic and chondroblastic in the upper alveolus. Although described in Paget’s disease of the bone, it probably is of more concern as a second cancer in survivors of head and neck irradiation primarily delivered to the oral cavity and nasopharynx. Most primary bone sarcomas occur infrequently in the mandible. In fact, if a bone tumor is suspected in the upper alveolar ridge, one should think of squamous cell carcinoma of the maxillary sinus. Chondrosarcoma does occur more frequently in the upper alveolar ridge than the mandible, often in conjunction with involvement of the maxillary sinus. It is not unusual for a chondrosarcoma to be first diagnosed as a chondroma, or other benign bone tumor, because of the difficulty of being sure about the presence of malignant cartilage (see Figure 4). The presence of bone destruction on the imaging study strongly suggests malignancy.
The general approach to the diagnostic evaluation of uncommon tumors of the bones surrounding the oral cavity depends somewhat on the nature and site of the original presentation. Of course, careful physical examination, with judicious use of mirrors and/or flexible scopes to assess the nasopharynx and larynx, is critical for accurate assessment of the presentation and extent of disease. Biopsy is usually required early to define the nature of the specific disease process, and this, in turn, will indicate the pattern of further workup and staging that is required, usually predicated on natural history, pattern of spread, and likelihood of multifocality. For the patient who presents with a mass or an ulcerating lesion, early biopsy is indicated. Sometimes, this is achieved early when a dentist makes the initial diagnosis by intention or by accident when addressing the problem of a tooth that has been displaced. For the patient presenting with a new mass, standard hematology and biochemistry will occasionally suggest the diagnosis – for example, a high uric acid or lactate dehydrogenase (LDH) level may suggest the possibility of lymphoma or myeloma; acute leukemia may be identified on a peripheral blood film; and circulating tumor markers may occasionally indicate the nature of the tumor (for example, raised levels of carcinoembryonic antigen or other tumor markers, such as CA 19-9 or CA 15-3, which might suggest the presence of a metastatic deposit from an adenocarcinoma at another site. Assessment of human immunodeficiency virus (HIV) status and of the expression of Epstein-Barr virus (EBV) may suggest, respectively, the presence of Kaposi sarcoma/AIDS-related lymphoma or a Burkitt’s lymphoma. A most important aspect of the diagnostic workup is afforded by judicious use of radiology. Plain bone films may indicate an asymptomatic cyst or the presence of bone destruction, and this can be defined more clearly by nuclear magnetic resonance imaging (MRI) or computed tomography (CT), both of which may also help define whether cervical
Figure 4 High-grade chondrosarcoma – Markedly atypical and occasionally multinucleated chondrocytes.
Metastases
Another presentation of malignancy in the mandible is an atypical pattern of metastasis from another site. This, in turn, usually reflects the histological type. For example, if a deposit is shown to represent squamous cell carcinoma, it will most commonly be from a bronchogenic primary; if adenocarcinoma is represented, mammary carcinoma should be considered; although, if the pattern includes clear cells, a renal cell carcinoma is a likely source.
90
HEAD AND NECK CANCER
lymph node involvement is part of the presenting syndrome. For tumors of the roof of the mouth, careful assessment of the base of the orbits by CT or MRI scanning will screen common patterns of extension. For tumors of the jaw, plain radiography is often a useful first step, augmented by CT or MRI scanning. In some cases, when it is suspected that the lesion is a metastatic deposit, workup for a source of a primary tumor at another site may be required.
Management The primary treatment of most tumors of the oral cavity, both common and uncommon, is by surgical resection or by radiotherapy. In some cases, treatment outcomes may be somewhat compromised by inadvertent incomplete primary treatment – for example, when a dentist attempts to unroof a purportedly “benign” lesion by curettage. In that situation, complete resection may be possible, or (depending on the histology) completion of therapy by radiotherapy may be appropriate.
Round Cell Tumors
Chemotherapy for Ewing’s sarcoma (small, round cell tumors) is generally used as initial treatment with the option of surgery determined by the patient’s response and the initial site and extent of presentation.8,9 Chemotherapy usually consists of vincristine, doxorubicin, and cyclophosphamide and/or actinomycin D, although more recently cisplatin or carboplatin have been employed with increasing frequency. In addition, ifosfamide and etoposide have activity in this disease.10 Radiation is often used to consolidate therapy, depending on the site and extent of the lesion. The principles of management are essentially the same as those for Ewing’s sarcomas at other sites. Of importance, the distinction from PNET should be made, as these tend to be more resistant to cytotoxic chemotherapy. In addition, careful histological exclusion of lymphoma (also a small, round cell tumor in some cases) is mandatory, sometimes by special stains or immunopathology, as the treatment is quite different.
Odontogenic Tumors
Malignant ameloblastomas are best treated with complete surgical excision because curettage is often associated with local relapse.7 These tumors can metastasize, and thus a staging workup may define that local therapy will not achieve cure. There is little published experience with the management of metastatic ameloblastomas, but it is probably reasonable to attempt a trial of cisplatin-based treatment or a regimen typically used for sarcoma (e.g. doxorubicinifosfamide-mesna) in otherwise healthy patients with a good performance status as these tumors represent part of a continuum between poorly differentiated carcinomas and soft tissue sarcomas. While selecting the treatment, it is important to be sure that an expert tumor pathologist has confirmed the diagnosis, ruling out the various other variants of uncommon (and common) oral malignancy. Similarly, the optimal treatment of odontogenic carcinoma is usually with complete surgical resection, and it is usually necessary for the surgical procedure to be radical. One should not be satisfied with curettage, no matter how complete, because of the high rate of local relapse. Radiotherapy may have a palliative role for local recurrence, and there is no defined role for chemotherapy. If referral to a tertiary referral center is not possible, cisplatin-based chemotherapy is a reasonable (but unproven) option. Mucoepidermoid and Giant Cell Tumors
Intraosseous mucoepidermoid carcinoma and giant cell tumors of bone are best managed with surgical resection. For the mucoepidermoid carcinoma, segmental resection may be adequate definitive treatment, but giant cell tumors are best managed with complete or radical resection. There is no defined role for adjuvant radiotherapy or for chemotherapy, and thus incompletely resected lesions are probably best handled by repeat surgical intervention by an expert, with radical clearance, and reconstruction as necessary.
Sarcomas
Prior to the introduction of chemotherapy, the 5-year survival rate for all osteogenic sarcomas was 20%. In tumors arising in the limbs, 65–75% 5-year survival is currently reported after treatment with preoperative chemotherapy followed by surgical tumor resection, with or without radiotherapy. Commonly used regimens incorporate cisplatin and doxorubicin, with or without methotrexate. However, in osteosarcoma of the head and neck, less benefit from chemotherapy has been reported, with a 5-year survival rate less than 50%.11 The successful use of radiation to achieve local control has been reported, but chemotherapy in association with an oncologic surgical procedure is currently the most common treatment. Of importance, some of the improvement in survival may be because of improved imaging modalities such as CT, MRI, magnetic resonance angiography (MRA), and positron emission tomography and computed tomography (PET-CT), which have improved the accuracy of staging, thus facilitating the optimal selection of treatment modalities and also leading to stage migration. Alternatively, improved supportive care and the use of salvage treatments could be other factors. Cures have also been reported for patients with radiationrelated osteosarcomas. Event-free survival for 23 patients has been reported as 41% at 8 years.12 The highest cure rates (70%) are in patients with clear surgical margins and lowgrade (chondroblastic) lesions. Improvement of survival in the nonosteogenic sarcomas of the oral cavity over the years has not been as a result of chemotherapy or irradiation, but more likely because of better staging and oncologic resection designed to obtain negative surgical margins. Five-year survival rate for lowgrade cancers is relatively high, above 70%, whereas for high-grade cancers it is only around 30%. The cause of death is usually local recurrence, unlike that of osteosarcomas.13
UNCOMMON TUMORS OF THE ORAL CAVITY AND ADJACENT STRUCTURES
MELANOMA Presentation Benign nevi and melanoma in situ occur in the mucosa of the oral cavity. Melanomas mostly occur on the gingiva and the hard palate. The diagnostic clue is an asymmetrical growth with irregular outline. The color may be black, gray, or reddish. Desmoplastic melanoma may occur on the gingival and buccal mucosa and is often characterized by a lack of pigmentation. This often causes microscopic confusion with fibrosarcoma and peripheral nerve sheath tumor. Also, the malignant nature of the spindle cells may even be overlooked. Positive S-100 and HMB-45 immunostains for melanoma are often helpful in pointing the diagnosis toward melanoma. Desmoplastic melanoma is rare. It has been reported that local recurrence and death from metastases is more common than that of other types of melanomas.14
Diagnosis The principles of diagnosis of melanoma of the oral cavity are the same as that for melanoma at other sites. Accurate biopsy, with definition of the depth of invasion and the status of margins is of critical importance. For melanoma of the oral cavity, the lymphatic drainage is important and will reflect, to some extent, the site and side of presentation. As for melanoma at other sites, a search for other possible sites of primary melanoma is important, and metastases should be excluded. Although a careful history and physical examination will be important, small metastases may be seen only on CT or MRI scans of chest and abdomen, with a particular focus on the lungs and liver. Occasionally, disseminated melanoma will be manifested by osseous metastases, which are usually lytic in nature, and thus small lesions may not show up on radionucleide bone scan.
Management The key to successful treatment are, when possible, early diagnosis, and adequate surgical resection, although the natural history is often determined by the depth of invasion and the extent of initial presentation. Although radiotherapy and chemotherapy may have roles in palliation, there is no defined role for combination therapy. In this setting, the role of immunotherapy remains uncertain, although occasional sustained responses in metastatic disease will be seen with interferon or interleukin-2. There have also been anecdotal reports of responses to “standard” chemotherapy regimens, based on dacarbazine, cisplatin, and the nitrosoureas, although usually these are quite toxic and of quite short duration. There are no unique features of management of melanoma at this site, and the appropriate treatment is predicated on strategies applied to melanoma at other sites. For superficial melanoma with clear margins, surgical resection is adequate. For deeply invasive melanomas of the oral cavity or those without clear margins, there is no clearly defined management pathway. In general terms, reexcision to achieve clear margins, if possible, is important, but this situation is likely
91
to be associated with a worse prognosis. Wherever possible, assessment of the draining lymph nodes should be effected, although this may be difficult in the case of melanomas of the gingival margin of the maxilla.
SALIVARY GLAND TUMORS There are more than 500 minor salivary glands distributed throughout the oral cavity, located beneath the squamous mucosa. The key unit of each gland is the lobule, consisting of acini lined by mucous cells. From acini, the secretory fluid drains to the intercalated, interlobular, and excretory ducts. Myoepithelial cells surround the ducts and are responsible for providing contractility. Immunohistochemically, the epithelial ductal cells express cytokeratin, whereas the myoepithelial cells express S-100 and smooth muscle actin (SMA), allowing the potential for histological distinction of the tumors that arise from these cells.
Presentation Salivary gland neoplasms in the oral cavity occur with an incidence of about 3 : 1 benign-to-malignant lesions. The most common benign neoplasm is the “mixed” tumor. The term “mixed” refers to the presence of epithelial ducts together with mesenchymal spindle cells, often associated with myxo-chondroid material (see Figure 5a and b). Salivary gland carcinomas originating from the minor salivary glands in the oral cavity present as submucosal “bumps,” most commonly on the hard and soft palates. Presumed benign cysts in the buccal mucosa are often ignored by the patient and the physician, until they grow. A mucous cyst of the sublingual glands in the floor of mouth is called “ranula”. This is the common location of the cystadenocarcinoma, which fortunately is usually of low grade. A history of a new mass, recent growth of an old mass, or rapid growth of any such lesion must be treated as significant. There is so much overlap in histological features among the salivary gland carcinomas; the presence of cysts, tubular structures, clear cells, and other features not being specific to one cancer type; and the exact histopathological diagnosis may be in dispute; and thus a complete excision or large incisional biopsy may facilitate accurate histological diagnosis. About one-third of salivary gland tumors in the oral cavity are malignant. Biopsy of small specimens of these tumors that are read as benign may be incorrect. The most common sites for such malignancy are the hard and soft palates. The most common benign tumor is one where errors of sampling occur, the pleomorphic adenoma, or the mixed tumor. The lack of a capsule, a feature of the parotid, sublingual, and submandibular salivary glands, can lead toward a wrong diagnosis of malignancy. Oncocytoma is a rare benign tumor, characterized ultrastructurally by the presence of profuse mitochondria. The high-grade cancers of the salivary glands include: • High-grade mucoepidermoid carcinoma • Adenoid cystic carcinoma
92
HEAD AND NECK CANCER
(a) Figure 6 Primary squamous cell carcinoma – Prominent cellular keratinization and loosely cohesive tumor cells.
(b) Figure 5 Pleomorphic Adenoma – (a) Epithelial islands and (b) ductal structures of a myxo-chondroid stroma.
• • • • • •
Salivary duct carcinoma Adenocarcinoma NOS Carcinoma ex-pleomorphic adenoma Small cell carcinoma Squamous cell carcinoma Undifferentiated carcinoma
The 10-year survival rate, combining major and minor salivary gland carcinomas, is about 25%. Clinical metastases are detected in 30% overall, but range from about 5% in low-grade squamous cell carcinoma (see Figure 6) to nearly 65% in undifferentiated carcinoma. Minor salivary gland carcinomas of the oral cavity are mostly small and of low grade (grade 1–2), distant metastases are documented in about 25%. The low-grade carcinomas are acinic cell carcinoma, polymorphous low-grade carcinoma (terminal duct carcinoma), myoepithelial carcinoma, and basal cell carcinoma. These carcinomas have a 10-year survival rate of 90% and an incidence of distant metastases of less than 20%. Some of the discrepancy in percentages of
local recurrence and distant metastases comes from the misdiagnoses of adenoid cystic carcinoma, wherein failure in distant metastases is high, but associated with local and/or regional failure in two of three patients.15,16 The mucoepidermoid carcinoma and the adenoid cystic carcinoma are equal in incidence.17 – 19 A history of rapid growth suggests poor differentiation. For mucoepidermoid carcinoma with few cystic spaces (high grade), the cytokeratin stain will be strongly positive and the positive mucicarmine stain will be scarce (see Figures 7a and b). The local recurrence patterns are as follows: low grade, under 10%; intermediate grade, 20% (often these will be large enough to have positive margins); and high grade about 80%, often with distant metastases. Adenoid cystic carcinoma is notorious for perineural invasion (see Figures 8a –c) with subsequent metastases at the base of the skull. Metastases to the lungs without first appearing in cervical lymph nodes and other “uncommon” places such as axillary and inguinal nodes may occur. In many salivary gland cancers, the exact histopathology differs amongst pathologists; hence the importance of staging. The staging for minor salivary gland carcinomas is not identical to that of the parotid, sublingual, and submandibular glands but depends on the anatomic site of origin. For the oral cavity, T1N0 (stage I) and T2N0 (stage II) measure up to 4 cm in greatest dimension. CT and MR imaging are part of the clinical staging. T3 is greater than 4 cm in greatest dimension (stage III T3N0 and T1–3N1); T4a invades bone, deep muscles of the tongue, maxillary sinus, and skin of face; T4b connotes unresectability. Polymorphous low-grade adenocarcinoma (also called terminal duct carcinoma) is predominantly seen in the salivary glands of the oral cavity, affecting predominantly females with a 2–3 : 1 ratio of females to males. It occurs most commonly in the fifth and sixth decades of life. It is nonencapsulated and therefore easily invades adjacent tissue. It is called polymorphous because it displays solid, cystic, tubular, and papillary features in its growth patterns (see Figure 9).
UNCOMMON TUMORS OF THE ORAL CAVITY AND ADJACENT STRUCTURES
(a)
93
In a report of over 2000 salivary gland cancers from all anatomical sites, 20-year survival differences do vary by histopathological type, with acinic cell adenocarcinoma showing nearly 80% survival, mucoepidermoid carcinoma 70%, adenocarcinoma, mixed malignant tumor, adenoid cystic and squamous carcinoma from nearly 30% to 45%.22 However, the range in survival figures is more dramatic with clinicopathological staging: stage I approximately 85%, stage II 55%, and stage III/IV 12%.23 Clinicopathological staging is more reproducible from center to center than histopathological type staging or grading. Practically, if the salivary gland cancer is greater than 3 cm, it is more than likely to have invaded adjacent structures and if solid nests of cells are a part of the cancer, a poor grade should be suspected. In the majority of clinical investigations, stage III/IV and poor grade cancers are not highly curable by any treatment.
Diagnosis
(b) Figure 7 Mucoepidermoid carcinoma – (a) The malignant cells are epidermoid, mucus, and intermediate. (b) Mucin stain highlights intracytoplasmic mucin (arrow ).
Its metastatic rate to lymph nodes and local recurrence rate is below 20%, which is much lower than that of adenoid cystic carcinoma, with which it is confused because of its tendency for perineural involvement.20,21 Distant metastases are uncommon in polymorphous low-grade adenocarcinoma and mucoepidermoid carcinoma. Polymorphous low-grade adenocarcinoma has histological diversity, as does adenoid cystic carcinoma, but the cytological uniformity is present, which also distinguishes it from adenoid cystic carcinoma. Adenosquamous carcinoma may occur in the floor of mouth, tongue, and buccal mucosa. The pathologic stage at presentation is often III or IV. Unlike the intermixing of the squamous mucoid and the glandular elements in mucoepidermoid carcinoma, the squamous elements form the superficial portion and the mucoid elements the deep portion of the adenosquamous carcinoma. Even low stage tumors tend to metastasize to regional nodes.
The general principles of diagnosis and investigation for salivary gland tumors are identical to those enumerated above for other tumors of the oral cavity. Careful history taking and physical examination, with emphasis on assessment of cervical nodes and the entire oropharynx and nasopharynx are crucial. Standard imaging approaches for head and neck malignancies should be employed. However, some specific nuances bear mention. For example, in the case of the diagnostic approach to a ranula, with its inherent risk of the occult presence of a cystadenocarcinoma, fine needle aspiration is often misleading even in distinguishing benign from malignant, probably because many of these cystadenocarcinomas are of low and intermediate grade, without severe cytological atypia. Thus, a more aggressive approach to biopsy is required. Adenosquamous carcinomas, irrespective of stage, tend to metastasize to lymph nodes, and thus careful imaging with MR is essential in planning subsequent management. Many studies of head and neck cancer suggest an increasing role for PET scanning in the definition of involvement of cervical lymph nodes, although most of the data relate to classical epithelial cancers of this region. Our general approach is to be cautious about the interpretation of negative scans, but to act on information gained from unequivocally positive PET scans in this setting.
Treatment Excisional biopsy of small (1–2 cm) tumors and incisional biopsy of large submucosal masses are recommended. Larger tumors have a higher risk of being high grade and more invasive of soft tissue and bone, thereby necessitating a radical resection, rather than a wide excision. If a surgeon other than the one doing the biopsy is chosen to do the radical resection, his or her design for radical surgery can be more accurate with most of the cancer intact. There is no standard role for chemotherapy in the primary management of salivary gland tumors.24 In cases of locally recurrent or metastatic disease, cisplatin-based therapy may provide palliation;25 although, it should be
94
HEAD AND NECK CANCER
(a)
(b)
(c) Figure 8 Adenoid cystic carcinoma – (a) Cribriform growth pattern. (b) Tubular – cribriform growth with mucohyaline stroma. (c) Perineural invasion.
Figure 9 Polymorphous low-grade adenocarcinoma – Combinations of solid, ductal, trabecular, and tubal growth can be seen.
noted that other agents such as the taxanes, doxorubicin, and cyclophosphamide will occasionally yield short-term responses.
Grading is decisive in choosing between a wide excision (e.g. a segment of hard palate for grade 1 and 2) and a more aggressive oncologic procedure (e.g. maxillectomy for grade 3). Patients who die of intraoral mucoepidermoid carcinoma are few, but most of them have grade 3 cancers. In 85% of patients treated with surgery alone, the rate of local recurrence is low and postoperative irradiation is unnecessary. Patients with recurrence of low-grade cancers may be salvaged by further surgery plus or minus radiation. Mucoepidermoid carcinomas occurring in the tongue or floor of the mouth reportedly have a worse outcome. The 5-year disease-free survival rate with low and intermediate grade cancers is 80%, whereas it is 20% with high-grade cancers.17,18 This disease is potentially curable by surgery, and the roles of radiotherapy and chemotherapy remain illdefined. For the occasional patient with widely recurrent or distant metastatic disease, the literature is unclear regarding chemotherapy, but we have occasionally seen responses to cisplatin-based regimens, although no optimal regimen has been defined. Adenoid cystic carcinoma of the minor salivary glands has an overall survival rate of 20% at 20 years. Adverse factors are greater than 4-cm osseous invasion (T3, T4), and a solid growth pattern meaning scarce in cribriform or
UNCOMMON TUMORS OF THE ORAL CAVITY AND ADJACENT STRUCTURES
tubular pattern (high grade). The high local recurrence rate of 30–40% after surgery alone has guided most to recommend postoperative radiation for all adenoid cystic carcinomas. Whether this improves survival is not clear. There are no randomized studies but most retrospective investigations show a survival benefit for postoperative irradiation.19 Among 99 malignant salivary gland carcinomas of the soft palate, one-half were adenoid cystic carcinomas. Tenyear overall survival of all cancers was about 65%; for patients with negative margins it was 90%; for patients with positive margins, 50%. Positive margins were most likely if the cancer involved the antrum, nasal cavity, or pterygoid region. For all soft palate cancers, no evidence of disease (NED) survival at 10 years was 54%. Postoperative radiation was delivered only for advanced or “poor” histology. The local recurrence rate for adenoid cystic carcinoma was still 30%, with polymorphous low-grade and mucoepidermoid carcinoma it was below 10%. In the entire series of 99 patients, 15% died of cancer and 21% died of intercurrent disease.26 Once again, the role of chemotherapy is unclear, but cisplatin-based regimens are occasionally tried, usually with only short-term responses. In the case of polymorphous low-grade carcinoma, surgical excision is the key to cure. Postoperative irradiation is not needed, in contrast to what is required in adenoid cystic carcinoma, unless margins are positive and repeat surgery is not possible. In adenoid cystic carcinoma of the oral cavity, even with postoperative radiation, there is local recurrence in 20% of cases. In addition, cervical metastases occur in 10%, almost always associated with distant metastases. For polymorphous low-grade and mucoepidermoid carcinoma, most often an oncologic surgical procedure is necessary because these cancers may invade bone and soft tissue. However, patients with local recurrence, uncommon as they are, can often be salvaged by further surgery. As stated above, distant metastases are rare and if it occurs, histopathological review would be important to rule out adenoid cystic carcinoma or other high-grade salivary gland cancers. In the case of adenosquamous carcinoma, even if the presentation is of low stage, T1 or T2, elective node dissection is indicated as all these tumors have a tendency to spread to regional nodes. Imaging studies of the cervical lymph nodes may show small nodes with necrotic centers as might occur in any high-grade salivary gland carcinoma. Five-year survival is 25%, a figure that has not really been influenced by adjuvant radiotherapy or chemotherapy. For tumors that recur extensively or are metastatic, the role of chemotherapy also remains unclear, and most clinicians believe that cisplatin-based regimens offer a small chance of remission, although the literature provides few structured data to support this.
HEMATOLOGIC TUMORS Presentation
95
(a)
(b) Figure 10 Non-Hodgkin’s lymphoma – (a) Sheets of malignant lymphoid cells. (b) CD20 positive malignant B cells.
hard and soft palates, as do plasmacytomas. Lymphoma masquerading as necrotic gingivitis may be missed, as the lymphoreticular elements are mixed with the intense inflammatory process. In addition, amyloid can invade the gingiva. An “infectious” swelling around the teeth in an HIV patient may actually be a lymphoma. Necrosis of the hard palate may be part of lymphoma-like diseases such as extranodal natural killer T-cell lymphoma (NK-T) cell lymphoma, nasal type “lethal midline granuloma”, Wegener’s granulomatosis, or angiocentric lymphoma. Differential diagnostic considerations include melanoma or undifferentiated carcinoma, which may rarely have a necrotic appearance. Cancers of the nasal fossa or maxilla can ulcerate the hard palate. The important point here is that small biopsies of a cancerous necrotic lesion may be misleading, revealing only inflammation.
Non-Hodgkin’s Lymphoma
Extramedullary Plasmacytoma
Non-Hodgkin’s lymphomas of the mucous membranes (see Figures 10a and b) occur most frequently as lumps on the
Extramedullary plasmacytoma has been reported in the tonsillar and soft palate regions, and in the oral cavity,
96
HEAD AND NECK CANCER
usually presenting as a small, nondescript mass. Conversion to multiple myeloma over time occurs in as many as one-third of the patients with plasmacytoma. Lymphomas
Lymphomas of the oral cavity are rare compared with those having a more common origin in Waldeyer’s ring (nasopharynx, tonsil, base of the tongue).27,28 They present as a “bump” most often on the hard palate, gingiva, and buccal mucosa. In the upper and lower alveolus, the tumor develops in the odontogenic tissue attacking the tooth sockets. The teeth become displaced, mobile, and then dislodged and embedded in the soft tissue of the tumor. This would occur with a rapidly growing tumor. A more slowly growing lymphoma would present to the dentist with a painful tooth, for which he would do a root canal or extraction. The pain would persist or recur and a diagnosis of osteomyelitis would initiate a course of antibiotics. The pain would get worse in association with “angry” soft tissue about the tooth or tooth socket, which would be biopsied. The diagnosis may be missed or delayed for additional biopsies. These nonHodgkin’s lymphomas of the alveolus and oral mucosa are most often widespread (at least 50%) as demonstrated by staging or by metastases after treatment with radiation alone.
(a)
Kaposi Sarcoma
Kaposi sarcoma of the oral mucosa, most often the hard palate, consists of red/purple macules, plaques, and nodules. Although seen in acquired immunodeficiency syndrome (AIDS), renal transplant patients on immunosuppression also suffer from Kaposi sarcoma and lymphoma. Biopsy reveals spindle cells and slits, which are vascular spaces.
Diagnosis As for the other lesions of the oral cavity, histology is critically important. In the instance of non-Hodgkin’s lymphoma, the principles of history taking, physical examination, and diagnostic workup are similar to those for other sites (see also Chapter 49, Rare Lymphomas). History and examination may reveal the extent of disease and other sites of involvement. Examination of the blood may indicate a raised uric acid level, LDH, presence of paraproteins, and/or impairment of renal function, the latter suggesting the possibility of more disseminated disease with infradiaphragmatic compromise of the ureters, an extremely uncommon situation with this presentation. Imaging studies should include a CT scan of the chest and abdomen and either a gallium scan or PET scan, depending on available resources. Depending on the histological subtype of lymphoma, imaging of the brain and meninges may also be indicated. In the situation of a histological diagnosis of plasmacytoma, a monoclonal gammopathy may be present (see Figures 11a and b). Bone marrow aspiration and biopsy should show less than 10% plasma cells. Imaging studies should not reveal osteolytic lesions; in the case of a positive
(b) Figure 11 Extramedullary plasmacytoma – (a) Sheets of neoplastic plasma cells. (b) κ Light chain restriction of the neoplastic plasma cells.
bone scan, it is likely that the plasmacytoma has, in fact, got converted into multiple myeloma.
Management As outlined in Chapter 49, Rare Lymphomas, atypical lymphomas are managed along similar lines of all lymphomas, with the caveat that Kaposi sarcoma and HIVrelated lymphoma require special consideration because of the associated immunodeficiency. In general terms, treatment is predicated on a rational combination of chemotherapy and radiotherapy to affected nodes, although in some indolent lymphomas, radiotherapy alone is used (particularly in elderly patients). Even for stage IAE non-Hodgkin’s lymphoma, chemotherapy followed by irradiation is the most common practice. Most commonly cyclophosphamide, vincristine, doxorubicin, and prednisone (CHOP) for six cycles are used for intermediate or high-grade non-Hodgkin’s lymphomas,29,30 more recently augmented by radioimmunotherapy.
UNCOMMON TUMORS OF THE ORAL CAVITY AND ADJACENT STRUCTURES
The present drug treatment of AIDS has decreased the frequency of incidence of Kaposi sarcoma.31 Treatment of symptomatic Kaposi sarcoma includes intralesional therapy with vinblastine for localized disease.32,33 For more extensive disease, treatment with α-interferon, especially when the CD4 count is greater than 200 uL−1 , or chemotherapy using combinations of doxorubicin, bleomycin, vincristine, or vinblastine should be considered.34 For true plasmacytomas of the oral cavity, surgical excision may be all that is necessary. If the margins are positive, radiation therapy should be added. If because of tumor size a radical type of oncologic resection is proposed, radiation therapy is preferred over surgery.35,36
SOFT TISSUE SARCOMAS Presentation Soft tissue sarcomas often present as painless, enlarging masses, distorting the facial contour. On inspection, some can be seen to originate from the oral cavity. Sarcomas of the tongue are diagnosed before they deform the facial contour because of alteration in speech or bleeding. Amyloid, causing macroglossia, can simulate a sarcoma. Sarcomas and carcinomas of the maxillary sinus can invade and ulcerate the upper alveolar ridge, masquerading as primary cancers of the oral cavity. Cancers of the parapharyngeal space, including the deep lobe of the parotid, can also bulge into the retromolar, anterior tonsillar pillar or buccal region, simulating a primary oral cavity cancer, the true size and origin revealed by contrast-enhanced CT or MRI. Soft tissue sarcomas of the oral cavity, except for hemangiopericytoma, have a poorer outlook than their counterparts in the extremities. Soft tissue sarcomas can infiltrate considerable distances compared to those of bone, which have a more rounded edge when growing outwards from bone to penetrate the soft tissue.
97
example, the poor survival of angiosarcoma in the extremities, as well as, generally, in the head and neck is apparently not so when it occurs in the oral cavity (tongue, lip, hard and soft palates).37 Leiomyosarcoma
Leiomyosarcoma is an aggressive cancer that presents mostly as a painful mass in the upper and lower jaws, but can also be found in the tongue, floor of the mouth, cheek, and hard and soft palates. Its circumscribed appearance is deceptive, as it is highly infiltrative38 (see Figures 12a and b). Fiveyear survival is 23% with no evidence of disease, 8% are alive with disease, and 69% are dead from disease. The local recurrence is nearly 40%, distant metastases 40%, and cervical metastases 15%. Liposarcomas
Liposarcomas of the oral cavity have a 5-year survival of 50%.39,40 Grading of this sarcoma is prognostic with low
(a)
Hemangiopericytomas
Hemangiopericytomas occur at the peripheral boundaries of the oral cavity. Because of their slow growth, they are often painless. Microscopically, they appear as tightly packed spindle cells between blood vessels with normal endothelial cells. The spindle cells are the malignant pericytes. The stain for factor VIII is positive. The local recurrence rate is high (50%) but the death rate from this neoplasm is below 10%. The presence of negative surgical margins is important. This tumor invades local structures, which can be shown by a gadolinium MRI study. Hemangiomas in the deep lobe of the parotid or masseter muscle can bulge into the oral cavity. CT or MRI will show that the site of origin is outside the oral cavity. In a large series of soft tissue sarcomas arising predominantly in the extremities, histology proved to be an important prognosticator of survival, the highest being dermatofibroma protuberans (100%) and the lowest being rhabdomyosarcoma and angiosarcoma (under 45%). MFH and fibrosarcoma were in the middle (65%). The rarity of these tumors in the oral cavity allows series with relatively few patient outcomes. For
(b) Figure 12 Leiomyosarcoma – (a) High-grade pleomorphic sarcoma in which smooth muscle differentiation is difficult to discern. (b) Smooth muscle actin (SMA) is useful in recognizing smooth muscle differentiation.
98
HEAD AND NECK CANCER
cavity for one. Of the six patients with follow-up (median duration 6.5 years), three have had local recurrence and one has metastasized.43 Hamakawa described an intraosseous EHE of the mandible.44 The most common oral site is the gingiva, presenting with bone erosion. The pattern of solid growth and the epithelioid appearance of the endothelium frequently lead to the mistaken diagnosis of metastatic carcinoma.42 The presence of CD31 and the absence of keratin immunostains distinguish this cancer from carcinoma. Histopathologic features that may be associated with aggressive clinical behavior include significant cellular atypia, and one or more mitosis per 10 hpf (high-power fields).45 Fibrosarcoma
Figure 13 Myxoid liposarcoma – Fine arborizing vascular pattern, small uniform cells, and myxoid stroma.
Figure 14 Pleomorphic liposarcoma – Atypical lipoblasts with vacuolated cytoplasm.
grade and myxoid (see Figure 13) having a high survival, and pleomorphic (see Figure 14) and, especially, round cell a low survival. Paradoxically, neither grade nor size was found to be prognostic in liposarcoma of the oral cavity in one study; whereas, opposite conclusions were reported in another study. Infiltrating lipomas with atypical cells may be diagnosed as liposarcoma. The pathologist time spent on making the exact cellular diagnosis of a soft tissue sarcoma rarely impacts treatment. Epithelioid Hemangioendothelioma
Epithelioid hemangioendothelioma (EHE) has two counterparts, epithelioid angiosarcoma and epithelioid hemangioma.41 Only 26 of 41 cases appear to arise from a vein. EHE often has an indolent course and can demonstrate multiple local recurrences. In the original series reported by Weiss and Enzinger, 6 of 31 patients developed metastases.42 Of the eight patients described by Nayler, EHE was in the oral
Fibrosarcoma is one of the malignant spindle cell tumors and must be distinguished from synovial sarcoma, liposarcoma, MFH, and sarcomas of the nerve sheath. The more undifferentiated the sarcoma, the more difficult this becomes.46 In a series of 32 sarcomas of the oral and maxillofacial region, three were fibrosarcomas.47 Most arose spontaneously. Those arising in areas of prior irradiation have a latency period of 10 years, and those arising in old burn scars, 30 years. A mass in the oral cavity may have deglutition complaints, whereas in the neck, fibrosarcoma presents as a painless enlarging mass. Fibrosarcoma of bone can arise within the mandible or skull bone. Fifteen to 30% of all osseous fibrosarcomas are associated with benign lesions such as fibrous dysplasia, bone infarcts, cysts, and Paget’s disease. Among 214 sarcoma patients from the head and neck sarcoma registry, the fibrosarcoma patients had a 60% survival at 5 years; 60% had local recurrence within 2 years of resection.48 Local failure was the most common cause of death. Thirty percent of patients developed distant metastases, most commonly to the lungs. Regional lymph adenopathy was found in 3%. Poor prognostic factors are as follows: Paget’s disease, medullary (vs periosteal) type primary, >5 cm, high grade, and lymph node metastases.49 Malignant Fibrous Histiocytoma
MFH is the most common soft tissue sarcoma; it is characterized by the presence of spindle cells in a storiform pattern, myxoid stroma, and myofibroblasts (see Figure 15); and it can be confused with fibrosarcoma, liposarcoma, and a pleomorphic rhabdomyosarcoma. It has a high recurrence rate and a high rate of distant metastases. Oral MFH has been reported to arise within prior radiation fields, developing 2.5–11 years after radiotherapy. Doses ranging from 24 to 80 Gy have been reported.50 The presentation was a tender, reddish, rubbery, lobulated mass with surface ulceration. The mean survival time after diagnosis was 17 months. MFH has been reported to have a variable natural history, with different series recording small-to-large percentages of cases with involved cervical nodes.51 Rhabdomyosarcoma
Embryonal rhabdomyosarcoma occurs commonly in the head and neck in children. It is composed of small round “blue”
UNCOMMON TUMORS OF THE ORAL CAVITY AND ADJACENT STRUCTURES
99
(a) Figure 15 Malignant fibrous histiocytoma – Storiform pattern is often focally present in this cellular sarcoma.
cells (see Figures 16a and b). Leukemia and lymphoma in the soft tissues of the oral cavity should be ruled out. It responds well to chemotherapy and radiation. It can occur in the orbit causing proptosis, in the nasopharynx mimicking adenoids, and as a polyp in the external ear canal initiating serosanguinous drainage and otitis media. A botryoid type simulates a nasal polyp. Access to the meninges may be through the sites of origin, such as nasopharynx and middle ear, causing cranial nerve palsies and invasion of the brain stem, which can be fatal. Five-year survival for orbital rhabdomyosarcoma is 90% and that of parameningeal sites (nasopharynx, middle ear/mastoid, nasal and paranasal sinuses) 70%. Five-year survival for local disease in children is 90%, and 30% if the tumor has metastasized. Survival after relapse (commonly distant metastases) is only 17% at 2 years.46
Diagnosis As in other tumors of the head and neck region, the diagnostic approach to sarcomas follows a pretty standard protocol. Careful history; chronicling the time course of the tumor and possible risk factors; followed by detailed physical examination with emphasis on the features of the primary tumor, cervical, and submental/submandibular lymph nodes; as well as endoscopy of the oropharynx and nasopharynx are very important. There are no specific serological tests that are helpful in the diagnosis of soft tissue sarcoma, although the clear demonstration of some epithelial tumor markers may exclude the diagnosis of sarcoma. Immunohistochemical stains of biopsy specimens may be helpful in characterizing the type of sarcoma, and testing for CD117 (c-KIT) may suggest potential utility of imantinib mersylate (“Gleevec”).
Management To ensure success, the surgical margins must be clear. In the oral cavity, this means that the sarcoma must be small (under 5 cm) and must be of low or intermediate grade to ensure that soft tissue invasion peripheral to the mass is limited.
(b) Figure 16 Embryonal rhabdomyosarcoma – (a) Small “blue” round cells are typical of this type. (arrow shows mitosis). (b) MyoD1 stain is helpful in distinguishing this sarcoma from other small “blue” round cell tumors.
The resection of the larger lesions needs to be planned carefully and may include sacrificing structures essential to either cosmesis or function. Clearance should be measured and stated in the pathology report, for example, 1–3 mm clearance. If multiple margins are positive or if gross tumor is left behind, postoperative irradiation will not give matching results to those achieved with negative margins.47,52 – 54 An oncologic surgical procedure that results in negative margins is most important. Radiation therapy is given to patients with high-grade and narrow and/or positive surgical margins. It is impossible in this venue to determine its worth. A report from a large series of extremity sarcomas demonstrates a 10% decrease in local recurrence without corresponding increase in survival. The role of chemotherapy for sarcoma is primarily reserved for cases of locally advanced recurrent disease and for metastatic disease.54 In these cases, standard regimens would include agents such as ifosfamide and doxorubicin.
100
HEAD AND NECK CANCER
In some cases of high-grade tumor, adjuvant therapy can be considered albeit definitive proof of the effectiveness is not available.55 Neoadjuvant chemotherapy can be considered, especially, in locally advanced tumors in an attempt to improve resectability and to test for chemoresponsiveness as a strategy for selecting adjuvant therapy.56 Complete wide resection to prevent local recurrence with a lymph node dissection in the presence of clinical or radiographic adenopathy is recommended for EHE. Two of 14 cases in a recent report developed regional lymph node metastases.45 Weiss and Enzinger42 described 6 of 31 patients with metastases, 1 with regional lymphadenopathy, and 5 with distant metastases. For fibrosarcomas, complete surgical excision with ample margins is the treatment of choice. Irradiation is reserved for close or positive margins or for high-grade lesions, but proof of efficacy is lacking. If submandibular disease or neck metastases are present, then neck dissection is warranted. At the MD Anderson Cancer Center, 799 patients who underwent postoperative radiotherapy for soft tissue sarcoma (mostly extremities) were retrospectively reviewed. Factors associated with an inferior local control were positive resection margins; treatment of recurrent disease; primary location in the head and neck or deep trunk; histopathology of MFH, neurogenic sarcoma, or epithelioid sarcoma; and postoperative radiation dose <64 Gy. A meta-analysis of 14 adjuvant chemotherapy trials showed improvement in recurrence-free interval, though survival advantage was not statistically significant. Death was mostly due to lung metastases.57 Because of the effectiveness of chemotherapy and radiation, an initial oncologic resection in children is recommended only if organ function can be preserved.58 Daya discusses reoperation and second look procedures. Lymph node metastases predict local failure.59 More than two metastatic sites markedly reduce survival.60 In adults, rhabdomyosarcoma is much different than that found in children. Its presence in the oral cavity is very rare. There is always some difficulty in making the diagnosis. Liposarcoma and MFH are always considered, as skeletal muscle may be difficult to demonstrate in undifferentiated cancers, despite the availability of many immunohistochemical stains. Since radiation and chemotherapy are much less effective in pleomorphic rhabdomyosarcoma, which is the common type found in adults, the distinction from embryonal rhabdomyosarcoma is important and leads to the recommendation of wide negative margins even at the risk of ablating organ function. The finding of this cancer in the head and neck in adults is in the ethmoid sinus; an oral cavity presentation is so rare as to be questioned.
SUMMARY Uncommon cancers of the oral cavity are a diverse group of tumors, representing a spectrum from epithelial to mesenchymal origin. They are characterized by an absence of the usual epidemiological and demographic features of the more common squamous malignancies. For the purpose of management, their occurrence is sufficiently rare as to make
definitive management recommendations difficult. We have thus attempted to correlate the experience from the literature with our own patterns of practice to develop algorithms that may be helpful to the clinician faced with one of these complex problems of diagnosis and therapy.
REFERENCES 1. Robbins KT, Yao SF. Uncommon epithelial tumors of the oral cavity. In Raghavan D, et al. (eds) Textbook of Uncommon Cancer, 2nd ed. Chichester, England: Wiley-Liss, 1998: 117 – 130. 2. Taylor AM, et al. Malignant odontogenic tumors. A retrospective and collaborative study of seven cases. Med Oral 2003; 8: 110 – 21. 3. de Lange J, van den Akker HP, Klip H. Incidence and disease-free survival after surgical therapy of central giant cell granulomas of the jaw in The Netherlands: 1990 – 1995. Head Neck 2004; 26(9): 792 – 5. 4. Messahel B, et al. Clinical features of molecular pathology of solid tumours in childhood. Lancet Oncol 2005; 6(6): 421 – 30. 5. Grier HE, et al. Addition of ifosfamide and etoposide to standard chemotherapy for Ewing’s sarcoma and primitive neuroectodermal tumor of bone. N Engl J Med 2003; 348(8): 694 – 701. 6. Kilpatrick SE, et al. Langerhans’ cell histiocytosis (histiocytosis X) of bone. A clinicopathologic analysis of 263 pediatric and adult cases. Cancer 1995; 76(12): 2471 – 84. 7. Reichart PA, Philipsen HP, Sonner S. Ameloblastoma: biological profile of 3677 cases. Eur J Cancer B Oral Oncol 1995; 31 B: 86 – 99. 8. Quesada JL, et al. Ewings’ sarcoma of the mandible. J Laryngol Otol 2003; 117(9): 736 – 8. 9. Wexler LH, et al. Combined modality treatment of Ewing’s sarcoma of the maxilla. Head Neck 2003; 25(2): 168 – 72. 10. Grier HE, et al. Addition of ifosfamide and etoposide to standard chemotherapy for Ewing’s sarcoma and primitive neuroectodermal tumor of bone. N Engl J Med 2003; 348(8): 694 – 701. 11. Smith RB, et al. National Cancer Data Base report on osteosarcoma of the head and neck. Cancer 2003; 98(8): 1670 – 80. 12. Tabone MD, et al. Outcome of radiation-related osteosarcoma after treatment of childhood and adolescent cancer: a study of 23 cases. J Clin Oncol 1999; 17(9): 2789 – 95. 13. Ruark DS, Schlehaider UK, Shah JP. Chondrosarcomas of the head and neck. World J Surg 1992; 16(5): 1010 – 5. 14. Prasad ML, Patel SG, Busam KJ. Primary mucosal desmoplastic melanoma of the head and neck. Head Neck 2004; 26(4): 373 – 7. 15. Terhaard CH, et al. Salivary gland carcinoma: independent prognostic factors for locoregional control, distant metastases, and overall survival: results of the Dutch head and neck oncology cooperative group. Head Neck 2004; 26(8): 681 – 92. 16. Bradley PJ. Distant metastases from salivary glands cancer. ORL J Otorhinolaryngol Relat Spec 2001; 63(4): 233 – 42. 17. Auclair PL, Goode RK, Ellis GL. Mucoepidermoid carcinoma of intraoral salivary glands. Evaluation and application of grading criteria in 143 cases. Cancer 1992; 69(8): 2021 – 30. 18. Lopes MA, et al. A clinicopathologic study of 196 intraoral minor salivary gland tumours. J Oral Pathol Med 1999; 28(6): 264 – 7. 19. Kuhel W, et al. Adenoid cystic carcinoma of the palate. Arch Otolaryngol Head Neck Surg 1992; 118(3): 243 – 7. 20. Batsakis JG, el Naggar AK. Terminal duct adenocarcinomas of salivary tissues. Ann Otol Rhinol Laryngol 1991; 100(3): 251 – 3. 21. Evans HL, Batsakis JG. Polymorphous low-grade adenocarcinoma of minor salivary glands. A study of 14 cases of a distinctive neoplasm. Cancer 1984; 53(4): 935 – 42. 22. Spiro RH. Salivary neoplasms: overview of a 35-year experience with 2,807 patients. Head Neck Surg 1986; 8(3): 177 – 84. 23. Spiro RH, Huvos AG. Stage means more than grade in adenoid cystic carcinoma. Am J Surg 1992; 164(6): 623 – 8. 24. Bell RB, et al. Management and outcome of patients with malignant salivary gland tumors. J Oral Maxillofac Surg 2005; 63(7): 917 – 28. 25. Agulnik M, Siu LL. An update on the systemic therapy of malignant salivary gland cancers: role of chemotherapy and molecular targeted agents. Curr Med Chem Anti-Cancer Agents 2004; 4(6): 543 – 51.
UNCOMMON TUMORS OF THE ORAL CAVITY AND ADJACENT STRUCTURES 26. Beckhardt RN, et al. Minor salivary gland tumors of the palate: clinical and pathologic correlates of outcome. Laryngoscope 1995; 105(11): 1155 – 60. 27. Sunaba K, et al. Radiotherapy for primary localized (stage I and II) non-Hodgkin’s lymphoma of the oral cavity. Int J Radiat Oncol Biol Phys 2000; 47(1): 179 – 83. 28. Takahashi H, et al. Primary extranodal non-Hodgkin’s malignant lymphoma of the oral region: analysis of 11 autopsy cases. J Oral Pathol 1987; 16(5): 241 – 50. 29. Gurkaynak M, et al. Waldeyer’s ring lymphomas: treatment results and prognostic factors. Am J Clin Oncol 2003; 26(5): 437 – 40. 30. Mohammadianpanah M, et al. Treatment results of tonsillar lymphoma: a 10-year experience. Ann Hematol 2005; 84(4): 223 – 6. 31. Cattelan AM, et al. Acquired immunodeficiency syndrome-related Kaposi’s sarcoma regression after highly active antiretroviral therapy: biologic correlates of clinical outcome. J Natl Cancer Inst Monogr 2001; 28: 44 – 9. 32. Epstein JB, et al. Oral Kaposi’s sarcoma in acquired immunodeficiency syndrome. Review of management and report of the efficacy of intralesional vinblastine. Cancer 1989; 64(12): 2424 – 30. 33. McCormick SU. Intralesional vinblastine injections for the treatment of oral Kaposi’s sarcoma: report of 10 patients with 2-year follow-up. J Oral Maxillofac Surg 1996; 54(5): 583 – 7. 34. Morris AK, Valley AW. Overview of the management of AIDS-related Kaposi’s sarcoma. Ann Pharmacother 1996; 30(10): 1150 – 63. 35. Soutar R, et al. Guidelines on the diagnosis and management of solitary plasmacytoma of bone and solitary extramedullary plasmacytoma. Clin Oncol (R Coll Radiol) 2004; 16(6): 405 – 13. 36. Alexiou C, et al. Extramedullary plasmacytoma: tumor occurrence and therapeutic concepts. Cancer 1999; 85(11): 2305 – 14. 37. Fanburg-Smith JC, Furlong MA, Childers EL. Oral and salivary gland angiosarcoma: a clinicopathologic study of 29 cases. Mod Pathol 2003; 16(3): 263 – 71. 38. Schenberg ME, Slootweg PJ, Koole R. Leiomyosarcomas of the oral cavity. Report of four cases and review of the literature. J Craniomaxillofac Surg 1993; 21(8): 342 – 7. 39. Golledge J, Fisher C, Rhys-Evans PH. Head and neck liposarcoma. Cancer 1995; 76(6): 1051 – 8. 40. Fanburg-Smith JC, Furlong MA, Childers EL. Liposarcoma of the oral and salivary gland region: a clinicopathologic study of 18 cases with emphasis on specific sites, morphologic subtypes, and clinical outcome. Mod Pathol 2002; 15(10): 1020 – 31. 41. Weiss SW, et al. Epithelioid hemangioendothelioma and related lesions. Semin Diagn Pathol 1986; 3(4): 259 – 87. 42. Weiss SW, Enzinger FM. Epithelioid hemangioendothelioma: a vascular tumor often mistaken for a carcinoma. Cancer 1982; 50(5): 970 – 81.
101
43. Nayler SJ, et al. Composite hemangioendothelioma: a complex, lowgrade vascular lesion mimicking angiosarcoma. Am J Surg Pathol 2000; 24(3): 352 – 61. 44. Hamakawa H, et al. Intraosseous epithelioid hemangioendothelioma of the mandible: a case report with an immunohistochemical study. J Oral Pathol Med 1999; 28(5): 233 – 7. 45. Flaitz CM, et al. Primary intraoral epithelioid hemangioendothelioma presenting in childhood: review of the literature and case report. Ultrastruct Pathol 1995; 19(4): 275 – 9. 46. Barnes L, Kanbour A. Malignant fibrous histiocytoma of the head and neck. A report of 12 cases. Arch Otolaryngol Head Neck Surg 1988; 114(10): 1149 – 56. 47. Yamaguchi S, et al. Sarcomas of the oral and maxillofacial region: a review of 32 cases in 25 years. Clin Oral Investig 2004; 8(2): 52 – 5. 48. Wanebo HJ et al., Society of Head and Neck Surgeons Committee on Research. Head and neck sarcoma: report of the Head and Neck Sarcoma Registry. Head Neck 1992; 14(1): 1 – 7. 49. Bras J, Batsakis JG, Luna MA. Malignant fibrous histiocytoma of the oral soft tissues. Oral Surg Oral Med Oral Pathol 1987; 64(1): 57 – 67. 50. Wiesmiller K, Barth TF, Gronau S. Early radiation-induced malignant fibrous histiocytoma of the oral cavity. J Laryngol Otol 2003; 117(3): 224 – 6. 51. Barnes L. Surgical Pathology of the Head and Neck, 2nd ed. New York: Marcel Decker, 2001. 52. Freedman AM, Reiman HM, Woods JE. Soft-tissue sarcomas of the head and neck. Am J Surg 1989; 158(4): 367 – 72. 53. Chen SA, et al. Adult head and neck soft tissue sarcomas. Am J Clin Oncol 2005; 28(3): 259 – 63. 54. Tran LM, et al. Sarcomas of the head and neck. Prognostic factors and treatment strategies. Cancer 1992; 70(1): 169 – 77. 55. Scurr M, Judson I. Neoadjuvant and adjuvant therapy for extremity soft tissue sarcomas. Hematol Oncol Clin North Am 2005; 19(3): 489 – 500. vi. 56. Potter BO, Sturgis EM. Sarcomas of the head and neck. Surg Oncol Clin N Am 2003; 12(2): 379 – 417. 57. Sturgis EM, Potter BO. Sarcomas of the head and neck region. Curr Opin Oncol 2003; 15(3): 239 – 52. 58. Andrassy RJ. Advances in the surgical management of sarcomas in children. Am J Surg 2002; 184(6): 484 – 91. 59. Wharam MD, et al. Failure pattern and factors predictive of local failure in rhabdomyosarcoma: a report of group III patients on the third Intergroup Rhabdomyosarcoma Study. J Clin Oncol 2004; 22(10): 1902 – 8. 60. Breneman JC, et al. Prognostic factors and clinical outcomes in children and adolescents with metastatic rhabdomyosarcoma – a report from the Intergroup Rhabdomyosarcoma Study IV. J Clin Oncol 2003; 21(1): 78 – 84.
Section 2 : Head and Neck Cancer
7
Rare Tumors of the Larynx Samir S. Khariwala and Marshall Strome
INTRODUCTION The larynx is a specialized, multifunctional organ. The evolutionary origin of the larynx was as a sphincter, which prevented the entrance of water into the airway of the lungfish.1 Its most primitive function, therefore, was to protect the airway. Over time, dilator muscles evolved, which afforded active opening and closing of the airway. This important evolution enabled the larynx to perform additional functions. These included phonation, coughing, and the generation of the Valsalva maneuver. The larynx also has indirect roles in deglutition, coughing, and the gag reflex related to its afferent innervation. As a result, laryngeal neoplasms are frequently associated with alterations in voice, aspiration, and airway obstruction. The most common malignancy of the larynx is squamous cell carcinoma (SCC). The treatment of SCC of the larynx has a solid foundation in surgery but newer chemotherapeutic and radiation therapy protocols are also rapidly gaining favor. In addition to epidermoid derivatives, the larynx contains glandular and mesenchymal tissue. Less common malignancies can develop from any of these tissue components. Consequently, several non-SCC neoplasms have been described in the larynx. Each lesion has unique characteristics. Often, these characteristics relate to their location in such a highly specialized organ. Also, many malignancies behave differently in the larynx when compared to other anatomic sites. This chapter aims to characterize uncommon malignancies that occur in the larynx, discuss current therapeutic considerations, and provide prognostic information. In considering non-SCC neoplasms of the larynx, several exceedingly infrequent tumors have been described such that 10 or fewer cases have been reported. In an effort to provide a thorough review of the lesions discussed in the following text, we will focus on those tumors that are rare, yet not “reportable”.
ANATOMY An understanding of laryngeal anatomy is vital when considering laryngeal malignancies, as it is the anatomy that governs the pattern of spread, recurrence, and thus management decisions. The superior border of the larynx is the tip
of the epiglottis. The lower border is the inferior edge of the cricoid cartilage. The larynx is bordered posteriorly and laterally by the hypopharynx. The strap muscles lie anterior to the thyroid and cricoid cartilages. The larynx is divided into three subsites based on embryologic development. The supraglottis extends from the tip of the epiglottis to the apex of the ventricle. Structures that lie in the supraglottis include the epiglottis, aryepiglottic folds, false vocal cords, and portions of the arytenoid cartilages. The glottis includes the area from the apex of the ventricle to a point 1 cm inferior to the apex. The main structures that comprise the glottis are the true vocal cords. Lastly, the subglottis is bordered superiorly by the glottis and extends to the inferior border of the cricoid cartilage. These subdivisions each drain to different nodal basins.
Laryngeal Skeleton Each subdivision of the larynx receives support from portions of the laryngeal skeleton (see Figure 1). The hyoid bone supports the larynx and stabilizes the hypopharynx. Its two ends project posteriorly via the greater cornua. The thyrohyoid membrane forms an attachment between the hyoid bone and the thyroid cartilage. The thyroid cartilage is composed of two halves, which meet in the midline to form the laryngeal prominence that can be seen externally. The superior cornu meets the thyrohyoid membrane and the inferior cornu articulates with the cricoid cartilage. Lastly, the cricoid cartilage provides support for the subglottis. The cricoid cartilage is the only complete ring in the upper airway. It therefore provides circumferential rigid support for the airway. The conus elasticus and the quadrangular membrane are fibroelastic membranes that are important components of the larynx (see Figures 2, 3). The conus elasticus attaches laterally to the cricoid cartilage and projects to the edges of the thyroid cartilage and the vocal process of the arytenoid cartilage. The quadrangular membrane connects the epiglottis to the arytenoid cartilages. The superior edge of this membrane is contained within the aryepiglottic fold while the inferior edge forms part of the false vocal fold. These membranes are clinically relevant as they form the
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
RARE TUMORS OF THE LARYNX
103
Hyoid Thyrohyoid membrane
Thyroid cartilage
Cricothyroid membrane Cricoid cartilage Tracheal ring
Figure 1 External laryngeal skeleton (Courtesy of Mark Sabo).
Epiglottis
False vocal cord Arytenoid
Conus elasticus
Ventricle
True vocal cord
Cricoid cartilage
Figure 3 Posterior view of endolaryngeal structure (Courtesy of Mark Sabo).
only abductor of the glottis. Adductors of the glottis include the lateral cricoarytenoid, thyroarytenoid, and interarytenoid muscles. The cricothyroid muscle is responsible for tensing the vocal cords. The vagus nerve supplies the larynx with both motor and sensory functions. The superior laryngeal nerve divides into an internal and external branch. The internal branch provides sensation to the supraglottis while the external branch supplies motor fibers to the cricothyroid muscle. All the other intrinsic laryngeal muscles are innervated by the recurrent laryngeal nerve. In addition, the recurrent laryngeal nerve provides sensation to the glottis and subglottis.2 The mucosa of the upper airway is primarily composed of respiratory epithelium. The exception to this is the surface of the true vocal cords, which are lined by squamous epithelium.
TUMORS OF EPITHELIAL ORIGIN Verrucous Carcinoma
Tracheal ring
Figure 2 Midsagittal section of larynx (Courtesy of Mark Sabo).
medial and inferomedial borders of the paraglottic space. As such, they play a role in preventing the spread of laryngeal malignancies.
Musculature, Innervation, and Histology The intrinsic muscles of the larynx primarily control movement of the vocal cords. The posterior cricoarytenoid is the
Biology and Epidemiology
Verrucous carcinoma was first described in 1948 by Lauren V. Ackerman, based on a series of patients with carcinoma of the oral cavity.3 In the past, it has often been termed “Ackerman’s tumor.” The term verrucous carcinoma is meant to describe the gross appearance of an exophytic lesion that is heaped above the epithelial surface with a papillary micronodular appearance.4 It is a variant of SCC with specific clinical and morphologic features and is felt to exist within the continuum between benign squamous hyperplasia and frank SCC. Verrucous carcinoma makes up only 1–2% of laryngeal carcinomas. The larynx is the second-most common site of occurrence in the head and neck after the oral cavity.5,6
104
HEAD AND NECK CANCER
The exact etiology of verrucous carcinoma is unclear. Most verrucous lesions are associated with adjacent mucosal abnormalities. Also, frank SCC can develop from highly proliferative verrucous lesions. These observations suggest that verrucous carcinomas may develop from a benign precursor.7 A clear association has been established between the use of tobacco products and verrucous carcinoma.8,9 In addition, many have theorized a role for human papilloma virus (HPV) in its development. This association is, however, far from established. While multiple studies have documented the presence of HPV in verrucous carcinomas, several studies have also found no trace of HPV in these tumors.10,11,12 Therefore, the role played by HPV is as yet unclear. Pathology
Grossly, verrucous carcinoma appears as a fungating, papillomatous, shaggy, grayish-white neoplasm (see Figure 4). Under microscopic magnification, verrucous carcinoma appears as a well-differentiated SCC with minimal cytologic atypia. The surface of the lesion is densely keratinized with a well-circumscribed deep margin. The borders are often described as “pushing”, rather than infiltrative. The borders are usually surrounded by an inflammatory infiltrate. Mitoses are rare and the usual cytologic criteria of malignancy are lacking.5,9,12 Occasionally, tumors may contain elements of both verrucous carcinoma and SCC. Misdiagnosis may occur in these instances when biopsy samples do not contain both elements. It is often only after recurrence that some tumors are correctly identified as containing SCC. These tumors are known as hybrid tumors and should be treated as SCCs.
Clinical Presentation and Diagnostic Considerations
In his initial report, Ackerman described verrucous carcinoma as “a variety of squamous cell carcinoma whose behavior is unique and which has a typical clinical course with characteristic gross and microscopic findings. We have designated this type of carcinoma as verrucous carcinoma and feel that it should be separated from other epithelial carcinomas for, even when extensive, with proper treatment, the prognosis is excellent.” Ackerman’s initial assessment of this tumor’s behavior has proven accurate. Hoarseness is the most common presenting symptom. Pain and dysphagia may be present but are far less common. Vocal cord paralysis is rare but can occur in cases of large tumors.13 Tumors most commonly present as stage I lesions.6 Despite occasional local destruction, disease is locally contained in greater than 90% of cases.6 In general, verrucous tumors have a very low rate of regional and distant metastases. Mortality related to this tumor is also rare and most commonly associated with anaplastic transformation.13,14 Behavior and Treatment
Treatment for most verrucous tumors is primary surgery. Radiotherapy can be added for larger tumors. Transoral resection of T1 and T2 tumors using the CO2 laser is sufficient in nearly all cases. Use of endoscopic laser surgery is appropriate for this tumor, as it is known to be less aggressive than SCC. Use of endoscopic surgery allows removal of the tumor with less morbidity while preserving the ability to treat recurrent disease if necessary.15 Regional and distant spread of verrucous carcinoma is exceedingly rare.13,14 In the past, there has been a concern that radiotherapy leads to a high rate of anaplastic transformation.16 Other studies have disputed this theory.13,17 Regardless, treatment with radiation alone appears less efficacious when compared to surgery. Five-year survival rates for surgical treatment of laryngeal verrucous carcinoma were found to be 94 versus 66% 5-year survival in patients treated with radiation alone.6 Chemotherapy has no defined role for this lesion, and in the rare situation where extensive or recurrent spread requires consideration of chemotherapy, referral to a specialist center is indicated. Overall 5-year survival for laryngeal verrucous lesions (regardless of treatment) has been reported to be 86.9%.
Spindle Cell (Sarcomatoid) Carcinoma Biology and Epidemiology
Figure 4 Verrucous carcinoma of the larynx, H&E, 50× (Courtesy of Richard Prayson, MD).
Spindle cell (sarcomatoid) carcinomas of the larynx are uncommon tumors (see Figure 5). They make up 1.3–2.7% of tumors at this site.18,19 Owing to the fact that this tumor is difficult to diagnose, there are several terms used in the literature to describe them. These include: carcinosarcoma, pseudosarcoma, pseudocarcinoma, pseudocarcinosarcoma, pseudosarcomatous carcinoma, spindle cell carcinoma, spindle cell variant of squamous carcinoma, SCC with pseudosarcoma, pleomorphic carcinoma, metaplastic carcinoma, and polypoid SCC. The tumor is composed of a spindle component and an epithelial component. Many
RARE TUMORS OF THE LARYNX
105
variable immunostaining lead to difficulties in making this diagnosis. Clinical Presentation and Diagnostic Considerations
Figure 5 Spindle cell carcinoma of the larynx, H&E, 200× (Courtesy of Richard Prayson, MD).
theories have been suggested regarding the histogenesis of this tumor. These include the following possibilities: the existence of two synchronous tumors, a carcinoma and a sarcoma from nearby sites (collision tumor); the origin of the tumor from an undifferentiated pluripotential cell which differentiates toward both squamous epithelium and stroma (carcinosarcoma); the spindle cells may be a non-neoplastic reaction of the stroma to the presence of the carcinoma (pseudosarcoma), or the spindle cells may be modified malignant epithelial cells (spindle cell carcinoma).20 The latter theory is currently most widely accepted. This tumor is significantly more common in males than females with a ratio of 13 : 1. Patients report a history of tobacco use (87%) and heavy alcohol consumption (65%). This tumor usually develops in the seventh decade of life.19 Pathology
Grossly, the majority of these tumors appear polypoid and rarely sessile or ulcerated. Most of these lesions are carcinomas with spindle-shaped cells. Spindle cell (sarcomatoid) carcinoma appears as a biphasic tumor composed of two components: one which is SCC (invasive or in situ); another composed of a bland or pleomorphic spindle cell stroma. When present, the squamous portion makes the diagnosis fairly routine. Unfortunately, the SCC portion is often elusive. When both are clearly present, the sarcomatoid and carcinomatous components commonly abut each other with minimal areas of blending. The sarcomatoid portion can vary in appearance such that the tumor will imitate malignant fibrous histiocytoma, leiomyosarcoma, fibrosarcoma, or fibromatosis. Immunophenotypic analysis of these tumors produces variable results. Thompson et al. noted variable staining with smooth muscle actin (32.5%), muscle-specific actin (15.4%), S-100 (4.9%), and desmin (1.6%). Epithelial markers were also variably reactive.19 Two recent case reports have demonstrated deviation from this pattern.21,22 These studies demonstrate that the combination of a variable microscopic appearance and
Hoarseness is the most common presenting symptom when spindle cell tumors occur in the larynx.11 Dyspnea, hemoptysis, and a variety of nonspecific laryngeal symptoms may also be present. Because of the nature of this tumor, patients usually present within 1 year of symptom onset. Laryngoscopy usually reveals a polypoid, exophytic lesion with partial glottic obstruction. Palpation of the neck may localize cervical metastases. A computed tomography (CT) scan is recommended to define laryngeal extent and better define cervical pathology. The differential diagnosis for this lesion is quite challenging. Lesions that must be considered include fibrosarcoma, malignant fibrous histiocytoma, leiomyosarcoma, rhabdomyosarcoma, malignant peripheral nerve sheath tumor, osteosarcoma, and chondrosarcoma, to mention only a few. In addition to the pathologic features described above, the correct diagnosis is made after considering several factors that include age and tumor location. Behavior and Treatment
Study of the true incidence and behavior of this tumor is difficult due to frequent misdiagnosis. In general, these tumors behave similarly to SCC and are treated in the same way. Based on the available literature and the senior author’s experience, small lesions are generally best treated by conservative resection or radiation alone, whereas larger lesions are treated with partial or total laryngectomy with or without postoperative radiation.23 Overall 5-year survival irrespective of tumor-node-metastasis (TNM) classification was recently found to be 58.8%. The percentage of patients who died of disease based on staging T1–T4 were 22.5, 36.0, 56.5, and 66.7%, respectively.19 This data supports a worsening progress with larger lesions. Tumors occurring in patients with a history of radiation treatment appear to be more aggressive. With respect to metastases, these tumors should generally be approached with the squamous component in mind. That is, regional and distant metastatic potential is present thus requiring neck dissection for all but early tumors.21 Thompson noted a 19.3% overall rate of metastases of tumors at any time during the course of their treatment. This rate varied based on laryngeal subsite as follows: glottis: 12.1%, supraglottis: 35.7%, subglottis: 100%.19 The issue of chemotherapy is complex. There are no reported large case series to provide guidance in management. In general terms, our team favors the use of combination regimens that incorporate platinum analogs and/or fluoropyrimidines, when palliative therapy is required. As yet, the role of novel agents, such as gemcitabine and the taxanes, has not been proven. In some instances, repeat biopsy, to define whether a sarcomatoid outgrowth dominates, will refine the choice of chemotherapy.
106
HEAD AND NECK CANCER
Basaloid Squamous Cell Carcinoma Biology and Epidemiology
Basaloid SCC is a rare tumor first described in 1928 by Montgomery.24 Basaloid SCC of the upper aerodigestive tract was first characterized by Wain et al.25 Alternate terms existing in the literature include basosquamous carcinoma and metatypical carcinoma. The tumor is thought to arise from the pluripotential basal cell layer of the epithelium. Since the initial report, over 150 cases have been described, most of them occurring in the upper aerodigestive tract. Within the larynx, the most common subsite is the supraglottis. Wain’s report included four laryngeal tumors, all of which were epiglottic. The typical patient is an elderly male with a history of tobacco and/or alcohol abuse. Epstein-Barr virus has been detected in three cases of nasopharyngeal basaloid squamous carcinoma, but a causative role has not yet been established.26 Pathology
treatment of SCC, that is, multimodality treatment should always be considered for advanced lesions. Cervical metastases are common and neck dissection is required in the treatment of T2 or larger tumors. Survival data is difficult to gather due to the rarity of this tumor and to the disparate views regarding its nature. A study of basaloid tumors of the head and neck reported survival results in 20 patients. Although follow-up ranged from 9 months to 8 years, five patients had died of the disease and four patients were alive with the disease.26 This study, as well as others, suggest a neoplasm with aggressive characteristics. Therefore, it is safest to err on the side of aggressive management.
TUMORS DERIVED FROM SALIVARY GLANDS Biology and Epidemiology Adenoid Cystic Carcinoma
The larynx contains subepithelial minor salivary glands, which can occasionally lead to the development of sialogenic neoplasms. Adenoid cystic carcinoma is the most common malignant tumor of the minor salivary glands and there are approximately 120 cases of laryngeal adenoid cystic carcinoma reported in the literature. Males and females are equally affected. The etiology of minor salivary gland tumors is unclear but several factors have been postulated to play a role. These include: lead, asbestos, alcohol, and ionizing radiation.34,35,36,37 When occurring in the larynx, the subglottis is the most common subsite.28
Grossly, basaloid squamous carcinoma appears as an exophytic lesion, which is sometimes ulcerated. Microscopically, Wain et al. described the tumor as being composed of two portions, basaloid and squamous. The basaloid component consists of growths of small, crowded cells in a lobular arrangement, close to the mucosa. These cells have scant cytoplasm, and round to oval nuclei with small basophilic nucleoli. Small cystic spaces containing mucinous-like material are interspersed within these lobules. Mitoses are common. The squamous component is either frank invasive SCC or foci of squamous differentiation, squamous dysplasia, or carcinoma in situ.25 The tumors commonly stain for p53 with variable focal S-100 positivity. Basaloid squamous carcinoma can be confused with adenoid cystic carcinoma (particularly the solid subtype) or small cell undifferentiated (neuroendocrine) carcinoma if not carefully examined. Also, limited biopsy specimens may make correct diagnosis difficult. The identification of mitoses, nuclear pleomorphism, and necrosis point toward a diagnosis of basaloid squamous carcinoma.27
Mucoepidermoid carcinoma was first described in 1945 by Stewart et al.38 The first case involving the larynx was reported in 1963.39 These tumors are believed to develop from ductal components of the submucosal salivary glands.40 The most common sites include the floor of the laryngeal ventricle, the false vocal folds, and the anterior commissure. The true vocal cords are rarely involved.41 Males and patients aged 40–80 years are more commonly affected.42
Clinical Presentation and Diagnostic Considerations
Pathology
Patients with basaloid SCC of the larynx most often present with hoarseness. Larger tumors may also cause pain and vocal cord fixation. Examination will usually reveal an exophytic mass often partially obstructing the larynx. CT scanning is frequently useful in determining the extent of the tumor as well as the presence of cervical metastases.
Adenoid Cystic Carcinoma
Behavior and Treatment
Since its first description, basaloid squamous carcinoma has commonly been described as a high- grade or aggressive subtype of SCC with a high incidence of nodal and distant metastases at the time of presentation.28,29,30 Recently, several reports have challenged this assertion, suggesting that this tumor behaves similarly to SCC.31,32,33 Therefore, the behavior of these tumors may vary. In general, treatment should be administered based on principles that guide the
Mucoepidermoid Carcinoma
Grossly, the lesions usually appear as exophytic masses with intact mucosal coverage. Batsakis initially defined four histopathological patterns: cribiform, tubular (glandular), solid, and hyaline (cylindromatous). These patterns were refined by the creation of a grading system for adenoid cystic carcinoma. Grade I is predominantly tubular, grade II is cribiform, and grade III is solid.5,43 The usual microscopic pattern is described as cribiform with nests and columns of bland cells arranged concentrically around glandlike spaces filled with periodic acid Schiff (PAS)-positive material. True glands may also be seen (see Figure 6). Mucoepidermoid Carcinoma
Mucoepidermoid carcinomas are composed of several different cell types, which include clear cells, mucoid cells,
RARE TUMORS OF THE LARYNX
107
Clinical Presentation and Diagnostic Considerations Patients presenting with sialogenic tumors of the larynx most commonly present with dysphagia and hoarseness.45 The duration of symptoms at the time of presentation can vary from 2–3 months to 2–3 years.45 In the case of adenoid cystic carcinoma, a detailed examination of cranial sensory and motor nerves is important in determining the presence of perineural invasion. Preoperative determination of nerve involvement helps to better counsel patients regarding postoperative expectations. Laryngeal mucoepidermoid carcinoma occurs most commonly in the supraglottis where the density of minor salivary glands is highest. Patients with intermediate or high-grade mucoepidermoid carcinoma may present with cervical metastases and this should be investigated prior to determining treatment. Figure 6 Adenoid cystic carcinoma of the larynx, H&E, 100× (Courtesy of Richard Prayson, MD).
Figure 7 Mucoepidermoid carcinoma of the larynx, H&E, 200× (Courtesy of Richard Prayson, MD).
columnar cells, epidermoid cells, and intermediate cells (see Figure 7). The classification and grading of the tumor depends on the relative presence of each cell type.43 Mucoepidermoid carcinomas have been divided into three different grades: low grade, intermediate grade, and high grade. Low-grade mucoepidermoid carcinoma is composed of well-formed glandular or cystic spaces lined by a single layer of mucin-producing cells and flattened epidermoid cells. Low-grade tumors do not feature pleomorphism or mitoses. Intermediate grade tumors have a greater tendency to form solid nests of cells. They are more cellular and pleomorphic with greater numbers of intermediate cells and occasional mitoses. High-grade mucoepidermoid tumors exhibit depletion of mucoid elements. There are numerous mitoses and solid nests of intermediate or epidermoid cells.44 This lesion is often difficult to distinguish from SCC.40
Behavior and Treatment Adenoid Cystic Carcinoma
Adenoid cystic carcinomas readily demonstrate perineural invasion. The submucosal growth pattern often leads these tumors to present at a later stage. Spread to lymphatics draining the larynx is rare. Still, cervical spread has been documented and is more common in tumors with a solid pattern.45,46 This has dictated treatment protocols that emphasize local excision with neck dissection only in the presence of clinically abnormal nodes or pathologically proven nodal metastases.47 Local excision techniques include conservation surgery for small, amenable lesions. Larger lesions require total laryngectomy. Not uncommonly, adenoid cystic carcinoma demonstrates distant metastases to the lungs (most common), bone, and liver. This tumor has a long natural history such that it is often measured in decades rather than years. Local recurrence occurs almost invariably and is often easily treated.46 Still, late metastases and mortality are not uncommon. With regard to laryngeal lesions, Alavi reported 2-year and 5-year survival rates of 100 and 75%, respectively.48 In contrast, Mahlstedt reported a 5-year survival rate of 33% for laryngeal adenoid cystic carcinoma.45 These variations are likely due to small sample sizes of this rare lesion. Radiation therapy can offer increased disease control in some patients. In patients who have complete surgical resection or microscopic disease, postoperative photon therapy can provide excellent local control rates (86% at 10 years).49 Neutron beam radiation has shown increased efficacy over photon radiation for advanced disease in which there is gross residual tumor, unresectable disease, lymph node metastases at presentation, or skull base involvement. The benefit gained in these patients treated by neutron beam may outweigh the increased incidence of adverse reactions associated with this treatment.50,51 Still, the increased local control rates achieved with neutron beams have not led to increased overall survival in patients with adenoid cystic carcinoma of the head and neck. The goal of improved survival will not likely be reached until improvements in systemic chemotherapy are developed.50 Agents that have been used, without great success, include doxorubicin, cyclophosphamide, the platinum
108
HEAD AND NECK CANCER
analogs, and fluoropyrimidines, although occasionally shortterm remissions will be obtained. The role of the taxanes, gemcitabine, mitomycin C, and topoisomerase inhibitors for metastatic adenoid cystic carcinomas has not been defined. Mucoepidermoid Carcinoma
Treatment of mucoepidermoid carcinoma of the larynx follows principles similar to those that guide its treatment in other areas of the head and neck. The primary mode of treatment is surgery for most cases. Decisions regarding the use of partial laryngectomy are based on tumor size. Endoscopic use of the CO2 laser for mucoepidermoid tumor removal has been described.52 Treatment of the neck is performed in the presence of clinically positive disease. In the clinically negative neck, addressing the possibility of cervical disease is somewhat controversial. Some authors recommend dissection of firstechelon nodes in all cases,53 others recommend irradiation of patients with high-risk tumors,54 while still others recommend neck dissection for all high-grade tumors.55,56,57 Our recommendation is that patients with high-grade tumors and clinically negative necks should have neck dissections. Postoperative radiation to the primary site and neck are also recommended for high-grade tumors. Radiotherapy alone is not considered optimal treatment for this lesion regardless of tumor grade.41 Still, moderate success with radiotherapy has been achieved in some cases.40 Once again, the role of chemotherapy is ill defined, with cisplatin and 5-fluorouracil having been a suboptimal standard regimen for many years. More recently, some short-lived responses have been reported from the combination of paclitaxel and carboplatin for mucoepidermoid cancers of the salivary glands,58 and there is potential for structured trials to address this combination in the setting of laryngeal tumors of this histological type. Survival of patients with laryngeal mucoepidermoid carcinoma rests, in large part, on tumor grading. Five-year survival has been reported as 100% for low-grade tumors but only 53% at 3 years for patients with high-grade mucoepidermoid carcinoma.59
LESIONS OF MESENCHYMAL ORIGIN Chondrosarcoma Biology and Epidemiology
Travers is credited with describing the first cartilaginous tumor of the larynx in 1816. It was described as “a case of ossification and bony growth of the cartilage of the larynx.”60 For several years thereafter, all cartilaginous tumors were termed “chondroma”. The term chondrosarcoma was first used to describe a laryngeal tumor in 1935.61 In the larynx, chondrosarcoma is the most common nonepithelial tumor (see Figure 8).62,63 It makes up 0.1–1% of all laryngeal neoplasms.63,64,65 The etiology of chondrosarcoma is unclear. Their occurrence has been described following Teflon injections and radiation therapy.66,67,68,69 Smoking is not considered to be an etiologic factor.64 The occurrence of
Figure 8 Chondrosarcoma of the larynx, H&E, 100× (Courtesy of Richard Prayson, MD).
chondrosarcoma is associated with ossification of the laryngeal cartilages. Because ossification commences at sites of muscle insertion (the posterior cricoid ring and the posterior thyroid lamina), chondrosarcomas occur more commonly in these areas.63,64,66,70 The timing of cartilaginous ossification also determines the age range for this tumor. Ossification begins in the third decade of life and increases with old age.71 As a result, patients between ages 50 and 80 years are most commonly affected but tumors have been reported in patients as young as 33 and as old as 91.72,73,74 It has been suggested that the majority of chondrosarcomas are juxtaposed with benign chondromas that have been exposed to a degree of ischemia. Thompson and Gannon noted that the ischemic areas were usually found directly abutting chondrosarcoma.64 This occurred in 62% of the tumors in their study. Pathology
Grossly, both chondroma and chondrosarcoma appear as smooth, firm lesions.63,66 The distinction between chondroma and chondrosarcoma is a difficult one to make. This often comes down to personal interpretation.64 Chondromas of the laryngeal cartilages will resemble normal cartilage, but the nuclei are often larger.75,76,77 Low-grade chondrosarcomas often contain chondroma patterns within them. Therefore, a single biopsy of a cartilaginous tumor may not provide adequate tumor sampling. Microscopically, the criteria for diagnosis were set by Lichtenstein and Jaffe.78 They described cells with plump nuclei, more than one cell with two such nuclei, giant cartilage cells with large single or multiple nuclei, or cells with clumps of chromatin. Evans grouped chondrosarcoma into grades I (well-differentiated), II (moderately differentiated), and III (poorly differentiated).79 The subdivisions are based on mitotic rate, cellularity, and nuclear size. Well-differentiated lesions often show only focal areas that meet criteria for malignancy. In contrast, malignant features are present in a greater area of moderate and high-grade
RARE TUMORS OF THE LARYNX
tumors. Additional subtypes of chondrosarcoma are dedifferentiated chondrosarcoma, clear cell chondrosarcoma and myxoid chondrosarcoma. Dedifferentiated chondrosarcoma is also known as chondrosarcoma with additional malignant mesenchymal component (CAMMC). It is considered a grade III lesion. These lesions demonstrate an abrupt transition into a malignant, highly cellular proliferation with a high mitotic count. Clear cell chondrosarcoma is extremely rare in the larynx with only three cases reported. They are characterized by rounded cells with a predominantly clear cytoplasm with a sparse intervening matrix. Two of the described cases involved recurrences but no distant metastases have been described.80,81,82 Myxoid chondrosarcoma features a myxoid background matrix production with the neoplastic chondrocytes arranged in a “string of pearls” distribution while including other required features of chondrosarcoma. It is required that >10% of the lesion demonstrate this pattern to be considered a myxoid chondrosarcoma.64 These lesions are designated as grade II tumors. Thompson and Gannon note that this designation did significantly affect patient outcome.64 Clinical Presentation and Diagnostic Considerations
Within the larynx, chondrosarcoma most commonly occurs in the cricoid cartilage, followed by the thyroid cartilage. Chondrosarcomas of the arytenoids cartilages and epiglottis are rare.64,70 Clinically, tumors arising from the cricoid cartilage usually present with progressive dyspnea. Because chondroma and chondrosarcoma are slow-growing lesions, symptoms develop slowly over time. If the lesion arises from the thyroid cartilage, the patient may have a palpable neck mass. Other symptoms include hoarseness, dysphagia, and stridor.64,83 Physical examination often reveals a submucosal mass, narrowing of the subglottis and/or dislocation of an arytenoid cartilage. These tumors most commonly demonstrate local growth and destruction. Regional and distant metastases are rare, but have been reported.64 With regard to imaging, 80% of chondromas and chondrosarcomas display calcifications.84 CT is superior to plain radiography in imaging these lesions. Using CT, precise determination of the size and extent of a laryngeal chondrosarcoma can be made. Magnetic resonance imaging (MRI) can be used to aid in determining the soft tissue extent of the lesion when appropriate.85 Behavior and Treatment
Discussion of treatment protocols for chondrosarcoma is important because the treatment of laryngeal lesions is much different from that for tumors of epithelial origin. Surgery is the primary treatment of choice for chondrosarcoma. In general, conservative surgery is advocated for low-grade lesions. Still, this should include wide excision of the tumor with attention to resection of the external perichondrium. Excisions which spare the external perichondrium are more likely to result in recurrence.86 Although total laryngectomy is usually reserved for high-grade and dedifferentiated lesions, it is sometimes required for large low-grade lesions or those that
109
require removal of a large part of the cricoid cartilage. To prevent the need for laryngectomy in these cases, techniques to reconstruct the larynx using local flaps or autologous rib grafts may be employed. Techniques that utilize thyroidtracheal anastomosis and two-stage tracheal autotransplantation have also been described.70,86 CO2 laser resection of primary or recurrent lesions has also been described.62,66 Because some patients may not tolerate or desire surgical management, selected patients have been treated with radiotherapy. This treatment is controversial and generally considered less effective than surgery. Long-term follow-up data has been published for only two patients treated with radiation. Although both of these patients achieved longterm remissions (3 years, 10 years), more research in this area is necessary to assess the efficacy of radiotherapy for chondrosarcoma.87 Laryngeal chondrosarcomas do not show a response to chemotherapy and this is therefore not considered to be a viable treatment option.63,86 Survival following treatment of laryngeal chondrosarcoma is generally quite good. Rinaldo reported a 5-year survival rate of 90%.66 A larger study noted overall survival of 96.3% at an average follow-up of 10.9 years. In this study, 5-year and 10-year disease-free survival rates were 78.9 and 47.8%, respectively.64 Interestingly, 5-year survival rates for grade I, II, and III tumors were reported as 78, 79 and 100%, respectively. Increasing tumor grade is associated with a higher risk of recurrence but does not seem to affect overall survival.
LESIONS OF NEUROENDOCRINE ORIGIN (I) Neuroendocrine Carcinoma Biology and Epidemiology
Neuroendocrine carcinoma (NEC) of the larynx was first described in 1969.88 This tumor has been described using several different terms. Terminology of these neoplasms has been a confusing and contentious issue.89 Wenig and colleagues in 198990,91 proposed a classification of these tumors into three groups based on clinical, histologic, immunocytochemical, and ultrastructural criteria. This classification may have practical utility in terms of prognosis and treatment options. Independently, Ferlito and Friedmann proposed a similar classification with only minor differences.92 Currently, neuroendocrine tumors are divided into the following subgroups: well-differentiated NEC (true carcinoid), moderately differentiated neuroendocrine carcinoma (MDNEC )(atypical carcinoid), and poorly differentiated NEC (small cell carcinoma). Moderately differentiated tumors, which are intermediate between the other (less common) extremes of NEC, make up the bulk of laryngeal NEC. Indeed, 90% of laryngeal NEC is moderately differentiated. The well-differentiated subtype is the least common.93 The histogenesis of laryngeal NEC is unknown. The only well-documented neuroendocrine structure present in the larynx is the paraganglion. However, paragangliomas are much less common when compared to NEC.93 A recent study aimed at identifying additional neuroendocrine cells in the larynx was successful but these cells do not appear to be the precursors of NEC.94
110
HEAD AND NECK CANCER
NEC most commonly occurs in the sixth or seventh decade with a predilection for males. Within the larynx, the supraglottis is most commonly affected. Affected patients commonly have a history of tobacco use. Pathology
The microscopic appearance of NEC varies based on the level of differentiation. In many cases, diagnosis is aided by the use of immunohistochemistry with both neuroendocrine and epithelial markers. Well-differentiated NEC of the larynx is characterized by cell nests composed of uniform cells separated by a fibrovascular or hyalinized connective tissue stroma. The nuclei are round or oval with a vesicular or stippled chromatin pattern and eosinophilic cytoplasm. These tumors demonstrate epithelial mucin staining and argyrophilia in addition to demonstrating cytokeratin, chromogranin, and neuron-specific enolase positivity. On ultrastructural definition, abundant neurosecretory granules, cellular junctional complexes, and inter- and intracellular lumina are seen. Moderately differentiated NEC is polypoid or nodular, displaying varying degrees of surface ulceration. The tumor is submucosal and growth characteristics including glandular, organoid, trabecular, acinar, solid, and nesting patterns are present. Light microscopic features of this tumor may be nonspecific. Combinations of patterns are frequently reported within the same specimen. Delicate fibrovascular or hyaline stroma separates nests of tumor. The tumor cells are large, round, or polyhedral cells containing eosinophilic cytoplasm with a round to oval pleomorphic eccentrically placed nucleus with a stippled chromatin pattern. Nucleoli may be present. Perineural, perivascular, and perilymphatic involvement is common. However, vascular or lymphatic invasion is rare. Histochemical, immunocytochemical, and ultrastructural examination of MDNEC demonstrate both epithelial and neuroendocrine differentiation. Neurosecretory granules are common but not abundant.89,95,96 Calcitonin is a useful immunohistochemical marker in moderately differentiated lesions. Calcitonin expression is reported to be as high as 80% in this subtype.91 In cases where calcitonin expression is present, the tumor may appear similar to medullary thyroid carcinoma. Poorly differentiated NEC lesions are characterized by sheets of undifferentiated small cells with minimal cytoplasm and pleomorphic, hyperchromatic nuclei with delicate chromatin and absent or inconspicuous nucleoli. Mitoses and individual cell necrosis are common. Vascular and perineural invasion are also quite common. Compared to the better-differentiated neuroendocrine tumors, the special histochemical, immunochemical, and ultrastructural investigations are less uniform due to the undifferentiated nature of this tumor type; nevertheless, evidence of both epithelial and neuroendocrine differentiation is clear.95 Scant neurosecretory granules are present.
tumor develops in the supraglottis and grows submucosally. As a result, supraglottic tumors more commonly present at later stages and patients do not experience hoarseness. Occasionally, supraglottic tumors will be pedunculated and therefore present earlier. Airway compromise can occur in the case of early subglottic tumors or large supraglottic lesions. Nonspecific symptoms associated with NEC include dysphagia, “sore throat” and globus sensation. When widespread disease is present, patients may have a neck mass and/or glossopharyngeal neuralgia. Cervical metastases are not uncommon, and patients with relatively small tumors may present with clinically palpable neck disease.97 Because this tumor can present with excess calcitonin production, differentiating NEC from medullary thyroid cancer can be challenging. Pentagastrin stimulation can be useful in this situation. Medullary cancers generally respond with a doubling of serum calcitonin.98 Behavior and Treatment
The pattern of behavior of laryngeal NEC with regard to local recurrence and distant metastases varies based on tumor differentiation. Well-differentiated tumors generally have an indolent course and respond well to surgery. Surgery is the most effective treatment modality as chemotherapy and radiation produce limited tumor regression.99,100 Conservation surgery is recommended when possible as these tumors more commonly involve the larynx focally, sparing the paraglottic space. In situations involving positive margins, radiotherapy may be considered if reoperation cannot be tolerated. Moderately differentiated tumors have a greater tendency toward local recurrence and metastatic disease. These tumors, like the well-differentiated subtype, are considered chemoand radio-resistant. Surgery is the primary form of therapy and total laryngectomy is often necessary for all but small lesions.101 Poorly differentiated tumors are much more aggressive and commonly metastasize to the bone, liver, lungs, brain, and adrenal glands.102 Because these tumors are so aggressive, chemotherapy and radiotherapy are often considered in a palliative fashion, analogous to the management of these tumors when they arise at other sites. Transient response is seen after treatment with several cytotoxic agents, including the platinum complexes, etoposide, the taxanes, gemcitabine, and ifosfamide. A specific role for neoadjuvant or adjuvant chemotherapy, while attractive in principle, has not been defined. Mean survival for patients with poorly differentiated NEC is less than 1 year.102 An unusual feature of laryngeal NEC is the predilection for cutaneous and subcutaneous metastases.103,104 This type of metastasis can occur regardless of tumor grade. It has been suggested that the occurrence of cutaneous metastases is related to dedifferentiation of a portion of the primary tumor. Prognosis for these patients is poor.103 It is important to differentiate these metastases from primary Merkel cell carcinoma and metastatic visceral undifferentiated small cell carcinoma.
Clinical Characteristics and Diagnostic Considerations
CONCLUSIONS
Patients presenting with early stage glottic NEC of the larynx most often present with dysphonia. More commonly, the
The rare tumors of the larynx discussed in this chapter are a diverse group of malignancies with unique characteristics.
RARE TUMORS OF THE LARYNX
Careful consideration of the relevant differential diagnoses generally helps determine the correct diagnosis. Because these tumors are uncommon, treatment parameters are less defined. When possible, conservation surgery should be attempted to spare lung-powered speech. This often involves endoscopic laser surgery for tumor removal. This type of surgery can usually be accomplished without tracheotomy thereby decreasing postoperative morbidity and hospital stay. Still, treatment of the more aggressive tumors described in this chapter may require laryngectomy and eradication of all disease must always remain the top priority for the treating physician.
REFERENCES 1. Negus YE. The Comparative Anatomy and Physiology of the Larynx. New York: Hafner, 1949. 2. Sanders I, et al. The innervation of the human larynx. Arch Otolaryngol Head Neck Surg 1993; 119: 934 – 9. 3. Ackerman LV. Verrucous carcinoma of the oral cavity. Surgery 1948; 23: 670 – 8. 4. Lawson W, Biller HF, Suen JY. Cancer of the larynx. In Myers EN, Suen JY (eds) Cancer of the Head and Neck. New York: Churchil Livingston, 1989: 533 – 593. 5. Batsakis JG. Tumors of the Head and Neck. Clinical and Pathological Considerations, 2nd ed. Baltimore, Maryland: Williams and Wilkins, 1979. 6. Koch BB, et al. Commission on cancer, American College of Surgeons. American Cancer Society. National survey of head and neck verrucous carcinoma: patterns of presentation, care, and outcome. Cancer 2001; 92(1): 110 – 20. 7. Ishiyama A, et al. Papillary squamous neoplasms of the head and neck. Laryngoscope 1994; 104: 1446 – 52. 8. Spiro R. Verrucous carcinoma, then and now. Am J Surg 1998; 175: 393 – 7. 9. Ferlito A, Recher G. Ackerman’s tumor (verrucous carcinoma) of the larynx: a clinicopathologic study of 77 cases. Cancer 1980; 46: 1617 – 30. 10. Miyamoto T, et al. Association of cutaneous verrucous carcinoma with human papillomavirus type 16. Br J Dermatol 1999; 140(1): 168 – 9. 11. Lubbe J, et al. HPV-11 and HPV-16-associated oral verrucous carcinoma. Dermatology 1996; 192: 217 – 21. 12. Fliss DM, et al. Laryngeal verrucous carcinoma: a clinicopathologic study and detection of human papillomavirus using polymerase chain reaction. Laryngoscope 1994; 104: 146 – 52. 13. McCaffrey TV, Witte M, Ferguson MT. Verrucous carcinoma of the larynx. Ann Otol Rhinol Laryngol 1998; 107: 391 – 5. 14. Orvidas LJ, et al. Verrucous carcinoma of the larynx: a review of 53 patients. Head Neck 1998; 20: 197 – 203. 15. Damm M, Eckel HE, Schneider D. CO2 laser surgery for verrucous carcinoma of the larynx. Lasers Surg Med 1997; 21: 117 – 23. 16. Ferlito A, Rinaldo A, Mannara GM. Is primary radiotherapy an appropriate option for the treatment of verrucous carcinoma of the head and neck? J Laryngol Otol 1998; 112: 132 – 9. 17. Tharp ME, Shidnia H. Radiotherapy in the treatment of verrucous carcinoma of the head and neck. Laryngoscope 1995; 104: 391 – 6. 18. Howell JH, Hyams VJ, Sprinkle PM. Spindle cell carcinomas of the nose and paranasal sinuses. Surg Forum 1978; 29: 565 – 8. 19. Thompson LDR, et al. Spindle cell (sarcomatoid) carcinomas of the larynx. Am J Surg Pathol 2002; 26: 153 – 70. 20. Friedmann I, Pyris J. The larynx in systemic pathology. In Symmers WS, Symmers C (eds) Nose, Throat and Ears, 3rd ed. Edinburgh: Churchill Livingstone, 1986. 21. Miyahara H, et al. Spindle cell carcinoma of the larynx. Auris Nasus Larynx 2004; 31: 177 – 82. 22. Marioni G, et al. Spindle-cell tumors of the larynx: diagnostic pitfalls. A case report and review of the literature. Acta Otolaryngol 2003; 123: 86 – 90.
111
23. Olsen KD, Lewis JE, Suman VJ. Spindle cell carcinomas of the larynx and hypopharynx. Otolaryngol Head Neck Surg 1997; 116: 47 – 52. 24. Montgomery H. Dermatopathology. New York: Harper & Row, 1967. 25. Wain SL, et al. Basaloid squamous carcinoma of the tongue, hypopharynx, and larynx. Hum Pathol 1986; 17: 1158 – 66. 26. Paulino AFG, et al. Basaloid squamous cell carcinoma of the head and neck. Laryngoscope 2000; 110: 1479 – 82. 27. Eryilmaz A, et al. Basaloid squamous cell carcinoma of the larynx. J Laryngol Otol 2002; 116: 52 – 3. 28. Ferlito A, et al. Basaloid squamous cell carcinoma of the larynx and hypopharynx. Ann Otol Rhinol Laryngol 1997; 106: 1024 – 35. 29. Raslan WF, et al. Basaloid squamous cell carcinoma of the head and neck: a clinicopathological and flow cytometric study of 10 new cases with review of the English literature. Am J Otolaryngol 1994; 15: 204 – 11. 30. Bahar G, et al. Basaloid squamous carcinoma of the larynx. Am J Otolaryngol 2003; 24: 204 – 8. 31. Bracero F, et al. Hypopharynx and larynx basaloid squamous carcinoma: Our experience with 6 cases. Acta Otorrinolaringol Esp 2001; 52: 229 – 36. 32. Luna MA, et al. Basaloid squamous carcinoma of the upper aerodigestive tract: Clinicopathologic and DNA flow cytometric analysis. Cancer 1990; 66: 537 – 42. 33. Erisen LM, et al. Basaloid squamous cell carcinoma of the larynx: a report of four new cases. Laryngoscope 2004; 114: 1179 – 83. 34. Ellis GL, Auclair PL, Gnepp DR. Surgical Pathology of the Salivary Glands. Philadelphia, Pennsylvania: W.B. Saunders, 1991. 35. Spitz MR, et al. Salivary gland Cancer A case-controlled investigation of risk factors. Arch Otolaryngol Head Neck Surg 1990; 116: 1163 – 6. 36. Saemundsen AK, et al. Epstein-Barr virus in nasopharyngeal and salivary gland carcinomas of Greenland Eskimos. Br J Cancer 1982; 46: 721 – 8. 37. Schneider AB, et al. Salivary gland neoplasms as a late consequence of head and neck irradiation. An Intern Med 1977; 87: 160 – 4. 38. Stewart FW, Footy FW, Becker WF. Mucoepidermoid tumors of the salivary glands. Ann Surg 1945; 122: 820 – 44. 39. Arcidiacono DG, Romeo DG. Tumor mucoepidermoidal salivary. Clin Otorinolaringoiatr 1963; 15: 95 – 108. 40. Cumberworth VL, et al. Mucoepidermoid carcinoma of the larynx. J Laryngol Otol 1989; 103: 420 – 3. 41. Prgomet D, et al. Mucoepidermoid carcinoma of the larynx: report of three cases. J Laryngol Otol 2003; 117: 998 – 1000. 42. Shonai T, et al. Mucoepidermoid carcinoma of the larynx: a case which responded completely to radiotherapy and a review of the literature. Jap J Clin Oncol 1998; 28: 339 – 42. 43. Batsakis JG, Luna MA, El-Naggar A. Histopathologic grading of salivary gland neoplasms: III. Adenoid cystic carcinoma. Ann Otol Rhinol Laryngol 1990; 99: 1007 – 9. 44. Healy WV, Persin KH, Smith L. Mucoepidermoid carcinoma of salivary gland origin. Classification, clinicopathologic correlations, and results of treatment. Cancer 1970; 26: 368 – 88. 45. Mahlstedt K, Ussmuller J, Donath K. Malignant sialogenic tumours of the larynx. J Laryngol Otol 2002; 116: 119 – 22. 46. Jones AS, et al. Adenoid cystic carcinoma of the head and neck. Clin Otolaryngol 1997; 22: 434 – 43. 47. Ferlito A, Barnes L, Myers EN. Neck dissection for laryngeal adenoid cystic carcinoma: is it indicated? Ann Otol Rhinol Laryngol 1990; 99: 277 – 80. 48. Alavi S, et al. Glandular carcinoma of the larynx: the UCLA experience. Ann Otol Rhinol Laryngol 1999; 108: 485 – 9. 49. Garden AS, et al. The influence of positive margins and nerve invasion in adenoid cystic carcinoma of the head and neck treated with surgery and radiation. Int J Radiat Oncol Biol Phys 1995; 32: 619 – 26. 50. Douglas JG, et al. Treatment of locally advanced adenoid cystic carcinoma of the head and neck with neutron radiotherapy. Int J Radiat Oncol Biol Phys 2000; 46: 551 – 7. 51. Huber PE, et al. Radiotherapy for advanced adenoid cystic carcinoma: neutrons, photons or mixed beam? Radiotherapy and Oncology 2001; 59: 161 – 7. 52. Lippert BM, et al. Mucoepidermoid cancer of the larynx. Case report and review of the literature. Laryngol Rhinol Otol 1992; 71: 495 – 9.
112
HEAD AND NECK CANCER
53. Bardwill J. Tumors of the parotid gland. Am J Surg 1967; 114: 498 – 502. 54. Frankenthaler RA, et al. Predicting occult lymph node metastasis in parotid cancer. Arch Otolaryngol Head Neck Surg 1993; 119: 517 – 20. 55. Armstrong JG, et al. The indications of elective treatment of the neck in cancer of the major salivary glands. Cancer 1992; 69: 615 – 9. 56. Ferlito A, et al. Management of clinically negative cervical lymph nodes in patients with malignant neoplasms of the parotid gland. ORL J Otorhinolaryngol Relat Spec 2001; 63: 123 – 6. 57. Medina JE. Neck dissection in the treatment of cancer of the major salivary glands. Otolaryngol Clin North Am 1998; 31: 815 – 22. 58. Airoldi M, et al. Paclitaxel and carboplatin for recurrent salivary gland malignancies. Anticancer Res 2000; 20: 3781 – 3. 59. Damiani JM, et al. Mucoepidermoid-adenosquamous carcinoma of the larynx and hypopharynx: a report of 21 cases and review of the literature. Otolaryngol Head Neck Surg 1981; 89: 235 – 43. 60. Travers F. A case of ossification and bony growth of the cartilages of the larynx. Med Chir Trans 1816; 7: 150. 61. New GB. Sarcoma of the larynx: report of two cases. Arch Otolaryngol 1935; 21: 648 – 52. 62. Saleh HM, et al. Laryngeal chondrosarcoma: a report of five cases. Eur Arch Otorhinolaryngol 2002; 259: 211 – 6. 63. Uygur K, et al. Chondrosarcoma of the thyroid cartilage. J Laryngol Otol 2001; 115: 507 – 9. 64. Thompson LDR, Gannon FH. Chondrosarcoma of the larynx: a clinicopathologic study of 111 cases with a review of the literature. Am J Surg Pathol 2002; 26: 836 – 51. 65. Wang SJ, et al. Chondroid tumors of the larynx: computed tomographic findings. Am J Otolaryngol 1999; 20: 379 – 82. 66. Rinaldo A, Howard DJ, Ferlito A. Laryngeal chondrosarcoma: a 24year experience at the Royal National Throat, Nose and Ear Hospital. Acta Otolaryngol 2000; 120: 680 – 8. 67. Lewy RB. Experience with vocal cord injection. Ann Otol Rhinol Laryngol 1976; 85: 440 – 50. 68. Glaubinger DL, et al. Chondrosarcoma of the larynx after radiation treatment for vocal cord cancer. Cancer 1991; 68: 1828 – 31. 69. Ghalib SH, Warner ED, DeGowin EL. Laryngeal chondrosarcoma after thyroid irradiation. JAMA 1969; 210: 1762 – 3. 70. Thome R, Thome DC, de la Cortina RA. Long-term follow-up of cartilaginous tumors of the larynx. Otolaryngol Head Neck Surg 2001; 124: 634 – 40. 71. Baatenburg de Jong RJ, van Lent S, Hogendoorn PCW. Chondroma and chondrosarcoma of the larynx. Curr Opin Otolaryngol Head Neck Surg 2004; 12: 98 – 105. 72. Nicolai P, et al. Laryngeal chondrosarcoma: incidence, pathology, biological behavior and treatment. Ann Otol Rhinol Laryngol 1990; 99: 515 – 23. 73. Lavertu P, Tucker HM. Chondrosarcoma of the larynx: case report and management philosophy. Ann Otol Rhinol Laryngol 1984; 93: 452 – 6. 74. Cantrell RW, et al. Conservative surgical treatment of chondrosarcoma of the larynx. Ann Otol Rhinol Laryngol 1980; 89: 567 – 71. 75. Batsakis JG, Raymond AK. Cartilage tumors of the larynx. South Med J 1988; 81: 481 – 4. 76. Chiu LD, Rasgon BM. Laryngeal chondroma: a benign process with long-term implications. Ear Nose Throat J 1996; 75: 540 – 9. 77. Devaney KO, Ferlito A, Silver CE. Cartilaginous tumors of the larynx. Ann Otol Rhinol Laryngol 1995; 104: 251 – 5. 78. Lichtenstein L, Jaffe H. Chondrosarcoma of bone. Am J Pathol 1943; 19: 553 – 74. 79. Evans HL, Ayala AG, Romsdahl MM. Prognostic factors in chondrosarcoma of bone: a clinicopathologic analysis with emphasis on histologic grading. Cancer 1977; 40: 818 – 31.
80. Kleist B, et al. Clear cell chondrosarcoma of the larynx: a case report of a rare histologic variant in an uncommon localization. Am J Surg Pathol 2002; 26: 386 – 92. 81. Obenauer S, et al. Unusual chondrosarcoma of the larynx: CT findings. Eur Radiol 1999; 9: 1625 – 8. 82. Said S, et al. Clear cell chondrosarcoma of the larynx. Otolaryngol Head Neck Surg 2001; 125: 107 – 8. 83. Cohen JT, et al. Hemicricoidectomy as the primary diagnosis and treatment for cricoid chondrosarcoma. Laryngoscope 2003; 113: 1817 – 9. 84. Burggraaff BA, Weinstein GA. Chondrosarcoma of the larynx. Ann Otol Rhinol Laryngol 1992; 101: 183 – 4. 85. Mishell JH, Schild JA, Mafee MF. Chondrosarcoma of the larynx. Diagnosis with magnetic resonance imaging and computed tomography. Arch Otolaryngol Head Neck Surg 1990; 116: 1338 – 41. 86. Delaere PR, Vertriest R, Hermans R. Functional treatment of a large laryngeal chondrosarcoma by tracheal autotransplantation. Ann Otol Rhinol Laryngol 2003; 112: 678 – 82. 87. Dailiana T, et al. Chondrosarcoma of the larynx; treatment with radiotherapy. Skeletal Radiol 2002; 31: 547 – 9. 88. Goldman NC, Hood CI, Singleton GT. Carcinoid of the larynx. Arch Otolaryngol Head Neck Surg 1969; 90: 64 – 7. 89. El-Naggar AK. Laryngeal neuroendocrine carcinoma: victims of semantics. Arch Pathol Lab Med 1992; 116: 237 – 8. 90. Wenig BM, Gnepp DR. The spectrum of neuroendocrine carcinomas of the larynx. Semin Diagn Pathol 1989; 6: 329 – 50. 91. Wenig BM, Hyams VJ, Heffner DK. Moderately differentiated neuroendocrine carcinoma of the larynx. Cancer 1988; 62: 2658 – 76. 92. Ferlito A, Friedmann I. Review of neuroendocrine carcinomas of the larynx. Ann Otol Rhinol Laryngol 1989; 98: 780 – 90. 93. Ferlito A, Shaha AR, Rinaldo A. Neuroendocrine neoplasms of the larynx: diagnosis, treatment and prognosis. ORL 2002; 64: 108 – 13. 94. Chung JH, et al. Neuroendocrine carcinomas of the larynx and an examination of non-neoplastic larynx tissue for neuroendocrine tissue for neuroendocrine cells. Laryngoscope 2004; 114: 1264 – 70. 95. Millroy CM, Ferlito A. Immunohistochemical markers in the diagnosis of neuroendocrine neoplasms of the head and neck. Ann Otol Rhinol Laryngol 1995; 104: 413 – 8. 96. Batsakis JG, El-Naggar AK, Luna MA. Neuroendocrine tumors of the larynx. Ann Otol Rhinol Laryngol 1992; 101: 710 – 4. 97. Clayman GL, et al. Head and neck cancer. In Holland JF, et al. (eds) Cancer Medicine, 4th ed. Baltimore, Maryland, Williams and Williams 1667 – 1670. 98. Ponder BAJ. Screening for familial medullary thyroid carcinoma: a review. J R Soc Med 1984; 77: 585 – 94. 99. Baugh RF, et al. Carcinoid (neuroendocrine carcinoma) of the larynx. Ann Otol Rhinol Laryngol 1987; 96: 315 – 21. 100. Patterson SD, Yarington CT. Carcinoid tumor of the larynx: the role of conservative therapy. Ann Otol Rhinol Laryngol 1987; 96: 12 – 4. 101. Gripp FM, et al. Neuroendocrine neoplasms of the larynx. Importance of the correct diagnosis and differences between atypical carcinoid tumors and small-cell neuroendocrine carcinoma. Eur Arch Otorhinolaryngol 1995; 252: 280 – 6. 102. Machens A, Holzhausen JH, Dralle H. Minimally invasive surgery for recurrent neuroendocrine carcinoma of the supraglottic larynx. Eur Arch Otorhinolaryngol 1999; 256: 242 – 6. 103. Schmidt U, et al. Well-differentiated (oncocytoid) neuroendocrine carcinoma of the larynx with multiple skin metastases: a brief report. J Laryngol Otol 1994; 108: 272. 104. Ottinetti A, et al. Cutaneous metastasis of the neuroendocrine carcinoma of the larynx: report of a case. J Cutan Pathol 2003; 30: 512 – 5.
Section 2 : Head and Neck Cancer
8
Nasopharyngeal Carcinoma in Non-endemic Populations June Corry and Bonnie Glisson
The first histologically confirmed case of cancer of the nasopharynx was probably that reported by Michaux,1 who, in 1845, described a 45-year-old male with carcinoma of the base of the skull. There is anthropological evidence, however, that the disease has existed for many centuries. For example, Strouhal2 described an ancient Egyptian skull from the cemetery at Naga-ed-Der in Upper Egypt with features consistent with extensive destruction by a nasopharyngeal cancer. The first English language review of the disease is contained in the textbook A Treatise on Disease of the Nose and Throat by Bosworth in 1889.3 Further details of the history of this fascinating disease may be found in the article by Muir.4 Cancer of the nasopharynx, in the generic sense, includes carcinomas, sarcomas, and lymphomas. Throughout the world, however, “nasopharyngeal cancer” pragmatically refers to carcinoma and to a specific category of carcinoma, abbreviated as NPC. Potentially a source of confusion, NPC designates nonglandular malignancies arising from the epithelium lining the surface and crypts of the nasopharynx. On the basis of ultrastructural features, all types of NPC may be regarded as variants of squamous cell carcinoma (see Figure 1) and are subclassifiable as groups based on their predominant pattern as viewed by the light microscope.5 – 9 NPC represents a nearly unique model in human neoplasia because of its etiopathogenic relationships with viral infection, neoplastic transformation, and immune response of the host.10 Its clinicopathologic aspects, histology, clinical staging systems, and genetic as well as environmental variables have further positioned NPC in a stalking-horse role in oncologic research.
with the posterior wall. The bony roof and posterior wall are formed serially by the basisphenoid, basiocciput, and anterior arch of the atlas. The lateral and posterior walls are, in part, upward extensions of the boundaries of the oropharynx. In the lower part of the lateral wall, the superior constrictor muscle sends its fibers posteriorly to attach to the basisphenoid. Between the upper border of the superior constrictor and the skull base is stretched the pharyngobasilar fascia with the eustachian tube lodged between the medial pterygoid plate and the superior constrictor. The tubal ampulla’s inward bulge creates a slitlike space between it and the posterior wall – the pharyngeal recess or fossa of Rosenm¨uller, filled, in part, by the levator palati muscle that lies between the pharyngobasilar fascia and the mucous membrane.11 The nasopharyngeal lymphoid tissue or adenoid is concentrated at the junction of the roof and the posterior wall of the postnasal space. There are other lymphoid aggregates about the tubal openings. The sensory nerve supply to the postnasal space is provided by the glossopharyngeal and maxillary nerves. Beneath the mucous membrane of the roof is the vestigial pharyngeal hypophysis. Lying in the midline near the vomero-sphenoidal articulation, its presence is a reminder of the embryologic origin of the anterior pituitary from Rathke’s pouch. The hypophyseal vestige may be partly or completely surrounded by the basisphenoid. Also in the midline but dorsal along the roof and separated from the pharyngeal hypophysis by adenoidal tissue is an epithelial recess, sometimes called the pharyngeal bursa, an occasional locus of inflammation or cyst formation. Believed to be formed by a tethering of pharyngeal endoderm to the tip of the embryonic notochord, the recess has no relationship with Rathke’s pouch.11
ANATOMY OF THE NASOPHARYNX
Microscopic Anatomy
Gross Anatomy
Batsakis et al.5 have summarized our current knowledge of the histology of the nasopharynx mucosa of the nasopharynx. It is composed of three basic cell types: pseudostratified
INTRODUCTION
The roof of the nasopharynx begins behind the posterior nasal choanae and slopes downward where it becomes continuous
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
114
HEAD AND NECK CANCER
Figure 1 Electron micrograph of undifferentiated carcinoma of nasopharynx showing epidermoid characteristics, for example, cell junctions and tonofilaments. Original magnification: ×9000.
columnar (respiratory) cells, squamous cells, and intermediate (pseudostratified) cuboidal cells. All three types are found during fetal development, with the respiratory type being the first to evolve. There is an increase in squamous epithelium until, in the adult nasopharynx, the dominance of respiratory over squamous epithelium is reversed. A commensurate increase in the intermediate type of epithelium is not seen, but it persists at junctions between respiratory and squamous epithelia with its greatest density at the junction of the oropharynx and nasopharynx. In the adult, squamous epithelium covers approximately 60% of the entire nasopharyngeal surface. The intermediate epithelium is aptly named since it is intermediate in a topographic as well as cytologic sense. Investigations in nonhuman species have indicated a change of some of the intermediate cells to either ciliated respiratory or squamous cells.5 Resembling the intermediate cell layers in both respiratory and squamous epithelia, the intermediate epithelium’s greatest density is in the sites of predilection for nasopharyngeal carcinomas (NPC) and it is also the closest normal histologic homologue of the nonkeratinizing or undifferentiated carcinomas of the nasopharynx.5
EPIDEMIOLOGY AND ETIOLOGY OF NPC The etiology of NPC is very likely multifactorial: genetic, environmental, and viral. There are at least three major risk factors: (i) a genetically determined predisposition allowing an Epstein-Barr virus (EBV) infection of the type that permits (ii) integration of the genome of the virus into the chromosomes of some nasopharyngeal epithelial cells, thereby priming them for (iii) neoplastic transformation by some environmental cofactor. Alternatively, the environmental agent(s) may trigger the viral genome in the cells to oncogenic activity. Although environmental factors appear to be essential, the high frequency in disparate ethnic groups points to different operative agents for each group.5,7,10 As judged by agespecific incidences, Scandinavians and American blacks and whites appear to have a different etiologic origin for NPC
from that of Chinese Americans, or Hong Kong or Singapore Chinese. In the Chinese, the incidence curves rise sharply after the third decade; those for non-Chinese show a rise after the fourth or fifth decade. Native Alaskans have a curve pattern similar to that of Chinese, but Tunisians have a bimodal curve, with an early peak in the second decade. In all ethnic groups, the incidence in males is two to three times greater than that in females. On the basis of data published in the latest edition of Cancer Incidence in Five Continents,12 the worlds highest incidence rates are found in Hong Kong (particularly among the boat people). With its population primarily derived from Guandong Province in southeastern China, Hong Kong has consistently reported very high rates of about 25–30 per 100 000 in males and 10–15 per 100 000 in females. Cantonese Chinese who have migrated to other parts of the world retain a high predisposition to the disease, although at a lesser rate than those living in China.13 Intermediate rates (5–15 per 100 000 in males) are reported in other peoples of Southeast Asia (Malays, Indonesians, Filipinos, Thais, and Vietnamese), Eskimos, and North African Arabs. A relatively low-to-moderate incidence (1–5 per 100 000) occurs in northern Chinese, Polynesians, Maltese, and central Africans. The disease is rare, <1 per 100 000 in Caucasians, Japanese, Koreans, and the population of the Indian subcontinent (see Table 1). In Europe, the populations of the southern countries (Spain, Italy, France, Balkan states) are at a relatively higher risk than those from northern countries. In the past, the increased risk was linked to populations living on the Mediterranean Sea coast. The recent data do not seem to confirm this finding. In fact, only Malta and Israel (Jews born in Africa or Asia) show higher incidence rates, whereas the coastal populations of Spain, Italy, and the Balkan states have a lower incidence than those living in the interior parts of those countries. Levine and others14 have studied demographic patterns for NPC in the United States. These were obtained from the Third National Cancer Survey, the Surveillance Epidemiology and End Results (SEER) program, and the Connecticut Tumor Registry. Approximately seven of every eight Table 1 Age-adjusted incidence rates by ethnic grouping.
Incidence
Ratea
Ethnic grouping
Very high
>25
Indigenous southern Chinese Emigrant southern Chinese (Singapore, USA) Non-Chinese Southeast Asians, Arabs, Eskimos Northern Chinese, Polynesians, central Africans, Maltese Caucasians, people of Indian subcontinent, Japanese, Koreans
High
15 – 25
Intermediate
5 – 15
Moderate to low
1–5
Rare
<1
a
Per 100 000 in males.
NASOPHARYNGEAL CARCINOMA IN NON-ENDEMIC POPULATIONS
malignant tumors of the nasopharynx (1202 patients) were classified as NPC. While 84% of white patients with cancer of the nasopharynx had NPC, more than 90% of nonwhite patients had NPC. The preponderance of NPC in the case material held for all but the younger patients; sarcomas were more frequent among whites under 10 years of age. White patients had the highest frequency of squamous cell carcinomas. Undifferentiated carcinomas were more common in black and Chinese American patients and were relatively frequent in young patients of all races. Chinese Americans have a greater risk of developing NPC than any other racial/ethnic group in the United States. Whites and blacks have similar risks, except in younger age-groups where there is a minor postadolescent age peak in NPC risk that is more pronounced for blacks than for whites in the United States.15 A similar young age peak has been observed in other parts of the world such as India, Tunisia, and Sudan, but remains unexplained. The apparent high risk in the young black population in the United States has been considered to be related to rural residence and low socioeconomic status.16 The morbidity data from the United States do not show outstanding geographic or temporal variation in the NPC risk for whites.14,16 Only the relatively high mortality rate in Alaska stands out as a significant factor in mortality studies of the states and counties.
Epstein-Barr Virus and Biologic Implications EBV is ubiquitous in humans, and antibodies to polypeptides of the virus are present in over 80% of human serum samples from the USA, and in higher percentages from Asian and African populations.10,17 Practically no one escapes infection from this herpes-group virus. Primary infections often remain clinically inapparent or not recognized as being due to EBV, particularly in subjects under the age of 5. The consequences of EBV infection vary in different populations: it is associated predominantly with infectious mononucleosis in the western hemisphere, Burkitt’s lymphoma in Africa, and NPC in Asia. Occurrence of these three EBV-associated diseases is unusual outside their normally associated populations, strongly suggesting the role of additional factors in the populations at risk. The apparent differences in geographic distribution of the three main EBV-associated diseases have also prompted suggestions that different strains of EBV are prevalent in different areas, but this has not been verified and likely is not a factor.10 Whether clinically manifest or not, primary EBV infections establish a permanent EBV carrier state in the lymphatic system and also in the major salivary glands.18 This is reflected in the life-long persistence of EBV-specific antibodies, at almost constant titers, and an intermittent excursion of EBV into the oropharynx.10,18,19 The association between EBV and NPC was initially discovered in 1966 by Old et al.,20 who showed that patients with undifferentiated carcinoma of the nasopharynx had elevated IgG and IgA antibody titers against EBV early and viral capsid antigens (VCA). Since then a large number of studies have shown that essentially all cases of undifferentiated carcinoma of the nasopharyngeal type throughout the world contain the EBV genome regardless of the local incidence
115
of the tumor or the ethnicity of the patients. Epstein-Barr nuclear antigen (EBNA 1) is expressed in practically all NPCs. Evidence strengthening the causal association of EBV with NPC in genetically predisposed groups was provided by the demonstration that preinvasive dysplastic lesions of the nasopharynx contain monoclonal copies of the EBV genome (using the criterion of terminal repeat reiteration frequency), indicating that EBV infection is likely to be an early initiating event in the development of NPC.21 Reports on the association between EBV and differentiated (WHO type 1) NPC are contradictory. Although some studies have been unable to demonstrate the presence of EBV in squamous cell carcinomas, others have shown positive hybridization signals although expression of viral encoded transcripts appears to be downregulated, once tumor cells differentiate and produce keratin.22 Although EBV infection is clearly an important factor in the pathogenesis of NPC, its ubiquitous distribution contrasted with the distinct geographical epidemiology of NPC implicates a multistep process. Genetic predisposition is obviously a factor.23 It would also appear that in highrisk groups, environmental carcinogens such as salted fish may play a role. Genetic polymorphisms in cytochrome p450 enzymes, such as the c2 allele of CYP2E, which metabolically activates nitrosamines, may be responsible for this association with salted fish exposure.24 In geographic areas of sporadic incidence, tobacco and ethanol exposure increase the risk for development of WHO type 1 NPC, but not type 2 or 3.25 A population-based case –control study in the United States showed an odds ratio of 1.9 for NPC when the glutathione S-transferase M1 gene (GSTM1) is absent.26 Interestingly, although this enzyme is involved in the metabolism of tobacco carcinogens, the relationship between GSTM1 absence and risk was not modified by tobacco exposure and the risk was similar among Caucasians, African Americans, and Asians, and across WHO types.
Genetics Several genetic systems have been investigated in patients with NPC. In southern Chinese populations, a strong association with human leukocyte antigen (HLA) alleles A2, B14, and B46 has been observed. This susceptibility may be due to a gene closely linked to HLA loci as suggested by Lu et al.27 However, data from other races are inconclusive. Burt et al.28 investigated associations between NPC and HLA antigens at the HLA-A, -B, -C, and -DQ loci and alleles at the DRB1 locus in a population-based multicenter study in the United States. Data from 82 cases and 140 controls were presented, making this the largest study yet performed in an area of sporadic incidence of NPC. An analysis was undertaken to compare their results with previously published findings. This found a significant protective association with A2 antigen in non-Chinese, a protective association with A11 across all races, and increased risk associated with B5 in Caucasians. Associations were found to be more pronounced in younger patients. Familial segregation of NPC is noted in areas of endemic incidence and more rarely, in Caucasians. However, no specific inherited gene in familial NPC has been identified and
116
HEAD AND NECK CANCER
the possibility that shared exposure to an environmental carcinogen contributes to familial clustering cannot be excluded. Early karyotyping analyses of primary tumors and cell lines showed nonrandom structural or numerical abnormalities in defined regions of chromosomes 1, 3p, 3q, 5q, 9p, 11q, 12, 13q, 14q, and X.29 – 32 More recent studies utilizing comparative genomic hybridization in primary NPC have shown numerous genetic events, both gains and losses, suggesting the involvement of both oncogenes and tumor suppressor genes.29 The most critical of these in NPC appear to be inactivation of tumor suppressor genes on 3p (RASSF1A, FHIT, RARβ2), 9p (p15INK4B, p16/INK4A, 11q (TSLC1), 13q (EDNRB), 14q, and 16 q and increased copy number and expression of c-myc at 8q24 and EGFR at 7p12. On the basis of the accumulating evidence of genetic and epigenetic alterations, a theoretical model for the molecular pathogenesis of NPC has been proposed.29 Continued investigation in this area is critical not only to further the understanding of pathogenesis, but also to identify potential targets for biologic-based therapy.
PATHOLOGY Classification and Histology of Nasopharyngeal Carcinoma Regardless of geographic distributions, the nonglandular, nonlymphomatous, and nonsarcomatous malignancies are the most common neoplasms of the nasopharynx5 (see Table 2). In high-risk regions, these carcinomas dominate cancer statistics for the head and neck. Over the years, diversity of diagnostic nomenclature and an absence of a uniform histologic reporting system have bedeviled correlation with results of therapy and prognosis.5 Cognizant of this, in 1978 the World Health Organization (WHO) divided NPC into three histologic types: squamous cell (type 1), nonkeratinizing (type 2), and undifferentiated (type 3).5,7,9 Type 1 histology is uncommon in areas of endemic incidence while it represents approximately 25% of cases in North America.30 Type 3 tumors represent the vast majority of cases in southern China (90–95%) and both types 2 and 2 are associated with elevated EBV serology at diagnosis.31 The latter association with EBV and the finding that NPC tissue sometimes shows a mixed pattern of types 2 and 3 led to a revised WHO classification in 1991.32 This defines NPC as either squamous cell carcinomas (WHO type 1) or nonkeratinizing carcinomas. The second Table 2 Carcinomas of the nasopharynx.
Nasopharyngeal carcinoma (NPC) Keratinizing squamous cell carcinoma (WHO type 1) Nonkeratinizing carcinoma Differentiated nonkeratinizing carcinoma (WHO type 2) Undifferentiated carcinoma (WHO type 3) Adenocarcinomas Salivary type Nonsalivary type Neuroendocrine carcinoma Teratocarcinoma
Figure 2 Keratinizing squamous cell carcinoma of the nasopharynx, WHO type 1. H&E, original magnification: ×400.
group is subdivided into differentiated (WHO type 2) and undifferentiated (WHO type 3) carcinomas. Squamous cell carcinoma of the nasopharynx is like the squamous cell carcinoma in other anatomic sites of the upper aerodigestive tracts. The carcinomas manifest obvious and readily identifiable keratin products, and their growth pattern is typical of that found in any other squamous cell carcinoma (see Figure 2). In general, the carcinoma is moderately differentiated and is accompanied by a desmoplastic host response. Since it is preponderantly a surface growth, endoscopic examination of the nasopharynx usually identifies the carcinoma. The average age of patients with squamous cell carcinoma of the nasopharynx is somewhat higher than that of all NPC patients. It is rarely found in patients younger than 40 years.9 Like the squamous cell carcinomas, differentiated nonkeratinizing carcinomas exhibit variable degrees of differentiation within the limits of their definition. The cells have a maturation sequence that ends without evidence of squamous differentiation at the light microscopic level (see Figures 3 and 4). Growth may be papillary and/or plexiform. The cells have fairly well-defined cell margins and the neoplastic islands are usually quite well-delineated from the adjacent stroma. In some of the carcinomas, there is a pseudostratified arrangement of cells, not unlike that noted for the intermediate epithelium of the nasopharynx. While histologic differences between squamous cell carcinoma and nonkeratinizing carcinomas are sharp, those between nonkeratinizing and undifferentiated carcinomas are sometimes vague and may be arbitrary, thus their shared group in the most recent WHO classification. Undifferentiated carcinoma of the nasopharynx is composed of primitive cells whose most consistent feature is a single, prominent nucleolus, and a nucleus with distinct membrane and, in many cases, nuclear vesiculation (see Figure 5). In contrast with the other NPC types, the cell margins of this carcinoma are often indistinct and the tumor often has a syncytial appearance. The cellular arrangement, however, is variable, with masses, strands, or individual cells lying in a lymphoid stroma. The variety of cytoplasmic forms
NASOPHARYNGEAL CARCINOMA IN NON-ENDEMIC POPULATIONS
Figure 3 Nonkeratinizing carcinoma of the nasopharynx, WHO type 2. Note the sharp delimitation from surrounding lymphoid tissue and in this example, a spindle character to the neoplastic cells. H&E, original magnification: ×360.
Figure 4 Differentiated nonkeratinizing carcinoma, WHO type 2. This example manifests, in addition to neoplasm – stroma demarcation, clear cells and “intermediate type” cells. H&E, original magnification: ×400.
and growth patterns has given rise to descriptive terms such as anaplastic, clear cell, spindle cell, simplex, and lymphoepithelioma. Undifferentiated NPC has a striking invasive and metastasizing capability, and tissue reactions to the infiltrating tumor are usually limited. Fibrosis and desmoplasia, for example, are never prominent unless there has been prior radiation therapy.9 Usually, there is no discernible reaction and the carcinoma maintains an intimate relationship with lymphoid tissues. The presence or absence of lymphocytes in NPC is not a factor while making the diagnosis. It is now firmly established that the lymphocytes are not neoplastic or integral to the carcinomas. They can be found in all three of the WHO types, but are most often associated with undifferentiated
117
Figure 5 WHO type 3 NPC or undifferentiated carcinoma. Vesicular nuclei, prominent nucleoli, indistinct cell membranes, and a lymphoidlike character are manifested. Note intimate relationship with nonneoplastic lymphocytes. H&E, original magnification: ×420.
carcinomas. Approximately 98% of undifferentiated, 70% of nonkeratinizing, and 37% of squamous cell carcinomas of the nasopharynx are associated with lymphocytes.9 The lymphoid “stroma” is not entirely passive. Metastases of undifferentiated NPC to nonlymphoid tissues may also have an accompaniment of lymphoid cells. Several histologic findings of a host response to NPC merit mention. Some may have as yet an unknown prognostic value; others, in the presence of undifferentiated NPC, may mislead the surgical pathologist to a diagnosis of lymphoid neoplasm. A mixture of lymphoid cells and plasma cells, sometimes associated with polymorphonuclear leukocytes, is found in nearly all forms of NPC. A mild-to-moderate stromal eosinophilia is evident in about one-fourth of the carcinomas, most often with the undifferentiated types, where it may be a conspicuous feature. Some authors have also reported an amyloid-like material in the stroma and also sometimes in the cytoplasm of the carcinoma cells.33 The amyloid, unlikely to be of a secondary type, is found most often in association with nonkeratinizing NPC. In the lymphoid tissue immediately adjacent to undifferentiated NPC, there may be a predominance of T lymphocytes, a finding of possible significance because of the inherently B cell nature of the lymphoid tissue of the nasopharynx. T zone histiocytes (Langerhans’ cells and precursors) at primary carcinoma sites may also play a role in an immune reaction.34 In lymph nodes with or without metastases from NPC and, on occasion, in the nasopharynx itself, tuberculoid granulomas are found. Usually around the neoplasm, the epithelioid granulomas may be accompanied by large numbers of eosinophils, fibrosis, and caseous necrosis. Infective granulomas or Hodgkin’s disease may be simulated. Despite the variations in histologic appearance of the WHO types, their proposed mode of histogenesis, the lability and maturational tendencies of the nasopharyngeal epithelium, and clinicopathologic findings suggest all three types may be histologically homogeneous. The tendency for an epidermoid differentiation and the light optic findings of a mixed
118
HEAD AND NECK CANCER
cell or intermediate population in otherwise prototypic histologic classes support this homogeneity, as does ultrastructure. Shanmugaratnam et al.7 have indicated that features of more than one histologic type were present in 25% of all NPCs studied in a Singapore population. In such instances, classification is based on the predominant type found in the primary lesion. Carcinomas histologically similar or indistinguishable from NPC types 2 and 3 have been found elsewhere in the epithelium of Waldeyer’s ring,35 the larynx,36 the thymus,37 major salivary glands,18 and cervix.38 The role of EBV in some of these carcinomas is strongly suggested by serologic profiles, by the presence of EBV-associated nuclear antigen in the carcinoma cells, and by high levels of viral genomes in the DNA. A histomorphological feature common to all carcinomas is an intimacy between epithelium and lymphoid cells, not unlike that of the nasopharyngeal mucosa. This “lymphoepithelium” is found in the base of the tongue, tonsillar and adenoidal crypts, in association with salivary ducts, laryngeal “tonsil”, and obviously in the thymus. These “lymphoepithelial carcinomas” are infrequent, for example, less that 5% of the base of the tongue and tonsil carcinomas, but their biologic behavior and response to treatment further qualify them as carcinomas of nasopharyngeal type.
Immunologic Aspects of NPC In the 1980s, a collaborative prospective study of North American patients with different histopathological types of NPC identified certain biologic characteristics of EBV that have clinical importance for diagnosis and possibly for the management of the disease.10,39 Immunovirological tests having diagnostic significance include antibody titers to VCA and early antigen (EA). The VCA (IgA) is more specific, while the EA is more sensitive. These tests may help resolve the diagnosis in patients who present with metastatic carcinoma in a neck node from an unknown primary. In addition, antibody dependent cellular cytotoxicity (ADCC) assays titrating EBV-induced membrane antigen complex appear to be predictive of clinical outcome and prognosis in patients with NPC types 2 and 3. High ADCC titers at diagnosis are associated with a more favorable prognosis, regardless of the disease stage. The incidence of positive titers in NPC appears to be the same regardless of tumor size and hence can complement the diagnostic evaluation of patients. In southern China, positive titers for IgA to VCA have been used in screening programs, with predominantly early stage NPC diagnosed.40 Anti-EBV serologic findings distinguish WHO type 1 from types 2 and 3 carcinomas. Types 2 and 3 carcinomas manifest characteristic anti-EBV profiles and are more often small and submucosal, and may be clinically occult.9 They are more radiosensitive than WHO type 1 carcinomas. WHO types 2 and 3 carcinomas appear to occur at an earlier age, manifest longer disease-free periods after treatment, and have a better survival even though early and advanced metastases to the neck are more common, as is the risk of distant metastasis.9,16 Early results of NPC screening in high-risk populations using a transoral brush biopsy appear promising. The brush provides nasopharyngeal cell DNA, which is tested for EBV
genomic sequences using polymerase chain reaction (PCR). In a sample population of 178 (21 with NPC and 157 without), a sensitivity of 90% and a specificity of 99% were found for detecting NPC using this technique.41 There have also been interesting results using plasma EBV DNA for therapeutic monitoring. Levels greater than 500 copies mL−1 at 6 weeks posttreatment had a high predictive value for future distant failure.42
CLINICAL PRESENTATION AND DIAGNOSIS OF NASOPHARYNGEAL CARCINOMA NPC, especially of the WHO types 2 and 3, usually arises in the region of the fossa of Rosenm¨uller. The early symptoms of the disease are neither pathognomonic nor specific. The clinical presentation of NPC in the North American population has recently been documented in a collaborative prospective study.43 Over a third of patients will notice a mass in the neck as their first symptom and approximately an equal number will have a sensation of unilateral ear fullness or plugging and hearing loss. A persistent serous otitis media, especially if unilateral in an otherwise healthy adult, should arouse suspicion of a carcinoma of the nasopharynx. A cancer of the nasopharynx will seldom produce choanal or nasal obstruction but initial bleeding or bloody nasal drainage will be noticed by approximately one-fifth of the patients. The triad of a mass in the neck, a conductive hearing loss, and nasal obstruction with blood-tinged drainage will frequently be present by the time the diagnosis of NPC is made. The proximity of foramen lacerum and thus the floor of the middle cranial fossa allows for direct tumor extension into the cranium and involvement of adjacent nerves. One-fifth of patients will have symptoms of cranial nerve involvement at diagnosis. Facial pain and paresthesias suggest tumor infiltration of the trigeminal nerve branches, and diplopia from paralysis of the lateral rectus muscle is a sign of involvement of the abducens nerve. Involvement of cranial nerves III and IV indicates more advanced disease along the cavernous sinus. Tumor extension may occur laterally into the parapharyngeal space and involve cranial nerves IX, X, and XI, producing a jugular foramen syndrome. A persistent occipitotemporal headache, especially unilateral, is reported by one of every six patients. Only very occasionally will NPC invade the parotid gland and cause facial nerve paralysis. Proptosis will occur when cancer invades through the posterior portion of the orbit. Trismus is an indication of pterygoid muscle invasion and cancer extension into this space. At diagnosis, 9 of every 10 patients will have palpable lymph node metastases with bilateral involvement in half of them. The lymph nodes most frequently involved are in the subdigastric area and in the chain along the spinal accessory nerve in the posterior triangle. The sentinel lymph node for NPC is located underneath the upper insertion of the sternocleidomastoid muscle. Frequently the neck mass is large and painless, and it can enlarge quite rapidly because of necrosis or hemorrhage. Retropharyngeal lymph node metastasis, when extensive, produces a characteristic
NASOPHARYNGEAL CARCINOMA IN NON-ENDEMIC POPULATIONS
119
Table 3 Frequency of cranial nerve involvement in cancer of the nasopharynx.a
Cranial nerves I Number of patients Frequency (%) a
13 0.5
II
III
IV
V
VI
VII
VIII
IX
X
XI
XII
114 4.0
236 8.2
207 7.2
521 18.1
600 20.9
133 4.6
49 1.7
264 9.2
233 8.1
154 5.4
358 12.5
Based on 2871 patients of whom 641 (22.3%) manifested cranial nerve involvement. Adapted from Sawaki et al.44 1971 University of Tokyo Press.
syndrome of pain referred to the ipsilateral neck, ear, head, forehead, and orbit. It may be associated with a stiff neck or pain when cervical dorsiflexion is attempted. Physical examination includes inspection of the nasopharynx, either indirectly with a mirror or preferably by direct visualization through a fiberoptic endoscope. The tumor usually appears as an asymmetric mass with telangiectasia on its friable surface and centered in the fossa of Rosenm¨uller. Depending on the size of the primary tumor, distortion of the soft palate can occur. Straw-colored serous otitis media is usually unilateral. Evaluation of the cranial nerves may show subtle signs of tumor infiltration. The earliest signs are usually extraocular muscle dysfunction, especially lateral rectus palsy, and signs of trigeminal nerve involvement such as hyperesthesia and atrophy of masticatory muscles. The relative frequency of involvement of each cranial nerve is indicated in Table 3.44 Biopsy of the tumor in the nasopharynx can be done under topical anesthetic by cocaine applications through the nasal cavity. Palpation of the neck, especially in the upper third, will frequently reveal lymph node metastases that are often of large size, multiple, and sometimes fixed to the surrounding structures. Squamous cell carcinomas (WHO type 1) produce fewer lymph node metastases as opposed to the nonkeratinizing carcinomas (WHO types 2 and 3). Radiologic evaluation of the nasopharynx, the base of the skull, paranasal sinuses, and the neck is mandatory for appropriate staging and treatment. Computed tomography (CT) allows for the definition of the extent of primary tumor and the amount of invasion and infiltration of surrounding structures. Bone destruction at the floor of the sphenoid sinus, the adjacent middle cranial fossa, the clivus, and the pterygoid plates can be documented, as well as the extension of tumor through the foramen lacerum into the middle cranial fossa, laterally into the parapharyngeal space, anteriorly into the nasal cavity, and anterolaterally into the orbit. It is important to demonstrate invasion into the posterior ethmoid cells and adjacent portion of the maxillary antrum. Magnetic resonance imaging (MRI) further improves radiological staging of the disease, particularly for detecting base of skull invasion and intracranial extension. Ng et al. reported on 67 patients who had both staging CT and MRI scans. They found staging MRI demonstrated skull base invasion and intracranial extension in a further 20% patients compared with their staging CT.45 Poon et al. reported similar results.46 They compared results of 48 patients who had both staging CT and MRI and found that MRI demonstrated increased volume of disease and upstaged disease in 16 patients (33%), of whom 8 (25%) had their
Figure 6 Sagittal T1-weighted MRI image of a patient with clinically localized NPC showing invasion of the clivus by tumor. The tumor appears as a dark area within the bright signal of normal marrow.
T stage upgraded. Consequently MRI is now considered the optimal standard diagnostic staging tool in NPC (see Figure 6). CT and MRI scans may also reveal metastatic spread to lymph nodes that is undetectable clinically. For example, Figure 7 shows a patient in whom obviously involved retropharyngeal lymph nodes were demonstrated radiographically. Accurate delineation of disease is a sine qua non for optimal radiation dose delivery to the entire gross tumor volume. It is presumably the explanation for the fascinating finding of a prognostic impact of MRI in NPC. This was first demonstrated by Lee et al.47 In their series of 2687 patients, 860 (32%) were staged by MRI, which resulted in significantly better outcomes than in those staged by CT. Despite a significantly higher stage in the MRI group (T3–4 46 vs 27%, N2–3 37 vs 27%), their 5-year local failure-free rate and overall survival rate was significantly better than that of the CT-staged patients, 91 versus 87% and 80 versus 74%. Using multifactorial analysis, Corry et al. have also demonstrated a significant independent impact of staging MRI on
120
HEAD AND NECK CANCER Table 4 Distant metastatic sites of nasopharyngeal carcinoma: a study of 2637 patients.
Metastatic site
Number of patients (%)
Bones Lungs Liver Distant lymph nodes Brain
342 (41) 256 (30) 121 (14) 101 (12) 18 (2)
Modified from Huang and Chu.51 with permission from Elsevier.
Figure 7 Axial T1-weighted fat-saturated contrast-enhanced MRI image showing bilaterally enlarged retropharyngeal lymph nodes harboring metastases from NPC.
local control and overall survival in Asian and non-Asian patients.48 The majority of patients with NPC present with locoregionally advanced disease, and at least one-fifth of patients can be reckoned to have had occult distant metastases at presentation on the basis of relapse patterns following treatment.49,50 Sites of predilection for metastatic spread are presented in Table 4.51 The current US National Comprehensive Cancer Network Guidelines recommend imaging of chest, liver, and bone for patients with WHO types 2 and 3 NPC associated with N2–3 disease.52 Recently, Kumar et al. prospectively evaluated 139 patients with WHO type 3 NPC with chest X ray, liver ultrasonography, and bone scan.53 The incidence of occult metastatic disease increased with higher overall stage and N status, with less than 5% yield for patients with N0–2 disease compared to a 14.3% yield for patients with N3 disease. The utility of positron emission tomography (PET) scanning to screen for occult distant metastases is currently being investigated.54
Staging For many years three separate staging systems have been used for NPC. Among Caucasians, the American Joint Commission for Cancer (AJCC) system was most commonly used in the United States and the similar International Union Against Cancer (UICC) system elsewhere in the Western world. By contrast, in southeastern Asia the Ho system55 was
used. One of the major achievements of the past decade has been the successful development of a single staging system with international consensus. The new staging system was ratified by both the UICC and the AJCC in 1997 and is maintained in the sixth edition of the UICC Tumor-NodeMetastasis (TNM) classification56 and the AJCC staging manual.57 It embraces many of the features of the Ho system and its derivatives, and thus provides a staging system for NPC that is quite different from that of other head and neck cancers. Details of the current unified system are provided in Table 5 and the stage groupings in Table 6. Central to the staging system is the use of cross-sectional imaging with CT and/or MRI to demonstrate the extent of primary tumor and nodal involvement. The principle adopted was that data guiding the staging revision should be based on tumors whose extent was assessed using cross-sectional imaging since outcomes based on previous clinical and radiologic findings would likely be invalid. The main differences between the previous UICC/AJCC staging systems and the new classification are as follows: • T categories: – T1: Includes all tumors confined to the nasopharynx. The old classification recognized different subsites within the nasopharynx. The lack of validity of these Table 5 Unified TNM staging system.
Primary tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ T1 Tumor confined to the nasopharynx T2 Tumor extends to soft tissues of oropharynx and/or nasal fossa T2a without parapharyngeal extension T2b with parapharyngeal extension T3 Tumor invades bony structures and/or paranasal sinuses T4 Tumor with intracranial extension and/or involvement of cranial nerves, infratemporal fossa, hypopharynx, or orbit Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Unilateral metastasis in lymph node(s), 6 cm or less in greatest dimension, above the supraclavicular fossa N2 Bilateral metastasis in lymph node(s), 6 cm or less in greatest dimension, above the supraclavicular fossa N3 Metastasis in a lymph node(s) N3a greater than 6 cm in dimension N3b extension to the supraclavicular fossa
NASOPHARYNGEAL CARCINOMA IN NON-ENDEMIC POPULATIONS Table 6 Stage grouping: nasopharynx.11
Stage Stage Stage Stage
0 I IIA IIB
Stage III
Stage IVA
Stage IVB Stage IVC
Tis T1 T2a T1 T2a T2b T2b T1 T2a T2b T3 T3 T3 T4 T4 T4 Any T Any T
N0 N0 N0 N1 N1 N0 N1 N2 N2 N2 N0 N1 N2 N0 N1 N2 N3 Any N
M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1
subdivisions in prognosis has been demonstrated by many authors, for example, Sham et al.58 – T2: Includes involvement of the oropharynx or nasal fossa, which under the old system was categorized as T3. Also, recognition of parapharyngeal space extension is provided. This is defined as posterolateral infiltration beyond the pharyngobasilar fascia. – T3: Includes tumors invading the skull base or paranasal sinuses, which were previously categorized as T4. – T4: This now requires intracranial extension and/or involvement of cranial nerves or involvement of the infratemporal fossa, hypopharynx, or orbit. • N categories: The critical prognostic significance of level of lymph node involvement in the neck, as well as size, has been recognized. Multiplicity and clinical fixity of nodes are not criteria for staging.59 – N1: Unlike other head and neck cancers, the N1 category includes nodes up to 6 cm in greatest dimension, provided they are unilateral and above the supraclavicular fossa. – N2: Includes bilateral lymph nodes above the supraclavicular fossa of the neck up to 6 cm in size. – N3: Presence of nodes greater than 6 cm in diameter or any involvement of or extension to the supraclavicular fossa constitutes N3.
MANAGEMENT AND PROGNOSIS While there is good evidence now that concurrent chemotherapy and radiotherapy yields improved results in locally advanced disease, radiotherapy remains the backbone of treatment in NPC.
Radiotherapy Because of the rarity of NPC in nonendemic areas, most published series are relatively small and span long periods over which both diagnostic and therapeutic technologies have undergone a revolution. In the first edition of this textbook, a tabulation of all Caucasian series published between 1980 and 1986 was provided. Mostly these included patients diagnosed between the 1950s and 1970s, that is, in the era
121
preceding cross-sectional imaging. Reported overall 5-year survivals averaged around 40%. The largest reported radiotherapy series from a nonendemic area remains that from MD Anderson Cancer Center (MDACC) detailing the outcome of patients treated with radiation alone at that institution between 1954 and 1992.45,60 There were 378 patients, 85% were Caucasian or Hispanic, the remainder being Asian (8%), African American (5%), and Arabic (2%). Three-fourths of the patients presented with AJCC (1992) stage IV disease. The histology was 51% squamous carcinoma (including poorly differentiated tumors), 41% lymphoepitheliomas, and 8% unclassified. WHO type 2 cases are not discernible from this series as they are mixed with the squamous carcinomas and the lymphoepitheliomas (Dr. Fady Geara, Beirut, personal communication). The overall 5-, 10-, and 20-year actuarial rates of survival were 48, 34, and 18%, respectively, and the corresponding diseasespecific survival rates were 53, 45, and 39%. Primary site control: 5- and 10-year actuarial local control rates were 71 and 66%, the falloff between 5 and 10 years reflecting the propensity of this disease for late recurrence. A total of 100 patients experienced local tumor failure, of whom 17 had concurrent regional recurrence, 5 had distant metastases, and 9 had both regional and distant failure. Multivariate analysis revealed T stage, tumor differentiation, and cranial nerve palsies as independent prognostic determinants of local control. The 10-year local control rate was 79% for lymphoepithelioma compared to 54% for squamous carcinoma. As reflected in the new staging system, patients staged T4 on the basis of skull base involvement had a much better 10-year local control rate (50%) than those with cranial nerve palsies (26%). Neck control: the 5-, 10-, and 20-year regional control rates were 84, 83, and 83%, respectively. The great majority of patients with regional failure had concurrent local and/or distant failure with isolated regional recurrence in only nine (2.4%) cases. Distant metastases: The 5-, 10-, and 20-year rates of distant metastasis were 30, 32, and 32%, respectively. Multivariate analysis showed that the combination of nodal stage and level in the neck was the most potent determinant of distant metastasis, ranging from 10% for those with nodes less than 3 cm in size in the upper two-thirds of the neck to 65% for those with larger nodes involving the lower third of the neck. This is consistent with the current staging system. Although a large series, only patients in the last decade of that four-decade study period were CT staged, and the 1992 AJCC system was used. Additionally, this retrospective series did not classify patients by WHO type, given that many were diagnosed before this classification came into common use. There are now several large non-Caucasian series, staged by CT and the AJCC/UICC 1997 classification, detailing current results of treatment with radiotherapy. The largest is the Hong Kong experience reported by Lee et al.47 This series detailed 2687 CT-staged patients, the vast majority with WHO type 3 histology. There were 7% stage I, 41% stage II, 25% stage III, and 28% stage IVA –IVB patients. Radiotherapy alone to a median dose of 66 Gy was used to
122
HEAD AND NECK CANCER
treat 77% of patients. The 5-year local, nodal, and distant failure-free survival (FFS) rates for all patients were 85, 94, and 81%, respectively. The progression-free survival and overall survival rates for patients in stages I, II, III, and IV were 85 and 90%, 73 and 84%, 62 and 75%, and 44 and 58%, respectively. These two series demonstrate significantly different results and typify the inherent difficulties in comparing series over very different time frames. The differences in staging investigations, staging systems, pathologic classification, and treatment techniques are all relevant to the differing outcomes. However another very relevant and interesting point is the different histologies in the two series. There is general agreement that, stage for stage, patients with WHO types 2 and 3 NPC do better than patients with WHO type 1 tumors, especially as regards local control with radiation.61 Historically, however, WHO types 2 and 3 have been associated with greater risk for distant recurrence. It has also been assumed that better outcomes from endemic areas are at least in part due to the very low incidence of WHO type 1 in those series. It is not clear whether there is a difference in treatment outcome between Asian and non-Asian NPC patients with the same histology. A retrospective analysis of North American NPC patients with Chinese ancestry (parents born in China, Hong Kong, or Taiwan) compared to non-Chinese was performed by Su and Wang.62 The series included 131 non-Chinese and 41 Chinese patients treated at one institution with definitive radiation between 1979 and 1996. Only 20% of patients received any form of chemotherapy. Patients were staged according to the AJCC/UICC 1997 classification with data from their examination and CT imaging. Chinese patients were younger, more likely to have stage IV disease and WHO type 3 histology, and less likely to have ongoing tobacco exposure than non-Asians. In a multivariate analysis that controlled for stage, age, WHO type, and treatment, Chinese patients had a fourfold increased risk of distant metastases. Race did not predict overall survival or local control, though there was a trend to worse disease-specific and overall survival in the Chinese patients. Independent of race, WHO type 3 histology was associated with better local control and survival. In a recent series from Peter MacCallum Cancer Centre, Australia, treatment outcomes of Caucasian (born in Europe, Australia, Middle East, and Pacific Islands) and Asian (born in southern China or southeast Asia) NPC patients were compared.48 All patients had WHO types 2 or 3 disease and were staged using the 1997 UICC/AJCC criteria. The mean potential follow-up time was 9.6 years (range 1.0–18.5 years). There were 158 patients: 86 Asian and 72 Caucasians. Stage groupings were: I – 12 patients; II – 32 patients; III – 59 patients; and IV – 55 patients. Ninetynine percent of patients had staging CT and/or MRI. Female sex, age <45 years, and performance status of zero were more commonly observed in Asian patients. Other putative prognostic factors were not significantly different between the groups. Treatment consisted of radiotherapy alone in 30% (early stage disease), and chemotherapy and radiotherapy combinations in the remainder (locoregionally advanced disease), the majority receiving neoadjuvant chemotherapy
and concurrent chemoradiation. There were no significant differences in treatment between the two groups (although this may have reflected small case numbers). The 5-year rates for freedom from local recurrence (FLR), FFS, and overall survival (OS) for Asian and Caucasian patients were 74 versus 82%, 61 versus 55%, and 75 versus 63%, respectively. Corresponding 10-year figures were: 62 versus 82%, 43 versus 48%, and 58 versus 49%. There were no significant differences in FFS or OS between Asian and Caucasian patients, perhaps again reflecting the case numbers. However, the FLR interval was significantly worse in the Asian group (HR 2.37; 95% CI 1.11–5.06), while duration of freedom from distant metastasis tended to be better (HR 0.71; 95% CI 0.33–1.53). While this study provides no evidence that race is an independent prognostic factor for overall survival in patients with WHO types 2 and 3 NPC, it does suggest that relapse patterns may vary, with a higher rate of late primary failures (offset by a lower rate of distant failure) in the Asian population. Further confirmatory studies with larger patient cohorts are indicated. Comparison of the impact of race in patients with WHO type 1 NPC has been difficult given the rarity of this histological subtype in Asian patients. Marks et al. analyzed cases from the National Cancer Data Base (USA) diagnosed between 1985 and 1994.63 Their data showed the expected predominance of type 1 histology in non-Asians, and better survival for patients with type 2 and 3 tumors. However, no independent association of survival with race, independent of the histologic type, was identified. Using the SEER database from the US, a retrospective matched analysis of Caucasian and Asian patients with NPC was performed by Bhattacharyya.64 Pairs were matched for age, gender, AJCC stage, WHO type, and treatment modality. Of 171 matched pairs studied, 45% (77 in each group) of patients were WHO type 1 with 47 and 8% WHO types 2 and 3, respectively. The majority (82%) were stage III or IV. While overall survival was higher in those of Chinese ethnicity, no differences in disease-specific survival, overall and stratified by stage, were identified. Given the differences in racial definitions, it is perhaps not surprising that the data discussed above are contradictory in many aspects. Overall, they do not support an independent effect of race on survival of patients with NPC. Further investigations will be necessary to determine if the race influences relapse patterns in a consistent manner independent of stage, WHO type, and treatment.
Radiotherapy: Philosophy and Technique The propensity of cancer of the nasopharynx to metastasize and spread locally beyond the confines of the nasopharynx mandates large treatment volumes for all stages of disease. The primary tumor may extend anteriorly into the nasal cavity, superiorly into the floor of the sphenoid sinus or through the foramen lacerum into the cavernous sinus, anterosuperiorly into the posterior ethmoid air cells and orbits, laterally into the parapharyngeal space and sphenopalatine fossa, and inferiorly into the oropharynx (see Figure 8). Lymphatic spread most commonly involves the jugular chain of lymphatics and the posterior cervical chain. In addition,
NASOPHARYNGEAL CARCINOMA IN NON-ENDEMIC POPULATIONS
123
Figure 8 Potential routes of spread of primary tumor.
retropharyngeal nodes may be involved (see Figure 7). These lymphatic pathways are included in the target volume for all stages of disease. Delineation of the primary target volume is based on the extent of disease determined by clinical and radiologic evaluation. The latter should include both transverse and coronal CT cuts. As in all radiotherapy, the objective of treatment is to deliver a dose to the target volume tailored to the extent of disease present while respecting the tolerance of the normal tissues irradiated. Achieving this objective is technically difficult. The use of intensitymodulated radiotherapy (IMRT) has been a very exciting development in radiotherapy in general, and in the treatment of NPC in particular. Treatment Plan
Although standard treatment planning with parallel opposed fields and an off-cord reduction can provide good coverage of the planning target volume (PTV) for early stage disease, it produces unnecessary morbidity associated with irradiation of the major salivary glands, and the gross tumor volume (GTV) coverage is generally poor in locally advanced disease. A diagram detailing its use is given for historical context (see Figure 9), since it has given way to 3-D conformal treatment using parotid sparing techniques. The most sophisticated of these conformal techniques is IMRT. IMRT enables virtually any shape of dose distribution to be achieved; however, successful implementation requires both sophisticated equipment and a major commitment of time from both the oncologist (to define both PTVs and normal tissues to be spared on cross-sectional imaging) and the physicist to provide proper quality assurance. The technique used at the Peter MacCallum Cancer Centre is described below: Patients are immobilized supine in a cast, usually in a hard palate vertical position, and 3–5-mm CT slices are taken from the skull vertex to well below the sternal notch. These upper and lower levels need to be well beyond the treated area if noncoplanar fields are to be used. On the relevant axial slices the radiation oncologist outlines the gross disease and areas at risk of subclinical disease, particularly the draining
Figure 9 Simulation film for the previous standard plan of treatment of a patient with NPC stage T2aN1 included as historical reference (see text).
regional lymph nodes. Fusion of the CT images with MRI images is used to further improve the delineation of disease, particularly in T4 disease. IMRT allows for considerable flexibility in delineation of PTVs and their radiation dose prescription compared to the classical shrinking field technique. Our current standard is to demarcate all gross disease and add a 3–5-mm margin in three dimensions to the primary GTV to generate the PTV 70 Gy. We accept minus 5% prescribed dose within the anatomical boundaries of the nasopharynx, aiming for the gross disease to receive 70 Gy. In T1–2 disease, the upper level of the PTV reaches the floor of the sphenoid sinus. In more advanced primaries, IMRT is not magic and some compromise in margin (and dose) may be required to respect the brain stem and optic nerve/chiasm dose tolerance of 54 Gy in 2 Gy fractions. Nevertheless, when gross disease is closely applied to these structures, we increase the tolerance dose constraint to 59 Gy when given in 1.7 Gy fractions over 7 weeks (i.e. biologically equivalent dose to 54 Gy in 2 Gy fractions using an α : β ratio of 2). The margin applied to gross nodal disease is usually greater, 5–7 mm, where there is absence of critical dose limiting structures. The common exception to this is the lateral margin around the upper cervical and jugulodigastric nodes, where a tighter margin improves parotid sparing dose. An important consideration when large nodal disease extends into the lower neck/supraclavicular fossa is the brachial plexus. We demarcate this structure to be in the plane of the anterior border of the sixth (and lower) transverse vertebral
124
HEAD AND NECK CANCER
process(es), lateral to the external jugular vein. We limit the brachial plexus dose to 66 Gy. Beyond the PTV 70 Gy, we recommend defining a PTV 63 Gy to volumes adjacent to gross disease and to nodes that are not definitely involved but are at high risk (e.g. retropharyngeal, jugulodigastric and high posterior cervical nodal regions). A dose of 56 Gy over 7 weeks is adequate for other lower risk nodes treated electively. We have favored junctioning the IMRT plan at an anterior neck field with midline shielding as a mechanism for reducing the radiation dose to midline structures (larynx, pharyngo-esophageal axis). It is important to realize the complexities of junctioning an IMRT plan with an anterior lower neck field and the subsequent dose uncertainties at that junction. Another option is to utilize the IMRT plan for the whole treatment volume with the larynx and pharyngooesophagus outlined as structures to be spared. This hasn’t been an entirely satisfactory solution as current IMRT planning systems have a limited number of structures that can be denoted, and the priority of this avoidance structure is low enough for the dose to be significantly higher compared to an anterior field with midline shielding. Critical normal structures (optic nerve/chiasm, brain stem/spinal cord) and life quality structures (parotid glands, inner ear) all need to be demarcated. Our demarcation of the inner ear (essentially the cochlea) is a 5-mm circle, medial to the tympanic membrane, in the temporal bone. It is also necessary to outline structures such as the brain, oral cavity, and larynx to avoid inappropriate “dumping” of dose in these structures. An optimal plan is then developed using IMRT software, usually involving 7–10 fields (see Figure 10). Careful and critical assessment of such plans is required, aided by production of dose volume histograms (DVH) for each of the GTV, PTV, spinal cord, brain stem, optic nerve/chiasm, parotid glands, and inner ear. Results with IMRT The largest series of NPC patients treated with IMRT from nonendemic areas are from North America. Lee et al. reported on 67 patients, 55 (82%) of whom were Chinese, treated at the University of California at San Francisco.65 WHO types 2 and 3 histology were
Figure 10 IMRT beam arrangement and dose cloud.
evenly distributed with no WHO type 1 patients in the series. Concurrent and adjuvant chemotherapy was given to 75% of patients. The prescribed dose was 65–70 Gy to the GTV, 60 Gy to the clinical target volume (CTV, i.e. GTV plus a margin of potential microscopic spread), and 50–60 Gy to the clinically negative neck. Twenty-six patients were treated with a brachytherapy boost and one with γ knife radiosurgery following external beam treatment. The disease was in AJCC/UICC 1997 stage I in 8, stage II in 12, stage III in 33, and stage IV in 25 patients. T3 and T4 disease was identified in 15 and 14 patients, respectively (43%). Median follow-up was relatively short at 31 months, but the 4-year locoregional control was excellent at 98%, the 4-year overall survival was 88%. The median parotid dose was 34 Gy, but this still resulted in only grades 0–1 xerostomia at 24 months in 98% of the 41 evaluable patients. This effect of parotid sparing has been confirmed by preliminary results from a randomized trial of IMRT compared with conventional fractionation in endemic early stage NPC.66 A second North American series from the Memorial Sloan Kettering Cancer Center reports on 74 patients: 43% were Asian; 5% had WHO type 1, 30% WHO type 2, and 65% WHO type 3 histology; 6% stage I, 16% stage II, 30% stage III, 47% stage IV, and 43% T3–4.67 Concurrent and adjuvant chemotherapy was given to 93% of patients. IMRT was given with accelerated fractionation, either concomitant boost or dose painting to 70 Gy. Median follow-up was again relatively short at 35 months. The 3-year rates of local and regional control are 91 and 93%, respectively. The rate of local control was not significantly different from the historical control of 79% at this institution for patients treated before 1998 with 3-D planning. Local control was 100% for T1–2 and 83% for T3–4, with 5/6 local recurrences in the target volume. Progression-free and overall survival were 67 and 83%, respectively. The authors note that an expectation for improved local control of advanced T stage disease with IMRT may be na¨ıve if dose to the target volume is maintained at 70 Gy. Given that data with IMRT are in evolution, it is quite reasonable to utilize conventional conformal radiotherapy for the treatment of NPC. At the Peter MacCallum Cancer Centre
NASOPHARYNGEAL CARCINOMA IN NON-ENDEMIC POPULATIONS
125
Figure 11 Boomerang – two Arc fields. The Tri-Arc – three small arcs in each major arc.
various conventional conformal radiotherapy techniques have been used and the one that dosimetrically best approximates IMRT is the “Boomerang” technique, recently modified to become the “Tri-Arc” technique.68 (see Figure 11) The shape of the Boomerang isodose curve nicely approximates the extent of disease commonly seen in locally advanced nasopharyngeal cancer. The posteromedial extent of disease represents a “cold area” with standard techniques, typically receiving almost 30% less dose in the off-cord treatment phase. In the Boomerang technique, the patient is immobilized in a prone position, to obviate treatment through the treatment couch rails. The optimal patient position is one that places the posterior clivus (brain stem) in line with the spinal cord, as seen on the image intensifier (usually hard palate vertical). This allows two asymmetric arcs to rotate around the spinal cord/brain stem, typically from 340 to 80◦ and 280 to 20◦ .
Figure 12 Tri-Arc and IMRT plans for the same patient.
Dividing these two major asymmetric arcs into three minor arcs of 30–35◦ (i.e. the Tri-Arc technique), allows the dose to the inner ear to be reduced in two ways. Firstly, additional shielding can be introduced on each of the minor arcs. Secondly, the width of each minor arc can be varied to better conform the dose to the PTV. In the Boomerang technique, the major arc has to be wide enough to cover the PTV in all dimensions. With three minor arcs, the width of each arc can vary appropriately. This allows a 5–10% reduction from the prescribed dose, and is particularly relevant when treating to doses of 70 Gy with concurrent cisplatin. Using the arc technique for the entire treatment enables very good parotid sparing as the arc isodoses reduce laterally. Comparison of dosimetry for a patient planned with both the Tri-Arc and IMRT is shown in Figure 12. Centers that can treat through the treatment couch (or use modifications as we now do with a head and
126
HEAD AND NECK CANCER
neck localizer couch extension) can use this technique with patients in the supine position. Radiation Dose and Fractionation There are no randomized trials specifically addressing the question of optimal radiation dose in NPC. As mentioned above, the standard recommended doses (derived empirically) are 70 Gy in 2 Gy fractions to areas of gross disease, and 50 Gy in 2 Gy fractions to uninvolved nodal regions. Given the tight margins that are required around gross disease, we recommend a “buffer zone” of 60 Gy transitioning into the 50 Gy volume. When IMRT is used, it is very labor intensive to use a shrinking field technique. We therefore treat all volumes in 35 fractions over 7 weeks using differential dose per fraction to provide biologically equivalent doses, that is, 63 Gy and 56 Gy at 1.8 and 1.6 Gy per fraction respectively. With the increased availability of IMRT, it is an opportune time for a randomized trial to establish the optimal radiation dose required in the context of defined WHO histology, a common staging system, and use of concurrent chemotherapy. The question of whether altered dose fractionation is of benefit in treating NPC is unclear. Teo et al., randomized 159 patients with WHO type 3 NPC to receive either conventional radiotherapy (60 Gy in 2.5 Gy fractions over 30 days) or conventional/accelerated radiotherapy (20 Gy in 2.5 Gy fractions, a further 51.2 Gy using 1.6 Gy twice daily, total 71.2 Gy in 40 fractions over 31 days).69 This study did not show improved locoregional control or overall survival in the higher dose arm. However only half the planned accrual was achieved before the study was terminated because of unacceptable neurological toxicity in the high-dose arm. Both retrospective and prospective studies by Lee et al.70 suggested improved local control in patients with stage IV(A-B) disease treated with accelerated fractionation (66 Gy in 2 Gy fractions, delivered 6 days per week) compared to a historical control group treated with conventional fractionation. The prospective phase II study included induction chemotherapy with cisplatin and 5-fluorouracil and concurrent chemoradiation with intermittent cisplatin. At 3 years, locoregional and distant failure-free rates were 77 and 75% respectively, and overall survival was 71%. On the basis of these promising results, a subsequent 2 × 2 randomized study comparing conventional radiotherapy, accelerated radiotherapy, each with or without chemotherapy, was begun. However, the trial was closed prematurely due to poor accrual, without definitively answering the questions posed.
Role of Chemotherapy The current standard of care for locoregionally advanced NPC in North America was established by the Intergroup 0099 trial, a randomized comparison of radiation alone (70 Gy) versus concurrent chemoradiation with intermittent high-dose cisplatin for three cycles followed by adjuvant chemotherapy with cisplatin and 5-fluorouracil (5-FU) for three cycles.71 Patient eligibility included those with stage III/IV disease by the AJCC 1992 classification. Accrual to this trial was stopped prematurely when an interim analysis indicated significant benefit for survival in the chemotherapy arm. The study is summarized in Tables 7a –b. This trial was
criticized for a high rate of ineligibility, a relatively poor outcome with radiation alone, and early closure leading to only 69 and 78 eligible patients in control and experimental arms, respectively. Additionally, 30% of patients in this study had WHO type 1 histology, and there was substantial doubt that this approach would be beneficial in endemic NPC, which typically is composed of more than 95% WHO types 2 and 3, and thus, in general, is more radiosensitive. In the last 3 years, five randomized trials of chemoradiation versus radiation have been completed in patients with endemic NPC (Table 7a –b).72 – 76 Three of these utilized cisplatin as a single agent during radiation, either weekly or intermittently; one cisplatin and 5-FU; one oral tegafur and uracil. Adjuvant chemotherapy was included in two of the trials. Only one study replicated the treatment from the experimental arm of the Intergroup 0099 trial75 and this study is one of two showing statistically significant improvements in both progression-free and overall survival for chemoradiation. In this trial the benefit of chemotherapy was manifest in improved control of distant disease; locoregional control was identical and >90% in both arms. The second trial by Lin et al.73 included only concurrent treatment with cisplatin and 5-FU and showed improvements in both locoregional and distant control. Of the other three trials, one utilized only concurrent cisplatin intermittently, and this study showed an 11% improvement in locoregional control at 3 years but no impact on survival.74 The other two trials showed 11 and 9% survival improvements at 3 and 5 years, respectively72,83 but only nonsignificant trends to improved progression-free survival. Interestingly, a retrospective subset analysis for T stage, Ho’s T1–2, compared to Ho’s T3 in the former trial by Chan et al. showed significant improvement in progressionfree survival and distant control for Ho’s T3 patients treated with concurrent cisplatin. In general, these five studies from endemic geographic areas demonstrated better outcomes for patients receiving radiation alone and lesser degree of impact on survival and other endpoints than was observed in the North American Intergroup 0099 study. Prior to the era of chemoradiation brought about by the results of Intergroup 0099, neoadjuvant and adjuvant chemotherapy had been studied in seven randomized trials. These were all negative for a survival impact, though one neoadjuvant trial showed improved progression-free survival for the chemotherapy arm.77 A recent meta-analysis included these and three of the chemoradiation trials described above with a total of 10 studies and 2450 patients.78 Both neoadjuvant chemotherapy and concurrent chemoradiation resulted in significant improvements in locoregional and distant control, while adjuvant chemotherapy did not. The hazard ratio for death in all studies was 0.82 (p = 0.01), corresponding to an absolute survival benefit of 4% after 3 years. The use of chemoradiation was associated with a hazard ratio of 0.48 (p = 0.004) and a survival improvement of 20% after 3 years. Neoadjuvant and/or adjuvant chemotherapy did not affect overall survival. While the data from the meta-analysis recapitulate that obtained from similar studies in non-NPC squamous cancer of the head and neck, the findings are perhaps, counterintuitive given the relatively high risk of distant metastasis for WHO types 2 and 3 NPC.
NASOPHARYNGEAL CARCINOMA IN NON-ENDEMIC POPULATIONS
127
Table 7a Randomized chemoradiation trials in NPC – description.
Author
N
Al-Sarraf81
147
Chan72
350
Lin73
284
Lee74
348
Wee75
221
Kwong76
222
Eligibility Stages III/IV AJCC 1992 Ho’s N2 – 3 or N ≥ 4 cm
WHO type 1 (%) 30 1
Chemotherapy Concurrent
Adjuvant −2
CDDP 80 mg m × 1d FUb 1000 mg m−2 × 4d
CDDPa 100 mg m−2 × 3 CDDP 40 mg m−2 weekly
Stages III/IV AJCC 1992 Stages III/IV AJCC/UICC 1997
3
CDDP 20 mg m−2 × 4d FU 400 mg m−2 × 4d
0
CDDP 100 mg m−2 × 3
Stages III/IV AJCC/UICC 1997 Ho’s T3 or N2 – 3
0
CDDP 25 mg m−2 × 4 d ×3
1
UFTd 600 mg d−1
×3
Radiation (Gy) ×3
70 66 ± 10 – 20 (boost)
–
70 – 74
–
66 ± 10 (boost)
– CDDP 20 mg m−2 × 4d FU 1000 mg m−2 × 4d
×3
CDDP/FUc Alternating with VBM
70 66 – 68 ± 10 (boost)
a
CDDP, cisplatin. FU, 5-fluorouracil. Four-arm factorial study, two arms had adjuvant chemotherapy consisting of alternating cisplatin/5-fluorouracil and vincristine/bleomycin/methotrexate. d UFT, uracil and tegafur. Modified from Lee N and O’Meara WP. Advances in nasopharyngeal carcinoma. Curr. Opin. Oncol. 2005; 17: 225 – 30 Lippincott Williams & Wilkins. b c
Table 7b Randomized chemoradiation trials in NPC – results.
Author
Time point
Treatment arm
Locoregional control %
Distant control %
Progression-free survival %
Overall survival %
Al-Sarraf81
5
CRTa RTb
89 74
87 65
58 29
67 p < 0.001 37
Chan72
5
CRT RT
93 92
79 74
60 p = 0.16 52
70 p = 0.05 59
Lin73
5
CRT RT
88 70
81 71
72 p = 0.001 53
72 p = 0.002 54
Lee74
3
CRT RT
93 82
75 72
69 p = 0.24 61
78 p = 0.76 79
Wee75
3
CRT RT
92 91
84 66
80 p = 0.009 65
85 p = 0.006 78
Kwongc76
3
CRT RT
80 72
85 71
69 p = 0.14 58
86 p = 0.06 77
a b c
CRT, chemoradiation. RT, radiation. Data are reported for two combined arms treated with CRT and two with RT alone (± adjuvant chemotherapy).
Despite the lack of uniformity in trial results, there is general agreement at this time that the results of Intergroup 0099 are applicable to both sporadic and endemic NPC that is locally or regionally advanced. However, the conventional wisdom that concurrent treatment improves locoregional control and neoadjuvant and/or adjuvant chemotherapy reduces the risk of distant disease is not borne out in these randomized trials. This suggests that risk-based approaches, as reported by investigators at MDACC, while attractive from the standpoint of minimizing toxicity and individualizing therapy, will need to be more sophisticated than the simple consideration of T and N status.79,80 Further, although the Intergroup 0099 trial has popularized the use of adjuvant cisplatin/5-FU, it is widely recognized that compliance is quite low with this regimen because of toxicity, especially in the postchemoradiation setting. Neoadjuvant chemotherapy
with taxanes and epirubicin in doublets with a platinating agent has been reported and is the subject of current research. These regimens, devoid of infusional 5-FU, appear as effective as cisplatin/5-FU based on response rates, are logistically less complex, and associated with reduced mucosal toxicity.49,81,82 Given its increased feasibility and improved compliance relative to adjuvant chemotherapy, neoadjuvant treatment with these newer regimens may optimize control of distant disease. As locoregional control improves, prevention of distant failure is becoming increasingly important, especially for WHO types 1 and 3 NPC. Another potential advantage to the neoadjuvant, as opposed to adjuvant, approach is the ability to cytoreduce advanced T stage disease that encroaches on critical neighboring structures, such as brain, brain stem, and optic tracts, facilitating the safety of subsequent chemoradiation.
128
HEAD AND NECK CANCER
Pretreatment Evaluation and Follow-up Every patient should have an audiogram and a careful dental evaluation before treatment. Teeth that show signs of decay or periodontal disease either need to be restored or ablated. Every patient is submitted to a thorough prophylactic program.83 For patients with advanced disease in whom the pituitary and hypothalamus will be in the primary radiation beam, baseline endocrine assessments of hypophyseal function should be done prior to commencement of radiotherapy. This permits early identification of hormonal deficiencies and initiation of appropriate replacement therapy.84 Titers to Epstein-Barr viral antigens may be of value where facilities are available, in view of the application of such data to the prognosis of the disease.17,40 After completion of therapy, patients should be examined at regular intervals, every 3 months for the first 2 years, every 4 months for the third year, and every 6 months through the fifth year after treatment. A follow-up evaluation is done every year thereafter. During follow-up, the disease status is evaluated for recurrence at the primary site or the neck nodes and for clinical signs of distant metastases. Attention should also be directed to possible sequelae of treatment and prevention of infectious complications in the head and neck area. The external auditory canals will be deficient in normal cerumen production and all patients should be instructed about prevention of external otitis. The auditory canal will be dry and the normal migration of the epithelium within the auditory canal is impaired. As a result, debris tends to collect and may impact. Patients should be advised to avoid manipulation of the ear canal and to seek medical advice promptly if irritation develops. Patients with carcinoma of the nasopharynx often present with serous otitis media. After treatment, this may subside but may persist in a chronic form. If this sequela causes bothersome symptoms, it can be managed with indwelling tympanic membrane ventilation tubes, although this has to be weighed against the subsequent high risk of chronic otitis media. The incidence of chronic hearing loss following treatment of NPC has generally been underappreciated, as there have been few prospective studies. The need for permanent hearing aids must be assessed and they must be recommended as required in the follow-up of these patients. Dental and oral cavity care should be meticulous and the application of fluoride solutions in the form of stannous fluoride or sodium fluoride by custom-fitted carriers should be done routinely. There are products now available, such as high–fluoride-containing toothpastes (e.g. Neutrofluor 5000 TM ) and DentacalTM (casein phosphate peptide amorphous calcium phosphate), that protect against the deleterious effects of xerostomia on dentition and should be recommended prophylactically. Dental extractions after radiation therapy should be avoided whenever possible. If extractions are unavoidable, extreme precautions are necessary to minimize the risk of osteoradionecrosis.47 Radiation to the temporomandibular joints and masticatory muscles, especially in patients who receive systemic chemotherapy, may cause trismus. This usually does not set in before 3–6 months after treatment, but is progressive. Patients, especially teenagers and young adults, need to be instructed about
its prevention and encouraged to do active jaw exercises. The effect of irradiation on the mucosa of the sinonasal tract is a metaplastic transformation of the epithelium from ciliary columnar respiratory epithelium to cuboidal or squamous stratified epithelium with loss of ciliary function and very often loss of the mucous-secreting elements. In spite of these changes, it is rare that patients experience sinonasal infections. Nevertheless, if they do occur they should be treated promptly and aggressively to avoid undesirable necrosis of soft tissue and osteoradionecrosis of facial bones. Irradiation of part or the entire pituitary hypothalamic axis is unavoidable when there is cancerous bony invasion of the base of the skull. In such patients, an annual evaluation of the pituitary function and of the thyroid and pituitary adrenal axis is therefore recommended. Proper replacement therapy should be tailored to identified deficiencies.84
MANAGEMENT OF RECURRENT OR METASTATIC NASOPHARYNGEAL CARCINOMA Radiotherapy Locally recurrent NPC, especially if limited to the primary site, without intracranial extension, should be considered for retreatment with radiotherapy. Pryzant et al.85 reported the results of retreatment of 53 patients with megavoltage irradiation at MDACC between 1954 and 1989. Overall 5-year actuarial local control was 35%. Much better results were achieved in a subset of nine patients with recurrent disease confined to the nasopharynx in whom treatment with an intracavitary brachytherapy boost was possible: seven of these patients achieved durable local control. In this series, 8 of 53 patients sustained severe complications of retreatment, which were fatal in 5. The most significant factor predicting for severe complications was a total cumulative dose of external beam therapy greater than 100 Gy. Similar results were recently reported by Fu et al.82 in a series of 74 patients re-treated at the University of California at San Francisco between 1957 and 1995. Overall local–regional progressionfree survival was 40%. Significant factors predicting for local–regional control were histologic type (WHO type 1 worse), time to diagnose a recurrence (longer than 5 years best), and use of brachytherapy in patients with disease confined to the nasopharynx. Complications were significantly increased in patients who received cumulative doses greater than 120 Gy. The definitive study of risk factors for complications of retreatment is from Hong Kong, where Lee et al.86 reported on a series of 654 patients retreated between 1976 and 1992. Of these, 539 received external beam therapy alone. The biologically effective dose (BED) of the initial treatment and to a lesser extent the retreatment BED were significant determinants of risk; interestingly, there was no evidence that the time between treatments influenced residual tolerance. In all series, there is clearly less morbidity when a component of brachytherapy is used, reflecting the smaller volume of tissue receiving a high retreatment dose. The limitation of brachytherapy in treating disease that extends beyond the nasopharynx can be partially overcome by using modern stereotactic conformal techniques or heavy particle
NASOPHARYNGEAL CARCINOMA IN NON-ENDEMIC POPULATIONS
therapy.87 – 89 No long-term results of retreatment with conformal photon beam therapy have yet been published. While it is virtually certain that these techniques will reduce morbidity, there is also a greater risk of geographic miss of tumor, thus limiting overall salvage rates – for example, in the series treated with heavy particles the long-term local control rate was 45%, comparable to that achieved with conventional treatment.68 Radiotherapy also has a role in the palliative treatment of regional or distant metastatic sites in patients with incurable disease. The most common indication is for painful bony or liver metastases.
Surgery Technical advances in skull base surgery along with better imaging to define the extent of recurrent disease make salvage surgery also an option for patients with localized recurrence. A variety of approaches have been described but fall into three main groups: inferior/inferolateral,90 – 92 lateral,93,94 and anterolateral.95 Long-term control rates averaging 38% have been reported in recently published series,92,95 a figure that compares favorably with retreatment by radiotherapy. However, it is likely that a greater degree of selection is applied to patients being considered for surgical resection than those for re-irradiation. Isolated neck recurrences after treatment of NPC are rare (especially for WHO types 2 and 3). However, for patients who do relapse regionally, and in whom no distant disease can be demonstrated, neck dissection is indicated and can be curative. For example, Wei et al. reported on 51 patients who underwent radical neck dissection for persistent or recurrent neck disease following radiotherapy.96 Actuarial 5-year survival and neck control rates were 38 and 66%, respectively. More recently, Yen et al.97 published outcome data on 31 patients undergoing salvage neck surgery in Taiwan over a 14-year period (emphasizing the rarity of the condition). In this series, overall 5-year survival after neck dissection was 67%.
Systemic Therapy for Recurrent Disease For recurrence in the neck and distant metastatic sites, with or without failure at the primary site, systemic chemotherapy is generally indicated. Because NPC is uncommon in areas of sporadic incidence and, until recently, chemotherapy was not widely used in the endemic areas there is little information on the role of chemotherapy specific to the management of metastatic nasopharyngeal cancer. In older reports, patients with nasopharyngeal cancer were subsumed into series that included all patients with head and neck cancer, where they constituted only a small minority. More recently, several chemotherapeutic agents used primarily in combination treatment have been tested. Over the past 10 years, investigators from the Institut Gustave Roussy have conducted a series of phase II trials of chemotherapy for metastatic and/or locoregionally recurrent disease using various combinations of consolidative radiation after achieving a complete response to bleomycin, epirubicin, cisplatin, and 5-FU. Overall response rates between 46 and 75%, have been reported.98
129
Newer chemotherapy regimens including paclitaxel or gemcitabine in a doublet with either cis- or carboplatin have so far yielded similar response rates.99,100 In the experience of the group at the Institut Gustave Roussy, a small subset of patients (approximately 10%) were long-term survivors (disease-free for 82+ to 190+ months) and appear to have been cured. Seventy-five percent of the long-term survivors had isolated bone metastases and many of them were treated with consolidative radiation after a complete response to chemotherapy. Because the prognosis for most patients with recurrence, especially distant metastases, is quite poor, the investigation of molecular-targeted agents is attractive. Two series have shown that expression of the epidermal growth factor receptor (EGFR) is observed in >80% of endemic NPC; though prognostic import to expression was not consistent.101,102 A trial with cetuximab, a monoclonal antibody to EGFR, in combination with carboplatin in patients with recurrent disease previously treated with cisplatin demonstrated a partial response rate of 12%.103 This is nearly identical to response rates seen with cetuximab as a single agent, or combined with cisplatin or carboplatin in platin-refractory patients with squamous cancers of other head and neck sites and indicates the value of further investigation of EGFR-targeted approaches in NPC.104 Another obvious target for biologic therapy of NPC is the association with EBV. In a small series of 10 patients, Straathof et al. demonstrated the feasibility of developing ex vivo expanded EBV-specific cytotoxic T lymphocytes (CTL) from previously treated patients. Infusion of CTL was associated with disease regression in three of six patients evaluable for response, two of whom had durable complete remission.105 This approach clearly bears study in an expanded cohort. This immunologic approach may have the greatest impact in the adjuvant setting for patients with high risk of distant recurrence.
CONCLUSIONS Although nasopharyngeal cancer is a rare disease in Caucasians and other nonendemic populations, it is one in which there is strong evidence that the results of treatment are significantly influenced by the quality of medical care rendered. Results have improved dramatically over the past two decades and can confidently be predicted to improve further as a result of revolutions in medical imaging, radiotherapy technical capability, and understanding of the optimal way to combine radiotherapy and chemotherapy. To ensure that patients with NPC receive optimum treatment they should ideally be managed in a major cancer center where the experience and expertise to handle the complexities of the disease are available. Furthermore, they should, wherever possible, be enrolled in clinical trials designed to resolve unanswered questions surrounding the disease and its management. As the medical resources available to countries with high endemic rates of NPC increase, opportunities for international collaboration to test new treatment strategies in a timely way should not be lost.
130
HEAD AND NECK CANCER
REFERENCES 1. Michaux L. Carcinoma de base du crˆane. In Godfredsen E (ed) Ophthalmologic and neurologic symptoms of malignant nasopharyngeal tumours. Acta Psychiat Scand 1944; 34(Suppl. 1): 323. 2. Strouhal E. Ancient Egyptian case of carcinoma. Bull N Y Acad Med 1978; 54: 290. 3. Bosworth FH. A Treatise on Disease of the Nose and Throat, Vol. 1. New York: Wood, 1889. 4. Muir CS. Nasopharyngeal cancer – a historical vignette. CA Cancer J Clin 1983; 33: 180. 5. Batsakis JG, Solomon AR, Rice DH. The pathology of head and neck tumors: carcinoma of the nasopharynx, part II. Head Neck Surg 1981; 3: 511. 6. Michaels L, Hyams VJ. Undifferentiated carcinoma of the nasopharynx. A light and electron microscopical study. Clin Otolaryngol 1977; 2: 105. 7. Shanmugaratnam K, et al. Histopathology of nasopharyngeal carcinoma. Correlations with epidemiology, survival rates and other biological characteristics. Cancer 1979; 44: 1029. 8. Taxy JB, Hidvegi DF, Battifora H. Nasopharyngeal carcinoma: antikeratin immunohistochemistry and electron microscopy. Am J Clin Pathol 1985; 83: 320. 9. Weiland LH, Neel HB, Pearson GR. Nasopharyngeal carcinoma. Curr Hematol Oncol 1986; 4: 379. 10. Henle W, Henle G. Epidemiologic aspects of Epstein – Barr Virus (EBV-associated) diseases. Ann NY Acad Sci 1980; 354: 326. 11. Watson CRR. The anatomy of the post-nasal space: its significance in local malignant invasion. Australas Radiol 1972; 16: 118. 12. Parkin DM, Whelan SL, Ferley J, Raymond L, Young J (eds). Cancer Incidence in Five Continents, Vol VII, No. 143. Lyon: IARC, 1997. 13. Lee HP, et al. Recent trends in cancer incidence among Singapore Chinese. Int J Cancer 1988; 42: 159. 14. Levine PH, Connelly RR, Easton JM. Demographic patterns for nasopharyngeal carcinoma in the United States. Int J Cancer 1980; 26: 741. 15. Greene MH, Fraumeni JF, Hoover R. Nasopharyngeal cancer among young people in the United States: radical variations by cell type. J Natl Cancer Inst 1977; 58: 1267. 16. Easton JM, Levine PH, Hyams VJ. Nasopharyngeal carcinoma in the United States. A pathologic study of 177 US and 30 foreign cases. Arch Otolaryngol 1980; 106: 88. 17. Neel HB. A prospective evaluation of patients with nasopharyngeal carcinoma: a overview. J Otolaryngol 1986; 15: 137. 18. Wolf H, Haus M, Wilmes E. Persistence of Epstein – Barr virus in the parotid gland. J Virol 1984; 51: 795. 19. Lung ML, et al. Evidence that respiratory tract is a major reservoir for Epstein – Barr virus. Lancet 1985; 1: 389. 20. Old LJ, et al. Precipitating antibodies in human serum to an antigen present in cultured Burkitt’s lymphoma cells. Proc Natl Acad Sci USA 1966; 56: 1699. 21. Pathmanathan R, et al. Clonal proliferation of cells infected with Epstein – Barr virus in preneoplastic lesions related to nasopharyngeal carcinoma. New Engl J Med 1995; 333: 693. 22. Rajadurai P, et al. Undifferentiated, nonkeratinizing and squamous cell carcinoma of the nasopharynx: variants of EBV infected neoplasia. Am J Pathol 1995; 146(6): 1355. 23. Choi PHK, et al. Nasopharyngeal carcinoma: genetic changes, Epstein – Barr virus infection, or both. A clinical and molecular study of 36 patients. Cancer 1993; 72: 2873. 24. Hildesheim A, et al. CYP2E1 genetic pleomorphisms and risk of nasopharyngeal carcinoma in Taiwan. J Natl Cancer Inst 1997; 89: 12207 – 12. 25. Vaughan TL, et al. Nasopharyngeal carcinoma in a low-risk population: Defining risk factors by histological type. Cancer Epidemiol Biomarkers Prev 1996; 5: 587 – 93. 26. Goldsmith DB, West TM, Morton R. HLA association with nasopharyngeal carcinoma in Southern Chinese: a meta-analysis. Clin Otolaryngol 2002; 27: 61 – 7. 27. Lu SJ, et al. Linkage of a nasopharyngeal carcinoma susceptibility locus to the HLA region. Nature 1990; 346: 470 – 1.
28. Burt RD, et al. Associations between human leukocyte antigen type and nasopharyngeal carcinoma in Caucasians in the United States. Cancer Epidemiol Biomarkers Prev 1996; 5(11): 879. 29. Lo KW, Huang DP. Genetic and epigenetic changes in nasopharyngeal carcinoma. Semin Cancer Biol 2002; 12: 451 – 62. 30. Nicholls JM. Nasopharyngeal carcinoma: classification and histological appearances. Adv Anat Pathol 1997; 4: 71 – 84. 31. Krueger GRF, et al. Histological types of nasopharyngeal carcinoma as compared to EBV serology. Anticancer Res 1981; 8: 27. 32. Shanmugaratnam K, Sobin LH. Histological typing of tumours of the upper respiratory tract and ear. In Shanmugaratnam K, Sobin LH (eds) International Histological Classification of Tumours, 2nd ed. Genevea: WHO, 1991: 32 – 33. 33. Prathap K, Looi LM, Prasad U. Localized amyloidosis in nasopharyngeal carcinoma. Histopathology 1984; 8: 27. 34. Nomori H, et al. Histiocytes in nasopharyngeal carcinoma in relation to prognosis. Cancer 1986; 57: 100. 35. Moller P, et al. Lymphoepithelial carcinoma (Schmincke type) as a derivate of the tonsillar crypt epithelium. Virchows Arch A Pathol Anat Histopathol 1984; 405: 83. 36. Micheau C, et al. Lymphoepitheliomas of the larynx (undifferentiated carcinomas of nasopharyngeal type). Clin Otolaryngol 1979; 4: 43. 37. Rosai J. ‘Lymphoepithelioma-like’ thymic carcinoma. Another tumor related to Epstein – Barr virus? New Engl J Med 1985; 312: 1320. 38. Mills SE, Austin MB, Randall ME. Lymphoepithelioma-like carcinoma of the uterine cervix with inflammatory stroma. Am J Surg Pathol 1985; 9: 883. 39. Neel HB, et al. Applications of Epstein – Barr virus serology to the diagnosis and staging of North American patients with nasopharyngeal carcinoma. Otolaryngol Head Neck Surg 1983; 91: 225. 40. Zong YS, et al. Immunoglobulin A against viral capsid antigen of Esptein-Barr virus and indirect mirror examination of the nasopharynx in the detection of asymptomatic nasopharyngeal carcinoma. Cancer 1992; 69: 3 – 7. 41. Tune CE, et al. Nasopharyngeal brush biopsies and detection of nasopharyngeal cancer in a high-risk population. J Natl Cancer Inst 1999; 91(9): 796 – 800. 42. Chan AT, et al. Phase II study of neoadjuvant carboplatin and paclitaxel followed by radiotherapy and concurrent cisplatin in patients with locoregionally advanced nasopharyngeal carcinoma: Therapeutic monitoring with plasma EBV DNA. J Clin Oncol 2004; 22: 3053 – 60. 43. Neel HB. Nasopharyngeal carcinoma: clinical presentation, diagnosis, treatment and prognosis. Otolaryngol Clin North Am 1985; 18: 479. 44. Sawaki S, Sugano H, Hirayama T. Analytical aspects of symptoms of nasopharyngeal malignancies. In de-The G, Ito Y (eds) Nasopharyngeal Carcinoma: Etiology and Control, Vol. 147. No. 20. Lyon, France: IARC, 1978: 63. 45. Ng SH, et al. Nasopharyngeal carcinoma: MRI and CT assessment. Neuroradiology 1997; 39: 741 – 6. 46. Poon PY, Tsang VH, Munk PL. Tumour extent and T stage of NPC: a comparison of MRI and CT findings. Can Assoc Radiol J 2000; 51(5): 287 – 95. 47. Lee AW, et al. Treatment results for nasopharyngeal carcinoma in the modern era: the Hong Kong experience. Int J Radiat Oncol, Biol, Phys 2005; 61: 1107 – 16. 48. Corry J, et al. Relapse patterns in WHO 2/3 Nasopharyngeal cancer: is there a difference between ethnic Asian versus non-Asian patients? Int J Radiat Oncol, Biol, Phys 2006; 64: 63 – 71. 49. Vikram B, et al. Patterns of failure in carcinoma of the nasopharynx. Failure at distant sites. Head Neck Surg 1986; 8: 276. 50. Geara FB, et al. Carcinoma of the nasopharynx treated by radiotherapy alone: determinants of distant metastasis and survival. Radiother Oncol 1997; 43: 53. 51. Huang SC, Chu GL. Nasopharyngeal cancer: study 11. Int J Radiat Oncol Biol Phys 1981; 7: 713. 52. Forastiere AA, et al. Head and neck cancers. J Natl Compr Canc Netw 2005; 3: 316 – 91. 53. Kumar MB, et al. Tailoring distant metastatic imaging for patients with clinically localized undifferentiated nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2004; 58: 688 – 93.
NASOPHARYNGEAL CARCINOMA IN NON-ENDEMIC POPULATIONS 54. Chang JT, et al. Nasopharyngeal carcinoma staging by (18)Ffluorodeoxyglucose positron emission tomography. Int J Radiat Oncol Biol Phys 2005; 62: 501 – 7. 55. Ho JH. Stage classification of nasopharyngeal carcinoma: a review. IARC Sci Publ 1978; 20: 99. 56. Sobin LH, Wittekind C, (eds). TNM Classification of Malignant Tumours, 6th ed. New York: Wiley-Liss, 2002. 57. Greene FL, Page DL, Fleming ID, (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002. 58. Sham J, et al. Extent of nasopharyngeal carcinoma involvement inside the nasopharynx. Cancer 1992; 69(4): 854. 59. Lee A, et al. Staging of nasopharyngeal carcinoma: evaluation of Nstaging by Ho and UICC/AJCC systems. Clin Oncol 1996; 8: 146. 60. Sanguinetti G, et al. Carcinoma of the nasopharynx treated by radiotherapy alone: determinants of local and regional control. Int J Radiat Oncol Biol Phys 1997; 37(5): 985. 61. Reddy SP, et al. Prognostic significance of keratinization in nasopharyngeal carcinoma. Am J Otolaryngol 1995; 16: 103 – 8. 62. Su CK, Wang CC. Prognostic value of Chinese race in nasopharyngeal cancer. Int J Radiat Oncol Biol Phys 2002; 54: 752 – 8. 63. Marks JE, Phillips JL, Menck HR. The National Cancer Data Base report on the relationship of race and national origin to the histology of nasopharyngeal carcinoma. Cancer 1998; 83: 582 – 8. 64. Bhattacharyya Neil. The impact of race on survival in nasopharyngeal carcinoma: a matched analysis. Am J Otolaryngol 2004; 25: 94 – 7. 65. Lee N, et al. Intensity-modulated radiotherapy in the treatment of nasopharyngeal carcinoma: an update of the UCSF experience. Int J Radiat Oncol Biol Phys 2002; 53: 12 – 22. 66. Kam MK, et al. Impact of intensity-modulated radiotherapy on salivary gland function in early stage nasopharyngeal carcinoma patients (abstract 5501). Proc Am Soc Clin Oncol 2005; 23: 500s. 67. Wolden SL, et al. Intensity Modulated Radiation Therapy (IMRT) for nasopharynx cancer: update of the Memorial Sloan-Kettering experience. Int J Radiat Oncol Biol Phys 2006; 64: 57 – 62. 68. Corry J, et al. The “Boomerang” Technique: an improved method for conformal treatment of locally advanced nasopharyngeal cancer. Australas Radiol 2004; 8: 170 – 80. 69. Teo PM, et al. Final report of a randomized trial on alteredfractionated radiotherapy in nasopharyngeal carcinoma prematurely terminated by significant increase in neurologic complications. Int J Radiat Oncol Biol Phys 2000; 48: 1311 – 22. 70. Lee AWM, et al. Treatment of stage IV (A-B) nasopharyngeal carcinoma by induction-concurrent chemoradiotherapy and accelerated fractionation. Int J Radiat Oncol Biol Phys 2005; 63: 1331 – 8. 71. Al-Sarraf MML, et al. Chemo-radiotherapy vs radiotherapy in patients with locally advanced nasopharyngeal cancer: phase III randomized Intergroup study (0099) (SWOG 8892, RTOG 8817, ECOG 2388). J Clin Oncol 1998; 16: 1310. 72. Chan AT, Ngan R, Teo P. Final results of a phase III randomized study of concurrent weekly cisplatin-RT versus RT alone in locoregionally advanced nasopharyngeal carcinoma (Meeting abstracts) (abstract 5523). J Clin Oncol 2004; 22(14s): 5523. 73. Lin JC, et al. Phase III study of concurrent chemoradiotherapy versus radiotherapy alone for advanced nasopharyngeal carcinoma: positive effect on overall and progression-free survival. J Clin Oncol 2003; 21: 631 – 7. 74. Lee AW, et al. Prospective randomized study on therapeutic gain achieved by addition of chemotherapy for T1-4N2-3M0 nasopharyngeal carcinoma. (NPC) (Meeting abstracts) (Abstract 5506). J Clin Oncol 2004; 22(14s): 5506. 75. Wee J, et al. Randomized trial of radiotherapy versus concurrent chemoradiotherapy followed by adjuvant chemotherapy in patients with American Joint Committee on Cancer/International Union Against Cancer Stage III and IV nasopharyngeal cancer of the endemic variety. J Clin Oncol 2005; 23: 6730 – 8. 76. Kwong DL, et al. Concurrent and adjuvant chemotherapy for nasopharyngeal carcinoma: a factorial study. J Clin Oncol 2004; 22: 2643 – 53. 77. International Nasopharynx Cancer Study Group. Preliminary results of a randomized trial comparing neoadjuvant chemotherapy (cisplatin,
78.
79.
80.
81.
82.
83.
84.
85. 86.
87. 88.
89.
90.
91. 92.
93.
94. 95.
96.
97. 98.
99.
100.
131
epirubicin, bleomycin) plus radiotherapy vs. radiotherapy alone in stage IV undifferentiated nasopharyngeal carcinoma: a positive effect on progression-free survival. Int J Radiat Oncol Biol Phys 1996; 35: 463 – 9. Langendijk JA, et al. The additional value of chemotherapy to radiotherapy in locally advanced nasopharyngeal carcinoma: A metaanalysis of the published literature. J Clin Oncol 2004; 22: 4604 – 12. Johnson FM, et al. A phase II study of docetaxel and carboplatin as neoadjuvant therapy for nasopharyngeal carcinoma with early T status and advanced N status. Cancer 2004; 100: 991 – 8. Johnson FM, et al. A phase I/II study of neoadjuvant chemotherapy followed by radiation with boost chemotherapy for advanced T-stage nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2005; 63: 717 – 24. Rischin D, et al. Excellent disease control and survival in patients with advanced nasopharyngeal cancer treated with chemoradiation. J Clin Oncol 2002; 20: 1845 – 52. Fu KK, Hwang JM, Phillips T. Re-irradiation of locally recurrent nasopharyngeal carcinoma. In Proceedings of the UICC Workshop on Nasopharyngeal Cancer, Singapore, February 11 – 14 1998: 173 – 87. Daly T. Dental care in the irradiated patients. In Fletcher GH (ed) Textbook of Radiotherapy, 3rd ed. Philadelphia, Pennsylvania: Lea & Febiger, 1980: 229. Samaan NA, et al. Hypothalamic, pituitary and thyroid dysfunction after radiotherapy of the head and neck. Int J Radiat Oncol Biol Phys 1982; 8: 1857. Pryzant RM, et al. Re-treatment of nasopharyngeal carcinoma in 53 patients. Int J Radiat Oncol Biol Phys 1992; 22: 941. Lee AW, et al. Reirradiation for recurrent nasopharyngeal carcinoma: factors affecting therapeutic ratio and ways for improvement. Int J Radiat Oncol Biol Phys 1997; 34: 43. Buatti JM, et al. Linac radiosurgery for locally recurrent nasopharyngeal carcinoma: rationale and technique. Head Neck 1995; 17: 14. Cmelak AJ, et al. Radiosurgery for skull base malignancies and nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 1997; 37: 997. Feehan PE, et al. Recurrent locally advanced nasopharyngeal carcinoma treated with heavy charged particle irradiation. Int J Radiat Oncol Biol Phys 1992; 23: 881. Fee WE, Robertson JR, Goffinet DR. Long-term survival after surgical resection for recurrent nasopharyngeal cancer after radiotherapy failure. Arch Otolaryngol Head Neck Surg 1991; 117: 1233. Tu G-Y, et al. Salvage surgery for nasopharyngeal carcinoma. Arch Otolaryngol Head Neck Surg 1988; 114: 328. Morton RP, et al. Transcervico-mandibulo-palatal approach for surgical salvage of recurrent nasopharyngeal cancer. Head Neck 1996; 18: 352. Panje WR, Gross CE. Treatment of tumours of the nasopharynx: surgical therapy. In Thawley SE, Panje WR (eds) Comprehensive Management of Head and Neck Tumours, Vol. 1. Philadelphia, Pennsylvania: WB Saunders, 1987: 662. Fisch U. The infratemporal approach for nasopharyngeal tumours. Laryngoscope 1983; 93: 36 – 43. Wei WI, et al. Maxillary swing approach for resection of tumours in and around the nasopharynx. Arch Otolaryngol Head Neck Surg 1995; 121: 638. Wei WI, et al. Efficacy of radical neck dissection for the control of cervical metastasis after radiotherapy for nasopharyngeal carcinoma. Am J Surg 1990; 160: 439 – 42. Yen KL, et al. Salvage neck dissection for cervical recurrence of nasopharyngeal carcinoma. Head Neck Surg 1997; 123: 725 – 9. Fandi A, et al. Long-term disease-free survivors in metastatic undifferentiated carcinoma of nasopharyngeal type. J Clin Oncol 2000; 18: 1324 – 30. Yeo W, et al. A phase II study of combination paclitaxel and carboplatin in advanced nasopharyngeal carcinoma. Eur J Cancer 1998; 13: 2027 – 31. Ma BB, et al. Chemotherapy with gemcitabine-containing regimens for locally recurrent or metastatic nasopharyngeal carcinoma. Cancer 2002; 95: 2516 – 23.
132
HEAD AND NECK CANCER
101. Chua DT, et al. Prognostic value of epidermal growth factor receptor expression in patients with advanced stage nasopharyngeal carcinoma treated with induction chemotherapy and radiotherapy. Int J Radiat Oncol Biol Phys 2004; 59: 11 – 20. 102. Leong JL, et al. Epidermal growth factor receptor expression in undifferentiated carcinoma of the nasopharynx. Laryngoscope 2004; 114: 153 – 7. 103. Chan ATC, et al. Multicenter, phase II study of cetuximab in combination with carboplatin in patients with recurrent or metastatic nasopharyngeal carcinoma. J Clin Oncol 2005; 23: 3568 – 76.
104. Trigo J, et al. Cetuximab monotherapy is active in patients with platinum-refractory recurrent/metastatic squamous cell carcinoma of the head and neck: results of a phase II study (Meeting proceedings) (Abstract 5502). J Clin Oncol 2004; 22(14s): 488. 105. Straathof KC, et al. Treatment of nasopharyngeal carcinoma with Epstein-Barr virus-specific T lymphocytes. Blood 2005; 105: 1898 – 904.
Section 2 : Head and Neck Cancer
9
Esthesioneuroblastoma
Barbara A. Murphy, Joseph M. Aulino, Christine H. Chung, Kim Ely, Robert Sinard and Anthony T. Cmelak
INTRODUCTION Esthesioneuroblastoma (ENB), also termed olfactory neuroblastoma (ONB), is a rare malignancy of neural crest origin, which comprises about 6% of nasal cavity and paranasal sinuses tumors.1,2 First described by Berger and Luc3 in 1924, more than 1400 cases have since been reported in the literature.4 The neoplasm develops from the olfactory epithelium of the cribriform plate, the superior upper one-third of the nasal septum, and the upper surface of the superior turbinate, and has a propensity to invade the skull base and intracranial space.5 ENB has been known by many names including: esthesioneurocytoma, olfactory intranasal neuroblastoma, esthesioneuroepithelioma, olfactory esthesioneuroma, tumor of olfactory placode and neuroolfactory tumor.6 The rarity of this tumor has led to the slow progress in our understanding of its pathobiology and the lack of consensus regarding treatment. Currently, there are no clearly established risk factors or causal agents. ENB occurs across the age spectrum from toddlers to the elderly. No racial or gender predilection has been noted. Endocrine activity has been described in rare cases including Cushing’s syndrome7 and syndrome of inappropriate secretion of antidiuretic hormone (SIADH).8 The most common presenting symptoms include nasal congestion, anosmia, recurrent epistaxis, pain, headache, and diplopia.2 Because many of these symptoms are nonspecific and can be attributed to chronic sinusitis, diagnosis can be delayed for prolonged periods of time, and hence patients often present with locally advanced disease. Since ENB is a rare tumor, individual investigators see and treat only a handful of cases. Thus, pooling of cases is critical. A number of reviews published over the years have provided insight into the natural history of ENB and have helped in the development of treatment paradigms.9 – 16 More recently, two large meta-analyses have been conducted by Broich (1997)6 and Dulguerov (2001).2 Broich provides a detailed listing of 208 documented reports between 1924 and 1994 totaling 970 cases of ENB. Dulguerov reported on 26 selected studies that include 390 patients, the majority
of which have 5-year survival data. Together, these reports have provided important insight into this uncommon disease process. In general, ENB is considered to be a slow-growing tumor with a prolonged disease course. In a meta-analysis conducted by Broich et al.,6 the 5-year overall survival of 234 patients was as follows: 68% alive without disease, 13% alive with disease, and 19% dead. When compared to other nonsquamous cell neoplasms of the nasal cavity and paranasal sinuses, 5- and 10-year survival is relatively high (see Figure 1).17 There is a subset of patients with poorly differentiated tumors, whose disease course is characterized by rapid progression with high rates of metastatic disease. However, some feel that these rapidly growing tumors represent a different pathologic entity, in particular, the clinically aggressive sinonasal undifferentiated carcinomas (SNUC).15 Miyamoto argues that the designation of high-grade tumors as SNUCs instead of ENB explains the high survival rates reported in some series.18 Advances in molecular biology will hopefully answer this question in due time.
STAGING ENB usually involves the ethmoid sinuses, and commonly involves the maxillary sinuses, sphenoid sinuses, frontal sinuses, orbits, anterior cranial fossa dura, cavernous sinus, and frontal lobes. The primary tumor more commonly extends inferiorly into the nasal cavity and paranasal sinuses than superiorly to involve the brain.19 Involvement of these critical structures by tumor has profound implications for staging, treatment, and functional outcome. The original staging system for ENB was proposed by Kadish et al. in 197611 (see Table 1). Tumors were categorized into three stages: A – confined to the nasal cavity, B – confined to the nasal and paranasal cavities, and C – tumors extending beyond the nasal and paranasal cavities. Only 5% of patients presented with clinically evident nodal disease; however, with time, it has become evident that this cohort of patients has a significantly worse survival (29% for node positive vs 64% for node negative disease).2
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
134
HEAD AND NECK CANCER Table 1 Staging systems.
Prop surv 1 0.8 0.6
Salivary
0.4 0.2 0
Sarcoma 0
60
(a)
120
180
Time-in mos Prop Surv
1 0.8
Esthesio
0.6 0.4
Lymphoma
0.2
Kadish Stage A B C
Extension Limited to the nasal cavity Involving the nasal and paranasal cavities Extending beyond the nasal and paranasal cavities
Morita Stage A B C D
Extension Limited to the nasal cavity Involving the nasal and paranasal cavities Extending beyond the nasal and paranasal cavities Presence of metastases
Biller T Stage T1 T2 T3 T4
Extension Limited to the nasal and paranasal cavities Extension to periorbital tissue and cranial cavity Extending to the brain with good operability Nonoperable extension to the brain
Dulguerov T Stage T1 T2 T3 T4
Extension Nasal and paranasal cavities excluding the sphenoid T1 plus extension to the sphenoid Extending to the orbit and anterior fossa Extension to the brain
Melanoma 0 (b)
0
60
120
180
Time-in mos
Figure 1 Comparative survival for ENB and other tumors involving the base of skull.17 (a) Actuarial survival for patients with salivary-type cancer and sarcoma. (b) Actuarial survival for patients with melanoma, lymphoma, and ENB. (Reprinted from reference 17 with permission of John Wiley & Sons, Inc. 1995).
Thus, in 1993, Morita recommended the addition of stage D – presence of metastases to accommodate this cohort of patients.13 Biller et al. proposed the first staging system using the TNM classification in 1990.20 T3 tumors extended to include the brain, and were distinguished from T4 tumors by operability. Finally, in 1992, Dulguerov proposed a staging system that could be based on computed tomography (CT) or magnetic resonance imaging (MRI) findings thus allowing nonoperative staging.21 Because of the nonspecific nature of the presenting symptoms, patients may be symptomatic for prolonged periods of time prior to diagnosis. Therefore, the majority of patients present with locally advanced disease. In a meta-analysis conducted by Dulguerov, studies using the Kadish staging system reported the following stages at presentation: stage A – 12%, stage B – 27%, and stage C – 61%. Studies using the Dulguerov staging system reported the following T stages at presentation: T1 – 25%, T2 – 25%, T3 – 33%, and T4 – 17%.2 Patients uncommonly present with distant metastases; disseminated disease is usually a feature of recurrence.19,22,23 Hematogenous spread to the lung, bone, liver, pancreas, skin, mediastinum, and brain has been described.19,22 – 24 Unfortunately, other than the presence or absence of nodal disease, current staging systems fail to consistently predict outcome. In the meta-analysis conducted by Dulguerov, 5-year survival based on stage was as follows: Kadish stage A – 72% (SD 41), stage B – 59% (SD 44), and stage C –
47% (SD16); and Dulguerov stage T1 – 81% (SD 17), T2 – 93% (SD 14), T3 – 33% (SD 33), and T4 – 48% (SD 41).
PATHOLOGY Morphologic Characteristics ENBs are polypoid, red-gray to tan lesions that are composed of discrete lobules of monotonous cells set in a delicate neurofibrillary background (see Figure 2). Less commonly, these nodules may coalesce to produce a sheetlike growth pattern with minimal intervening connective tissue. The eosinophilic fibrillary stroma is present in variable amounts in approximately 86% of cases and has been shown ultrastructurally to represent neuronal cell processes. When this fibrillary material is prominent, Homer –Wright rosettes are typically found, with about 28% of instances containing well-developed pseudorosettes. True or Flexner –Wintersteiner rosettes may also be seen, being identified in about 5% or less of tumors. In rare examples, admixed ganglion cells,25 rhabdomyoblasts,26 and melanin pigment27 are evident. Cytologically, the neoplastic cells possess small, hyperchromatic, round to oval nuclei with a stippled or “salt and pepper” chromatin possessing inconspicuous nucleoli. The cytoplasm is usually sparse with indistinct cell membranes; however, occasional ENBs may have more abundant eosinophilic cytoplasm. Most cases display a mild to moderate degree of nuclear atypia and a low mitotic rate. Necrosis, dystrophic calcification, and lymphovascular invasion are unusual findings. When pleomorphism is pronounced and the proliferation index high, the possibility of another neoplasm should be entertained.
Microscopic Grading In 1988, Hyams and associates28 proposed a system for grading ENB, which takes into account the following morphologic features: lobular architecture, neurofibrillary background, formation of rosettes, nuclear pleomorphism, mitotic
ESTHESIONEUROBLASTOMA
135
nearly all ENBs in his series were labeled with one or more antibodies that detected neuronal cytoskeletal proteins (class III β-tubulin isotype – 82%, microtubule-associated protein – 73%, neurofilament 200 kD – 73%). These results are in agreement with those in other analyses.25,34 Recently, a few authors have noted that some ENBs have been found to stain for keratin. Approximately, 19 to 36% of cases have been positive for low-molecular-weight keratin (CAM 5.2).33,34 They are, however, usually negative or at most focal and sporadically positive for AE1/AE3.33,34 Immunoreactivity for EMA and CEA is typically absent.33,35
Electron Microscopy On ultrastructural examination, ENB is consistently characterized by numerous intracytoplasmic, neurosecretory, densecore granules, and cell processes.32 The neurosecretory, dense-core granules range in size from 80 to 230 nm (mean, 140 nm; most granules measuring, 100–180 nm) and contain electron-dense cores of relatively uniform shape. Longitudinally arranged neural tubules and an occasional synaptic junction are present within neuronal processes.36,37 In addition to these structures, sustentacular-like cells can be seen at the periphery of cell nests. These cells are usually devoid of dense-core granules and microtubules.32
(a)
Differential Diagnosis
(b) Figure 2 (a) Low magnification: ENB is composed of clustered nests of cells in fibrillary matrix; (b) high magnification: the uniform “small blue cells” have indistinct nucleoli. Cell borders are difficult to appreciate.
activity, necrosis, and dystrophic calcification. On the basis of these criteria, a numerical grade was assigned from I to IV, with grade I corresponding to the most differentiated lesions and grade IV the least differentiated tumors. Some have found this classification to be an accurate indicator of local recurrence and survival,2,13,14,18 while others have found it to be of limited value in predicting outcome.29 – 31
Immunohistochemistry The typical immunohistochemical profile of ENB includes reactivity for markers of neuroendocrine differentiation, with 64 to 94% labeling with synaptophysin32 – 34 and up to 100% with neuron-specific enolase (NSE).33,34 Chromogranin is less consistent, with only occasional studies reporting variable expression.33,35 Between 7333 and 88%32,34 of cases demonstrate a distinctive peripheral staining pattern with antibodies to S-100 protein. This distribution of S-100 positive cells corresponds to the Schwann cells at the tumor – stroma interface ultrastructurally. Frierson33 observed that
ENB is a member of the family of “small round cell tumors” and bears histologic resemblance to embryonal rhabdomyosarcoma, lymphoma, and Ewing’s sarcoma/peripheral neuroectodermal tumor (ES/PNET). Usually, it can be distinguished by close attention to morphological, immunohistochemical, and anatomical detail. Specifically, its neurofibrillary stroma, immunophenotype (NSE+, synaptophysin+, LCA−, desmin−, and 013−), and confinement to the olfactory epithelium are defining features. In addition, the presence of characteristic S-100 reactive dendritic cells around lobules of tumor cells in ENB is a finding not shared by these other small round cell neoplasms and contrasts with the diffuse, strong staining seen in malignant melanoma.
MOLECULAR BIOLOGY Because of the limitations of histological diagnosis by light microscopy, various molecular techniques have been applied to look for characteristic cytogenetic and molecular features of ENBs. With recent development of molecular technologies such as comparative genomic hybridization (CGH), genomics, and proteomics, the identification of ENB tumors and pathologic classification systems will become more sophisticated. The additional biological information will also greatly facilitate clinical management. Initial evidence indicated that ENB was related to PNET and ES due to the presence of the characteristic translocation found in PNET and ES, t(11 : 22)(q24;q12) causing a gene fusion EWS/FLI1, in ENB cell lines established from metastatic lesions.38 – 40 However, emerging molecular evidence indicates that ENB is a distinct entity. Further studies
136
HEAD AND NECK CANCER
of ENB demonstrated that tumors lack the characteristic EWS gene rearrangement (measured by fluorescent insitu hybridization (FISH), reverse transcriptase polymerase chain reaction (RT-PCR) and southern blot analyses), and immunohistochemical analyses indicate that ENB tumors also lack the ES-associated MIC2 antigen.34,41,42 Furthermore, in one study, expression levels of the human achaete-scute homolog (hASH1) gene, which is critical in olfactory neuronal differentiation and expressed in immature olfactory cells, appeared useful as a diagnostic marker.43 All 4 ENB were positive, and 19 poorly differentiated tumors of sinonasal region were negative. In addition, there was an inverse relationship between hASH1 level and grade of the tumor. CGH was performed in a study of 12 ENB patients, including 12 primary tumors and their 10 metastatic/recurrent lesions.44 This study showed numerous individual chromosomal abnormalities and also a characteristic pattern of chromosomal imbalance consisting of deletions on chromosomal 3p12-p14 and overrepresentations on 17q12 and 17q25 in almost all cases. Thus, ENB may be separated from other
(a)
(b)
small round cell tumor types through its distinct cytogenetic pattern. In a comparison between the primary tumor, a metachronous lymph node, and two intraspinal metastatic lesions, a high number of shared overlapping alterations were found. This result supports the hypothesis of an underlying clonal process. However, metastatic/recurrent lesions had a higher mean number of chromosomal aberrations per tumor (16 vs 23). Specific patterns of alteration associated with metastatic/recurrent disease could also be determined (deletions in chromosome 5, 6q, 7q, 11, and 15q21, and amplifications of 1p32-p34, 1q12, 2p22-p24). More importantly, these patterns had clinical implications associated and were associated with a worse prognosis.
IMAGING General Imaging Considerations Although ENB is an uncommon tumor, the imaging appearance has been well-documented (see Figures 3 and 4).24,45 – 48
(c)
Figure 3 Esthesioneuroblastoma with intracranial extension. (a) Axial T1-weighted image shows tumor within the superior left nasal cavity and left ethmoid air cells (asterisks). Mildly T1-hyperintense signal within the left sphenoid and posterior left ethmoid air cells reflects retained secretions (arrows). (b) Coronal T1-weighted postcontrast image reveals cyst formation adjacent to the intracranial component, characteristic of esthesioneuroblastoma. (c) Axial T2-weighted image shows edema and mass effect.
(a)
(b)
(c)
Figure 4 Coronal images through left nasal cavity esthesioneuroblastoma with cribriform plate involvement. (a) CT image in a bone window setting shows opacification of the left nasal cavity and left maxillary sinus (white asterisks). Left ethmoid air cells appear opacified as well (arrowheads). The bony left cribriform plate is slightly attenuated. (b) T2-weighted image [slightly anterior to (a)] shows the retained secretions within the left maxillary sinus (black asterisk). The left ethmoid air cells are not completely replaced by tumor (curved arrow), as suggested on the CT scan. (c) Postcontrast T1-weighted image shows tumor extension across the cribriform plate (double arrows), without nodular dural enhancement. Retained left maxillary sinus secretions (large arrow).
ESTHESIONEUROBLASTOMA
The determination of the extent of tumor is essential for staging and appropriate therapy. The differential diagnosis of sinonasal tumors is extensive, and unfortunately the imaging appearance of ENB is relatively nonspecific. The imaging differential diagnosis of a mass centered in the superior nasal cavity includes squamous cell carcinoma, minor salivary gland tumor (adenoid cystic carcinoma, mucoepidermoid carcinoma, undifferentiated carcinoma, adenocarcinoma), melanoma, ENB, sarcoma (especially rhabdomyosarcoma), neuroma, lymphoma, and meningioma.48,49 However, the diagnosis of ENB should be entertained prior to biopsy if certain clinical and imaging findings are identified. Given that ENB arises from the basal cells of the olfactory epithelium,50,51 the location of the tumor centered in the superior nasal cavity within the olfactory recess is predictable. However, the olfactory epithelium extends inferiorly to the middle turbinates and can be found in the paranasal sinuses, explaining the occasional atypical locations of these tumors12,47 including the sphenoid sinus, sellar and parasellar region, the nasopharynx, or the petrous apex.52 – 59
137
is a frequent presenting symptom for ENB, no hemorrhage is seen within these tumors by imaging, unless a recent biopsy has been performed.47,48,60,64 Face MR imaging is superior to CT for assessment of intracranial extension of tumor through the cribriform plate to involve the dura, leptomeninges, and the brain.46,67 – 71 Cystic changes adjacent to areas of intracranial extension of a sinonasal mass have been described as specific for the diagnosis of ENB.47,61 Although tumor involvement of the olfactory nerves has been reported,72 frank perineural tumor spread is not characteristic of this malignancy. The face MR imaging protocol of sinonasal tumors should include postcontrast coronal fat-suppressed T1weighted images for evaluation of intracranial, dural, and leptomeningeal involvement, and for determining orbit invasion. Dedicated, small field-of-view imaging through the orbits and anterior skull base allows improved resolution compared with routine neck MR imaging, which typically includes lower resolution images through the face. If CT images through the neck have not been obtained, it would be appropriate to perform MR imaging of the neck at the time of the face MRI, to help identify involved cervical nodes.
CT Imaging of Local Disease Facial CT images are most useful to identify bony changes of erosion and remodeling,48 and to identify calcifications. Images in the axial and coronal planes generated using a bone-contrast setting should be obtained and carefully inspected for subtle erosions of the cribriform plate, lamina papyracea, and other facial bone structures. With multidetector CT scanner hardware, high-resolution coronally reformatted images can be generated from axial source data, limiting radiation, and patient discomfort. Calcifications have been identified within ENB,11,24,45,46,60 and are better seen on CT than MR images. However, the imaging appearance of residual, partially destroyed bone could be confused with tumor calcifications.61,62 Rarely, ENB tumors may cause a hyperostotic reaction in adjacent bone.63 If MR imaging cannot be performed, enhanced CT imaging of the anterior skull base is required to help identify intracranial extension. The use of CT to determine the extent of tumor is limited by retained sinus secretions, which may be of the same density as the tumor, and enhancing, thickened mucosa may also mimic tumor. Enhanced CT (or MR) imaging of the neck is used to identify regional nodal involvement.
MR Imaging of Local Disease The MR imaging appearance of ENB is not specific for this tumor.47,48 On T1-weighted images the tumor is hypo- to isointense, and on T2-weighted images the signal is iso- to hyperintense, compared with the brain. Contrast enhancement of the tumor is variable but always present. MR imaging is useful to distinguish tumor from retained paranasal sinus secretions, usually displaying different signal characteristics. T2-weighted images are most useful to differentiate tumor from sinus secretions,48,64,65 although precontrast T1weighted images occasionally contribute information to make the distinction.48,65 Enhanced images may also be useful to separate secretions from tumor.46,48,60,66 Although epistaxis
CT versus MRI for Local Disease CT is superior to MRI for evaluation of bone detail, including bony anterior skull base destruction,68,70 paranasal sinus and facial bone erosion,68 and tumor calcifications. MRI outperforms CT for determining dural and leptomeningeal involvement48,70 as well as differentiating sinus secretions from tumor.46,48,65 MR and CT imaging of the face and orbits for evaluation of local disease are complementary, and should both be performed for accurate staging and surgical planning.68,72 The CT and MR images should be carefully inspected for lamina papyracea destruction, orbit periosteum involvement and displacement, extraocular muscle involvement and displacement, and orbit fat involvement. Unfortunately, both CT and MR imaging underestimate orbit involvement by sinonasal tumors when compared with intra-operative assessment.72,73 Therefore, intra-operative assessment remains the gold standard for the determination of the extent of local disease.72,74
Regional and Distant Disease CT has been shown to perform slightly better than MR imaging for the detection of nodes involved by metastatic squamous cell carcinoma,75 and this would similarly seem to be the case for other metastatic tumors as well. CT imaging is also more sensitive than physical examination for detecting involved nodes,76 – 78 and should complement the clinical evaluation routinely. Nuclear medicine bone scan has been used to characterize osseous involvement at distant sites,22 and CT scanning of the chest, abdomen, and pelvis may be useful to detect solid organ involvement. Nuclear medicine whole-body fluoro-deoxy-glucose positron emission tomography (FDG-PET) scanning appears sensitive to the detection of widespread disease. Although this newer technology remains to be explored, a relative lack of specificity may limit utility.79 Despite this limitation, FDG-PET
138
HEAD AND NECK CANCER
shows promise in evaluating for both regional and distant metastatic disease, and will likely become a routine diagnostic method for ENB staging.
Molecular Imaging Despite the theorized neural crest cell origin of the tumor, ENB does not reliably display increased uptake of radioiodinated metaiodobenzylguanidine (131 I-MIBG) by nuclear medicine imaging, although cases have been reported.80,81 Prado et al. presented a case of absent 131 I-MIBG uptake, possibly related to the histologically undifferentiated nature of the tumor.82 In that same patient, increased radiotracer accumulation of technetium-99methylcysteinate dimer (99m Tc-ECD) was identified during evaluation of cerebral blood flow related to the activity of the enzyme esterase within the tumor tissue. Similarly, strong uptake of 111 indium-labeled octreotide, a somatostatin analog, has been shown in primary and metastatic ENB, suggesting a future role for somatostatin receptor-directed radionuclide therapy.79,83
Recurrence and Surveillance Imaging Because of the low incidence of ENB, there are no prospective studies of imaging surveillance. Since a third of patients with regional recurrence are salvageable,2 aggressive imaging screening as an adjunct to physical examination for the detection of cervical lymphadenopathy is recommended.75 – 78 Late recurrences have been reported,11,30 and lifetime surveillance has been proposed.84,85 Lund et al. found that more than half of recurrences occurred adjacent to or within the orbit, suggesting dedicated orbit MR imaging for surveillance.86 Unfortunately, postoperative and postradiation granulation tissue can be indistinguishable from tumor in the paranasal sinuses,48,65 often requiring biopsy or serial MR imaging to detect an interval change. Cerebral radiation necrosis may mimic recurrence intracranially.87 FDG-PET imaging will likely play a future role in the distinction of postoperative and postradiation changes from recurrent tumor, both in the paranasal sinuses and intracranially. We suggest baseline face and brain MR imaging 8–16 weeks after surgery and radiation, then every 4–6 months for 5–8 years, with consideration of annual imaging surveillance thereafter. Although the survival rate of patients with disseminated disease is dismal,2 radiochemotherapy and bone marrow transplantation have been used successfully to treat metastatic disease.84 Koka et al. reported a 40% rate of distant metastases in their series, of which the most common site was bone (82% of patients with metastases).22 They therefore proposed nuclear medicine bone scans be used for routine screening of metastases. However, the metastatic rate to bone is substantially lower in most series, and the costeffectiveness of this approach is questionable.
TREATMENT Historical Perspective Historically, ENB was treated with either surgery or radiation therapy. A case review of 104 patients with ENB was
reported by Skolnick in 1966. Five-year survival was 64% for patients treated with surgery and 38% for patients treated with radiation therapy.88 Early reports documented a high rate of local recurrence with a significant percentage of patients experiencing multiple recurrences over extended periods of time.10,12 Although recurrence rates were high, some investigators reported high rates of salvage. Two advances in treatment have made a dramatic impact on outcome: the use of craniofacial resections and the use of combined modality therapy. Craniofacial resection, which allows the en bloc removal of tumor, is described below under surgical considerations.
Treatment of Early Stage Disease Early stage lesions below the cribriform plate that exhibit no bony erosion or cranial nerve involvement can be treated with definitive irradiation with good tumor control and preservation of function. However, the vast majority of these tumors will be resected with negative margins and without compromising olfaction or vision. Therefore, the standard practice is to observe or, more commonly, to irradiate postoperatively using moderate doses of 50 to 60 Gy with conformal treatment planning (3D or intensitymodulated radiation therapy – IMRT) to minimize dose to the optic apparatus, pituitary, and the brain.89 Elective nodal irradiation is generally not indicated in this setting if imaging (CT, MR, PET) is normal due to the low risk (less than 15%) of subclinical metastases.10
Treatment of Locally Advanced Resectable Disease Patients with more advanced local disease benefit from a multimodality approach incorporating surgery, radiation, and chemotherapy. In this setting, the risk of regional nodal and distant metastases becomes a factor in decision making for the clinician. In addition, functional impairment from radical resection and the ability to resect all gross disease with clear margins becomes more difficult. Historically, radical resection followed by postoperative irradiation has been the favored approach. Local control of 65 to 100% has been reported, with larger retrospective studies favoring the use of postoperative irradiation even in patients with negative surgical margins. In the Broich meta-analysis,6 combined modality therapy demonstrated a clear superiority to single modality therapy (72.5% for surgery and radiation therapy versus 62.5% for surgery alone and 53.9% for radiation alone). Similar results were reported in the Dulguerov metaanalysis2 (65% for surgery and radiation therapy versus 48% for surgery alone versus 37% for radiation alone). Although death rates were highest with radiation alone, this may represent a cohort with a high percentage of patients with advanced nonresectable disease. Foote and colleagues14 from the Mayo Clinic reported results in 49 patients, 32 of whom had advanced local disease, who underwent surgery with or without postoperative irradiation. The 5-year local control, disease-free survival, and overall survival were 65.3, 54.8, and 69.1%, respectively. Postoperative irradiation (mean 55.5 Gy) improved local tumor control even in patients with gross total resection. Some have argued against the combined
ESTHESIONEUROBLASTOMA
approach citing a high rate of complications.90 More recent reports using updated craniofacial surgical techniques and 3D conformal RT or IMRT do not corroborate these concerns and strongly advocate postoperative irradiation with modern techniques in essentially all cases.16,91
Unresectable and Marginally Resectable Disease Unfortunately, some patients will present with unresectable disease. Although criteria vary, patients with invasion of the optic chiasm, the cavernous sinus, and middle fossa are generally considered unresectable.92 In addition, patients with extensive or bilateral frontal lobe involvement are considered poor surgical candidates. In patients with tumors that are unresectable or marginally resectable, preoperative radiotherapy with or without chemotherapy has been attempted for tumor downsizing with some success. Because of the histological similarity to small cell carcinomas, investigators have utilized cisplatin-based chemotherapy regimens to treat ENB. Some reports indicate that systemic chemotherapy may be highly effective in previously untreated ENB. In a report by Kim et al., 9 of 11 patients had an objective response to a combination of cisplatin, ifosfamide, and etoposide.93 In a report by Mishima,94 9 of 12 previously untreated patients who received cyclophosphamide, doxorubicin, and vincristine with continuous infusion cisplatin and etoposide had at least a partial response after two cycles of therapy. Conversely, of three patients treated with induction therapy by investigators at the Johns Hopkins, none had a response.23 Koka22 reported the results of 40 patients with ENB treated at Institut Gustave Roussy. Sixteen patients received induction chemotherapy. Only 5 (31%) patients had a complete response to induction chemotherapy; however, response to chemotherapy predicted survival in those patients who were treated with definitive radiation therapy. Despite the mixed results, chemotherapy is being incorporated into a combined modality approach for patients with advanced disease. Investigators at the University of Virginia have been strong advocates for the use of preoperative radiation therapy or chemoradiation for patients with Kadish stage B and C disease. Preoperative doses of 45 to 50 Gy were utilized to remain within optic nerve tolerance. Polin et al. reported that two-thirds of patients experienced a significant reduction in tumor burden.95 Responding patients experienced an improved disease-free survival. Overall 5- and 10-year survival for the entire patient cohort was 81 and 54.5%, respectively. Others have used radiation with chemotherapy after tumor debulking in order to minimize surgical morbidity. Wieden et al. reported using this approach with postoperative radiation (55.8 Gy), cisplatin and 5-fluorouracil with good survival.96 Radiotherapy with chemotherapy has also been used as definitive treatment in locally advanced tumors as well. Fitzek et al. reported 5-year local control of 88% and 5-year survival of 74% in 19 patients treated with etoposide and cisplatin with concomitant-boost radiation.97 Bhattacharyya98 utilized a similar approach using daily radiation to 68 Gy, utilizing resection only for incomplete tumor response. Eight
139
of nine patients were able to forego surgery, and complications were low. No patients had experienced recurrent tumor with short follow-up.
Nodal Disease Approximately 5% of patients present with cervical node metastases.2 The presence of involved cervical lymph nodes at diagnosis is an important prognostic factor.2,19,22,23,99,100 In a study by Koka, the 2-year survival of patients with palpable lymph nodes at presentation was 0% compared with 78% for patients presenting with N0 disease.22 They also noted no correlation of advanced clinical stage (tumor size) with cervical nodal metastases, suggesting aggressive nodal evaluation even with small primary lesions. Treatment options include resection, radiation therapy, or a combination of both. The survival rate in this cohort of patients is poor. Because of the poor prognosis of patients with nodal disease, the question has arisen as to how to deal with the clinically negative neck. In a literature review evaluating neck recurrences, Beitler101 reported that 19% (21 of 110 patients) developed a neck recurrence. Thus, neck recurrence is relatively frequent. Similar results have been reported by others. In a review of 320 patients from 15 reports, Rinaldo100 reported that 23.4% of patients developed nodal disease. Neck recurrences can be delayed; 10 of 21 patients in Beitler’s study developed a recurrence within 24 months of diagnosis, and 18 of 21 had recurred within 60 months. Contralateral neck recurrence is common; hence, some authors have advocated bilateral neck treatment when nodal disease develops. Despite aggressive salvage therapy, a third of patients who develop neck recurrence will die of the disease. This has led some investigators to ask whether prophylactic treatment of the neck with either surgery or radiation would be beneficial.102 Koka22 reported a neck failure rate of 0% (0 of 12) in patients receiving elective neck radiation versus 19% (4 of 21) in patients who did not. Confirmatory data is limited; hence, some advocate treating the neck only when it is found to be clinically involved.100
Patterns of Recurrence After Primary Treatment The median time to first recurrence is generally less than 2 years; however, late recurrences are common. In one study, the longest time from presentation until recurrence was 13.3 years.15 Thus, follow-up must be protracted. The most common site of recurrence is local. Local recurrence rates have dropped significantly with the use of combined modality therapy and now occur in approximately 15–30% of patients.6,13,14,60,84,101 Salvage rates after local recurrence are approximately 30–50%.2 The second most common site of recurrence is in regional lymph nodes at 15 to 29%. The salvage rate for patients with nodal disease is about 30%. Although patients seldom present with distant metastases, 8 to 17% of patients will develop disseminated disease at some point during their disease process.19,22,23,60,84 Long-term survivors have been reported with high-dose chemotherapy and stem cell transplantation.
Chemotherapy for Recurrent Disease As expected, response rates in previously treated patients were substantially lower. In a report by investigators at
140
HEAD AND NECK CANCER
Mayo clinic,103 10 patients with Kadish stage C disease were treated with platinum-based chemotherapy. Nine of 10 patients had recurrent disease. Two of four patients with highgrade tumors responded to chemotherapy. None of the lowgrade tumors responded. Stewart et al. reported the results of 8 patients with recurrent ENB or poorly differentiated sinonasal tumors, treated with high-dose chemotherapy and autologous bone marrow transplant. At the time of reporting, three patients were alive, and another patient had survived five years, but relapsed and died after a second transplant. Treatment with high dose therapy remains experimental.104
Surgical Considerations The surgical management of ENB consists of an en bloc resection of the tumor, the olfactory epithelium from which it arises, the cribriform plate and its adjacent dural covering, and any other involved structures. This type of resection is best accomplished through what is called a craniofacial resection. There are several surgical approaches available to achieve this en bloc removal, and the choice of which one to use depends on the anatomical extent of tumor, the individual surgeon’s preference, and certain patient factors. Transfacial approaches to the anterior skull base can be viewed as a progression of increased exposure with extension of incisions that sequentially allow retraction of “modules” of the facial structure. These approaches are combined with a bifrontal craniotomy to gain superior exposure while protecting the frontal lobes of the brain. The lateral rhinotomy is the most basic unilateral approach to tumors of the anterior skull base. The incision begins under the medial aspect of the brow, curves medially and inferiorly between the nasal root and medial canthus, and then courses inferiorly between the nasal and medial cheek subunits, and then curves around the nasal ala. Superiorly, the incision is carried down to the underlying nasal bone and nasal process of the maxilla, while inferiorly the incision connects to the pyriform aperture. This allows removal of the medial maxilla for exposure, and the removed segment of bone can be replaced at the end of the case. This basic approach allows exposure for tumors involving the maxillary and ethmoid sinus and the anterior portion of the sphenoid sinus, with limited extension into the nasopharynx and clivus. When tumors spread posterolaterally into the infratemporal fossa, the incision can be extended under the eye out to the lateral canthus, providing improved lateral exposure at the price of potential ocular complications. Tumors with contralateral extension beyond the pterygoid plates require a more bilateral exposure such as that achieved with a midfacial degloving. This approach was designed to expose the maxilla without using facial incisions, primarily for repair of facial fractures. The approach begins with incisions in the nose and under the upper lip that separate the nasal pyramid and facial soft tissues from the underlying bone. Bone cuts are then made, as dictated by the extent of the tumor. The bicoronal flap is a critical step in the exposure of the anterior cranial fossa from above. It begins with an incision extending from one helical root of the ear to the
other, just posterior to the hairline. The pericranial tissues are then elevated forward to the supraorbital vessels and glabella medially and to the lateral orbital rims laterally. The craniotomy is then designed as dictated by the location of the tumor. Generally, a low craniotomy facilitates exposure by minimizing the amount of frontal lobe retraction needed to visualize the anterior cranial fossa. Tumor extension superiorly into the frontal sinus may preclude a low craniotomy, depending on the degree of pneumatization of the sinus. Preoperative imaging provides the information necessary for designing the craniotomy. Resection of the tumor at this point connects the cranial and facial approaches. After tumor removal, dural defects are repaired with grafts of fascia lata to make the closure watertight. The frontonasal bone segment is replaced and secured into position with miniplates. The frontal sinus is then “cranialized” by removal of the posterior wall and the frontal bone plate returned to its anatomic position and secured with miniplates or sutures through small drill holes. The list of complications that can occur with anterior skull base surgery includes those common to surgical procedures in general (such as infection, hemorrhage) and those specific to the anterior skull base (including cerebrospinal fluid (CSF) leak and pneumocephalus). As experience with the procedure and its modifications has grown over the decades, craniofacial resection is no longer considered a procedure of last resort. It is a viable, relatively safe procedure as long as careful attention is paid to the basic surgical principles of careful handling of tissues and accurate closure of dura to obtain anatomic separation of the sterile intracranial environment from the contaminants of the sinonasal cavities.
Radiation Therapy Considerations Radiotherapy should be delivered in fractions of 1.8 to 2.0 Gy to minimize risk of late normal tissue complications. Three-dimensional or intensity-modulated techniques should be utilized to limit dose to surrounding sensitive structures. Tissue homogeneity corrections are important in this group of patients, particularly after surgical resection where large air spaces may be present. Homogeneity within the high-dose region should be kept within 10% of the prescribed dose, if possible. This is more important when treating to doses over 60 Gy, when areas receiving greater than 110% dose will be at risk for radiation necrosis. Tolerance doses for brain, optic nerves, chiasm, retina, and pituitary need to be maintained to avoid catastrophic complications. In order to improve the therapeutic ratio, some have used conformal irradiation in combination with proton beam or radiosurgery.97 Zabel and colleagues showed this technique could simultaneously improve target coverage and lower doses to adjacent critical structures significantly.105
REFERENCES 1. Svane-Knudsen V, Jorgensen K, Hansen O, et al. Cancer of the nasal cavity and paranasal sinuses: a series of 115 patients. Rhinology 1998; 36: 12 – 4. 2. Dulguerov P, Allal AS, Calcaterra TC. Esthesioneuroblastoma: a meta-analysis and review. Lancet Oncol 2001; 2: 683 – 90.
ESTHESIONEUROBLASTOMA 3. Berger L, Luc G, Richard D. L’esthesioneuroepitheliome olfactif. Bull Cancer 1924; 13: 410 – 21. 4. Christmas DA, Mirante JP, Yanagisawa E. Endoscopic view of an Esthesioneuroblastoma that resembles a benign polyp. Ear Nose Throat J 2004; 83: 668 – 70. 5. Mishima Y, et al. Combination chemotherapy and radiotherapy with stem cell support can be beneficial for adolescents and adults with esthesioneuroblastoma. Cancer 2004; 101: 1437 – 44. 6. Broich G, Pagliari A, Ottaviani F. Esthesioneuroblastoma: a general review of the cases published since the discovery of the tumour in 1924. Anticancer Res 1997; 17: 2683 – 706. 7. Arnesen MA, Scheithauer BW, Freeman S. Cushing Syndrome secondary to olfactory neuroblastoma. Ultrastruct Pathol 1994; 18: 61 – 8. 8. Al Ahawal M, et al. Olfactory neuroblastoma: report of a case associated with inappropriate antidiuretic hormone secretion. Ultrastruct Pathol 1993; 18: 437 – 9. 9. Lewis JS, et al. Nasal tumors of olfactory origin. Arch Otolaryngol 1965; 81: 169 – 74. 10. Elkon D, et al. Esthesioneuroblastoma. Cancer 1979; 44: 1087 – 94. 11. Kadish S, Goodman S, Wang CC, Olfactory neuroblastoma, a clinical analysis of 17 cases. Cancer 1976; 37: 1571 – 6. 12. Shah JP, Feghali J. Esthesioneuroblastoma. Am J Surg 1981; 142: 456 – 8. 13. Morita A, Ebersold MJ, Olsen KD. Esthesioneuroblastoma prognosis and management. Neurosurgery 1993; 32: 706 – 14. 14. Foote RL, et al. Esthesioneuroblastoma: the role of adjuvant radiation therapy. Int J Radiat Biol Phys 1993; 27: 835 – 42. 15. Levine PA, Gallagher R, Cantrell R. Esthesioneuroblastoma: reflections of a 21-year experience. Laryngoscope 1999; 109: 1539 – 43. 16. Spaulding CA, et al. Esthesioneuroblastoma: a comparison of two treatment eras. Int J Radiat Biol Phys 1988; 15: 581 – 90. 17. Spiro JD, Soo KC, Spiro RH. Nonsquamous cell malignant neoplasms of the nasal cavity and paranasal sinuses. Head Neck 1995; 17: 114 – 8. 18. Miyamoto RC, et al. Esthesioneuroblastoma and sinonasal undifferentiated carcinoma: impact of histologic grading and clinical staging on survival and prognosis. Laryngoscope 2000; 110: 1262 – 5. 19. Dias FL, et al. Patterns of failure and outcome in esthesioneuroblastoma. Arch Otolaryngol Head Neck Surg 2003; 129: 1186 – 92. 20. Biller HF, et al. Esthesioneuroblastoma: surgical treatment without radiation. Laryngoscope 1990; 100: 1199 – 201. 21. Dulguerov P, Calcaterra T, Esthesioneuroblastoma: the UCLA experience 1970 – 1992. Laryngoscope 1992; 102: 843 – 9. 22. Koka VT, et al. Aesthesioneuroblastoma. J Laryngol Otol 1998; 112: 628 – 33. 23. Resto VA, et al. Esthesioneuroblastoma: the Johns Hopkins experience. Head Neck 2000; 22: 550 – 8. 24. Manelfe C, et al. Computer tomography in olfactory neuroblastoma: one case of esthesioneuroepithelioma and four cases of esthesioneuroblastoma. J Comput Assist Tomogr 1978; 2: 412 – 20. 25. Mills SE, Frierson HF. Olfactory neuroblastoma – a clinicopathologic study of 21 cases. Am J Surg Pathol 1985; 9: 317 – 27. 26. Slootweg PJ, Lubsen H. Rhabdomyoblasts in olfactory neuroblastoma. Histopathology 1991; 19: 182 – 4. 27. Curtis JL, Rubinstein LJ. Pigmented olfactory neuroblastoma. Cancer 1982; 49: 2136 – 43. 28. Hyams VJ, Batsakis VJ, Michaels L (eds). Tumors of the upper respiratory tract and ear. Atlas of Tumor Pathology, Vol. Fascicle 25. Washington, District of Columbia: Armed Forces Institute of Pathology, 1988: 240 – 248. 29. Papadaki H, et al. Relationship of p53 gene alterations with tumor progression and recurrence in olfactory neuroblastoma. Am J Surg Pathol 1986; 20: 715 – 21. 30. Diaz EM, et al. Olfactory neuroblastoma: the 22-year experience at one comprehensive cancer center. Head Neck 2005; 27: 138 – 49. 31. Ingeholm P, et al. Esthesioneuroblastoma: a Danish clinicopathological study of 40 consecutive cases. APMIS 2002; 110: 639 – 45. 32. Hirose T, et al. Olfactory neuroblastoma. An immunohistological, ultrastructural, and flow cytometric study. Cancer 1995; 76: 4 – 19. 33. Frierson HF, et al. Olfactory neuroblastoma. Additional immunohistochemical characterization. Am J Surg Pathol 1990; 94: 547 – 53.
141
34. Argani P, et al. Olfactory neuroblastoma is not related to the Ewing family of tumors: absence of EWS/FLI1 gene fusion and MIC2 expression. Am J Surg Pathol 1987; 22: 391 – 8. 35. Axe S, Kuhajda FP. Esthesioneuroblastoma. Intermediate filaments, neuroendocrine, and tissue-specific antigens. Am J Clin Pathol 1987; 88: 139 – 45. 36. Min K. Usefulness of electron microscopy in the diagnosis of “small” round cell tumors of the sinonasal region. Ultrastruct Pathol 1995; 19: 347 – 63. 37. Taxy JB, et al. The spectrum of olfactory neural tumors-a light microscopic immunohistochemical and ultrastructural analysis. Am J Surg Pathol 1986; 10: 687 – 95. 38. Whang-Peng J, et al. Translocation t(11;22) in esthesioneuroblastoma and their metastases. Cancer Genet Cytogenet 1987; 29: 155 – 7. 39. Cavazzana AO, et al. Olfactory neuroblastoma is not a neuroblastoma but is related to primitive neuroectodermal tumor (PNET). Prog Clin Biol Res 1988; 271: 463 – 73. 40. Sorensen PH, et al. Olfactory neuroblastoma is a peripheral primitive neuroectodermal tumor related to Ewing’s sarcoma. Proc Natl Acad Sci USA 1996; 93: 1038 – 43. 41. Mezzelani A, et al. Esthesioneuroblastoma is not a member of the primitive peripheral neuroectodermal tumour-Ewing’s group. Br J Cancer 1999; 81: 586 – 91. 42. Kumar S, et al. Absence of EWS/FLI1 fusion in olfactory neuroblastomas indicates these tumors do not belong to the Ewing’s sarcoma family. Hum Pathol 1999; 30: 1356 – 60. 43. Mhawawech P, et al. Human achaete-scute homologue (hASHI) mRNA level as a diagnostic marker to distinguish esthesioneuroblastoma from poorly differentiated tumors arising in the sinonasal tract. Am J Clin Pathol 2004; 122: 100 – 5. 44. Bockmuhl U, et al. CGH pattern of esthesioneuroblastoma and their metastases. Brain Pathol 2004; 14: 158 – 63. 45. Hurst RW, et al. Computer tomographic features of esthesioneuroblastoma. Neuroradiology 1989; 31: 253 – 7. 46. Li C, et al. Olfactory neuroblastoma: MR evaluation. AJNR Am J Neuroradiol 1993; 14: 1167 – 71. 47. Derdeyn CP, et al. MRI of esthesioneuroblastoma. J Comput Assist Tomogr 1994; 18: 16 – 21. 48. Schuster JJ, Phillips CD, Levine PA. MR of esthesioneuroblastoma (olfactory neuroblastoma) and appearance after craniofacial resection. AJNR Am J Neuroradiol 1994; 15: 1169 – 77. 49. Loevner LA, Sonners AI. Imaging of neoplasms of the paranasal sinuses. Magn Reson Imaging Clin N Am 2002; 10: 467 – 93. 50. Michaeu C. A new histological approach to olfactory esthesioneuroma. Cancer 1977; 40: 314 – 8. 51. Cantrell RW, Ghorayeb BY, Fitz-Hugh GS, Esthesioneuroblastoma: diagnosis and treatment. Ann Otol Rhinol Laryngol 1977; 86: 760 – 5. 52. Castro L, de la Pava S, Webster JH. Esthesioneuroblastoma: a report of 7 cases. Am J Roentgenol Radium Ther Nucl Med 1969; 105: 7 – 13. 53. Mashberg A, Thoma KH, Wasilewski EJ. Olfactory neuroblastoma (esthesioneuroblastoma) of the maxillary sinus. Oral Surg Oral Med Oral Pathol 1960; 13: 908 – 12. 54. Sawar M. Primary sellar-parasellar esthesioneuroblastoma. Am J Roentgenol 1979; 133: 140 – 1. 55. Chacko G, Chandi Chandy MJ. Primary sphenoid and petrous apex esthesioneuroblastoma: case report. Br J Neurosurg 1998; 12: 264 – 6. 56. Roy A, et al. Correspondence: aesthesioneuroblastoma arising in pituitary gland. Neuropathol Appl Neurobiol 2000; 26: 177 – 9. 57. Sharma SC, et al. Isolated esthesioneuroblastoma of sphenoid sinus. Am J Otolaryngol 2002; 23: 287 – 9. 58. Chirico G, et al. Primary sphenoid esthesioneuroblastoma studied with MR. J Clin Imag 2003; 27: 38 – 40. 59. Mariani L, et al. Esthesioneuroblastoma of the pituitary gland: a clinicopathological entity? J Neurosurg 2004; 101: 1049 – 52. 60. Pickuth D, Heywang-Kobrunner SH. Imaging of recurrent esthesioneuroblastoma. Br J Radiol 1999; 72: 1052 – 7. 61. Som PM, et al. Sinonasal esthesioneuroblastoma with intracranial extension: marginal tumor cysts as a diagnostic MR finding. AJNR Am J Neuroradiol 1994; 15: 1259 – 62. 62. Som PM, Lidov M. The significance of sinonasal radiodensities: ossification, calcification, or residual bone? AJNR Am J Neuroradiol 1994; 15: 917 – 22.
142
HEAD AND NECK CANCER
63. Roegenbogen VS, et al. Hyperostotic esthesioneuroblastoma: CT and MR findings. J Comput Assist Tomogr 1988; 12: 52 – 6. 64. Schroth G, et al. MR imaging of esthesioneuroblastoma. J Comput Assist Tomogr 1986; 10: 316 – 9. 65. Som PM, et al. Sinonasal tumors and inflammatory tissues: differentiation with MR imaging. Radiology 1988; 167: 803 – 9. 66. Lanzieri CF, et al. Use of gadolinium-enhancing MR imaging for differentiating mucoceles from neoplasms in the paranasal sinuses. Radiology 1991; 178: 425 – 8. 67. Sze G, et al. MR imaging of the cranial meninges with emphasis on contrast enhanced and meningeal carcinomatosis. AJNR Am J Neuroradiol 1989; 10: 965 – 75. 68. Kraus DH, et al. Complementary use of computer tomography and magnetic imaging in assessing skull base lesions. Laryngoscope 1992; 102: 623 – 9. 69. Eisen MD, et al. Use of pre-operative MR to predict dural, perineural, and venous sinus invasion of skull base tumors. AJNR Am J Neuroradiol 1996; 17: 1937 – 45. 70. Ishida H, Mohri M, Amatsu M. Invasion of the skull base by carcinomas: histopathologically evidenced findings with CT and MRI. Eur Arch Otorhinolaryngol 2002; 259: 535 – 9. 71. Maroldi R, Ambrosi C, Farina D. Metastatic disease of the brain: extra-axial metastases. Eur Radiol 2005; 15: 617 – 26. 72. Eisen MD, et al. Preoperative imaging to predict orbital invasion by tumor. Head Neck 2000; 22: 456 – 62. 73. Maroldi R, et al. MR of malignant nasosinusal neoplasm. Frequently asked questions. Eur Radiol 1997; 24: 181 – 90. 74. Som PM, et al. Ethmoid sinus disease: CT evaluation in 400 cases. Part III. Radiology 1986; 159: 605 – 9. 75. Curtin HD, et al. Comparison of CT and MR imaging in staging neck metastases. Radiology 1998; 207: 123 – 30. 76. Shingaki S, et al. Computer tomographic evaluation of lymph node metastases in head and neck carcinomas. J Craniomaxillofac Surg 1995; 23: 233 – 7. 77. Merritt RM, et al. Detection of cervical metastasis: a metaanalysis comparing computer tomography with physical exam. Arch Otolaryngol Head Neck Surg 1997; 123: 149 – 52. 78. Atula TS, et al. Assessment of cervical lymph node status in head and neck cancer patients: palpation, computer tomography and low field magnetic resonance imaging compared with ultra-sound-guided fine-needle aspiration cytology. Eur Radiol 1997; 25: 152 – 61. 79. Yu J, et al. Ectopic Cushing’s syndrome caused by an esthesioneuroblastoma. Endocr Pract 2004; 10: 119 – 24. 80. Kairemo KJA, et al. Imaging of olfactory neuroblastoma – an analysis of 17 cases. Auris Nasus larynx 1998; 25: 173 – 9. 81. Sasajima T, et al. High uptake of 123I-metaiodobenzylguanidine related to olfactory neuroblastoma revealed by single-photon emission CT. AJNR Am J Neuroradiol 2000; 21: 717 – 20. 82. Prado GLM, et al. Olfactory neuroblastoma visualized by technetium99m-ECD SPECT. Radiat Med 2001; 19: 267 – 70. 83. Ramsey HA, Kairemo KJ, Jekunen AP. Somatostatin receptor imaging of olfactory neuroblastoma. J Laryngol Otol 1996; 110: 1161 – 3. 84. Eden BV, et al. Esthesioneuroblastoma: long-term outcome and patterns of failure – the University of Virginia experience. Cancer 1994; 73: 2556 – 62. 85. Girod D, Hanna E, Marentette L. Esthesioneuroblastoma. Head Neck 2001; 23: 500 – 5.
86. Lund VJ, et al. Olfactory neuroblastoma: past, present, and future? Laryngoscope 2003; 11: 502 – 7. 87. Tran TA, et al. Delayed cerebral radiation necrosis. Am J Roentgenol 2003; 180: 70. 88. Skolnick EM, Massari FM, Tenta LT, Olfactory neuroepithelioma. Arch Otolaryngol 1966; 84: 644 – 53. 89. Chao KS, et al. Esthesioneuroblastoma: the impact of treatment modality. Head Neck 2001; 23: 749 – 57. 90. Olsen KD, De Santo L, Olfactory neuroblastoma. Acta Otolaryngol 1983; 109: 797 – 802. 91. O’connor TA, et al. Olfactory neuroblastoma. Cancer 1989; 63: 2426 – 8. 92. Levine PA, McLean WC, Cantrell RW. Esthesioneuroblastoma: the University of Virginia experience 1960 – 1985. Laryngoscope 1986; 96: 742 – 6. 93. Kim DW, et al. Neoadjuvant etoposide, Ifosfamide, and cisplatin for the treatment of olfactory neuroblastoma. Cancer 2004; 101: 2257 – 60. 94. Mishima Y, et al. Combination chemotherapy (cyclophosphamide, doxorubicin, and vincristine with continuous-infusion cisplatin and etoposide) and radiotherapy with stem cell support can be beneficial for adolescents and adults with esthesioneuroblastoma. Cancer 2004; 101: 1437 – 44. 95. Polin RS, et al. The role of pre-operative adjuvant treatment in the management of esthesioneuroblastoma: the University of Virginia Experience. Neurosurgery 1998; 42: 1029 – 37. 96. Wieden PL, Yarington CT Jr, Richardson RG. Olfactory neuroblastoma. Chemotherapy and radiotherapy for extensive disease. Arch Otolaryngol 1984; 110: 759 – 60. 97. Fitzek MM, et al. Neuroendocrine tumors of the sinonasal tract. Results of a prospective study incorporating chemotherapy, surgery, and combined proton-photon radiotherapy. Cancer 2002; 94: 2623 – 34. 98. Bhattacharyya N, et al. Successful treatment of esthesioneuroblastoma and neuroendocrine carcinoma with combined chemotherapy and proton radiation: results in 9 cases. Arch Otolaryngol Head Neck Surg 1997; 123: 34 – 40. 99. Homzie MJ, Elkon D. Olfactory esthesioneuroblastoma. Cancer 1980; 46: 2509 – 13. 100. Rinaldo A, et al. Esthesioneuroblastoma and cervical lymph node metastases: clinical and therapeutic implications. Acta Otolaryngol 2002; 122: 215 – 21. 101. Beitler JJ, et al. Esthesioneuroblastoma: is there a role for elective neck treatment. Head Neck 1991; 13: 321 – 26. 102. Ferlito A, Rinaldo A, Rhys-Evans PH. Contemporary clinical commentary: esthesioneuroblastoma: an update on management of the neck. Laryngoscope 2003; 113: 1935 – 8. 103. McElroy E, Buckner JC, Lewis JE. Chemotherapy for advanced esthesioneuroblastoma: the Mayo clinic experience. Neurosurgery 1998; 42: 1023 – 7. 104. Stewart FM, et al. High-dose chemotherapy and autologous marrow transplant for esthesioneuroblastoma and sinonasal undifferentiated carcinoma. Am J Clin Oncol 1989; 12: 217 – 21. 105. Zabel A, Thilmann C, Milker-Zabel S, et al. The role of stereotactically guided conformal radiotherapy for local tumor control of esthesioneuroblastoma. Strahlenther Onkol 2002; 178: 187 – 91.
Section 3 : Endocrine Tumors
10
Neoplastic Disorders of the Adrenal Glands K. Oberg, A. Goldhirsch and A. Munro Neville
INTRODUCTION Disorders of the adrenal glands have traditionally presented either with endocrine disturbances caused by the increased secretion of steroid hormones or catecholamines or with symptoms relating to an abdominal mass. However, computed tomography (CT) and ultrasonography have markedly increased the clinician’s dilemma as a wide variety of additional asymptomatic adrenal lesions are now being recognized, many when quite small. Such lesions referred to frequently as “incidentalomas” present a wide array of pathologies, including the so-called nonfunctioning nodule, ‘nonhormonal’ (nonfunctioning) carcinomas, myelolipomas, foci of calcification, metastatic carcinomas, cysts, and pseudocysts. Careful clinical and biochemical analyses are required prior to biopsy or open surgery to establish the correct diagnosis.1 – 4 Hyperfunction of the adrenal cortex (hypercorticalism) mainly manifests itself in three ways: Cushing’s syndrome, the adrenogenital syndrome or primary aldosteronism (Conn’s Syndrome). Bilateral adrenocortical hyperplasia, not neoplasia, is usually the cause of Cushing’s syndrome and the adrenogenital syndrome, while hyperaldosteronism with low plasma renin (Conn’s syndrome) is due generally to an adrenocortical adenoma. While bilateral adrenal medullary hyperplasia may also account for increased catecholamine production, this is more commonly due to a pheochromocytoma, which accounts for hypertension in 0.1–0.6% of all patients. The majority of pheochromocytomas are benign. Malignant adrenocortical and medullary tumors are, therefore, extremely rare representing approximately two and one new cases per million of the population per year, respectively. Their detection before metastases become overt is difficult. Antimetastatic therapy is far from satisfactory. Their early detection, while still limited to the adrenal gland, remains a future goal; surgical removal is the only known truly effective treatment.
ADRENOCORTICAL TUMORS Symptoms related to tumors of the adrenal cortex depend on whether the tumors are functional or nonfunctional (i.e. produce either biologically active or inactive hormones.5 Indeed, a wide variety of steroid hormone products are released by functional lesions, although it is mainly cortisol, aldosterone, and/or sex steroids that are produced. The unregulated hyperproduction of these hormones results in well-defined clinical syndromes. In adult patients with functioning tumors, the most common clinical presentation is Cushing’s syndrome, whereas in children the adrenogenital syndrome or precocious puberty predominates. A mixed picture of Cushing’s syndrome and virilism is not uncommon. Feminization is rare and is predominantly diagnosed in adult men with gynecomastia. An excess of mineralocorticoids (Conn’s syndrome) is seldom due to an adrenocortical carcinoma. ‘Nonfunctional’ tumors release biologically inactive hormone precursors and cause symptoms related to their mass, local extension, or metastases.6
Cushing’s Syndrome Cushing’s syndrome is due to a chronic increment in the level of circulating glucocorticoid hormones, in particular, cortisol. Adrenocortical tumors account for about one-fourth of all adult cases of Cushing’s syndrome, occurring most frequently in females particularly between the ages of 30 and 60 years. In children, tumors are more frequent than bilateral hyperplasia as a cause of the syndrome (see Table 1). Clinical Presentation
Cushing’s syndrome manifests itself as truncal obesity with the characteristic buffalo hump and florid rounded facies. Increased body weight, cutaneous striae, ecchymoses, increasing degree of tiredness and weakness, hypertension, and personality changes (depression or psychotic signs) are also often noted. In women, hirsutism, amenorrhea, and clitoromegaly may occur. Glucose intolerance with polyuria and
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
144
ENDOCRINE TUMORS
Table 1 Incidence of adrenal lesions in Cushing’s syndrome as a function of age.5
Incidence(%) Adrenal lesions Hyperplasia Adenoma Carcinoma
Adults
Children
78 13 9
42 12 46
Reproduced from Neville.5 With kind permission of Springer Science and Business Media.
polydipsia, alterations in immune function, thinning of the skin, osteoporosis, and renal calculi may be evident. These symptoms and signs, caused by glucocorticoid excess, are due to increased gluconeogenesis, inhibition of amino acid uptake and protein synthesis, and accelerated protein breakdown. Clearly, none of these symptoms or signs is unique to Cushing’s syndrome. Classical ‘Cushingoid’ features occur with each etiological type. Adenomas in adults are generally associated with a ‘pure’ form of the syndrome while in children, as noted above, virilism may be an added feature. Carcinomas, on the other hand, in addition to the classical stigmata, are frequently associated with virilism and hypertension. While this chapter is devoted to neoplasia-induced Cushing’s syndrome, it is worth remembering that the syndrome is more frequently due to bilateral adrenocortical hyperplasia (see Table 1),5 including the several forms of nodular hyperplasia. Some have a familial genetic basis while others exhibit adrenocorticotropic hormone (ACTH) independence and cause the overproduction of cortisol through the ectopic expression of neuroendocrine receptors. The features of Cushing’s syndrome can also be an iatrogenic disorder due to the therapeutic administration of excessive amounts of glucocorticoids or ACTH, the ectopic production of ACTH by tumors such as those of the lung, and may also be met in chronic alcoholism due to defective elimination of steroid hormones and their metabolites. Pathology Adenomas These tumors are generally single and unilateral. If bilateral and/or multiple, the possibility of bilateral nodular hyperplasia should be excluded.7 However, true bilateral adenomas have been reported. In one case, there were several years between the appearance of the first and second tumors. The appearances of the attached gland are the clue to the diagnosis. It is atrophic with adenomas and hyperplastic with nodular disorders. Adenomas are small, rounded lesions with a yellow cut surface in which brown foci are present. Occasionally, a black cut surface is noted in the so-called ‘black’ adenoma. Necrosis and hemorrhage are rare. They usually weigh less than 50 g and often less than 30 g. However, adenomas weighing up to 250 g have been recorded. Microscopically, a tenuous capsule surrounds the lesion. The yellow areas correspond to lipid-laden clear cells, morphologically similar to the cells of the zona fasciculata of the normal adrenal cortex (see Figure 1). The brown
0.1 mm
Figure 1 Cushing’s syndrome – adrenocortical adenoma (5 g). A typical adenoma is illustrated composed predominantly of clear lipid-laden zona fasciculata type cells arranged in small nests. Such tumors frequently contain frequently small foci of eosinophilic lipid-sparse compact cells (upper left), similar to those of the normal zona reticularis (H&E, original magnification:×240).
areas consist of compact cells with eosinophilic lipid-poor granular cytoplasm, similar to the cells of the normal zona reticularis. Usually, clear cells predominate but compact cells are the sole or dominant component of black adenomas. The tumor cells are arranged in small cords or alveoli. Nuclear or cellular pleomorphism is uncommon. The ipsilateral and contralateral adrenal cortex associated with adenomas is always atrophic. Carcinomas These tend to be large lesions, most weighing in excess of 100 g and measuring over 6 cm in diameter. This may be related, on a cellular basis, to their relatively ineffective production of cortisol. However, not all carcinomas are large, and weight alone cannot be rigidly used to assist in the differentiation between benign and malignant lesions. Carcinomas tend to affect the right more often than the left gland and are usually encapsulated and soft in consistency, often with a pink lobulated cut surface. Areas of necrosis, hemorrhage, cystic change, and calcification are not uncommon, and occur more frequently as the lesions increase in size and weight. In larger tumors, there may be obvious microscopic evidence of capsular penetration with infiltration of the related neighboring tissues including the ipsilateral adrenal, kidney, liver, diaphragm, and venous system. Satellite tumor nodules may also be present. Microscopically, the typical large carcinoma consists solely of compact cells with eosinophilic, granular, lipid-poor cytoplasm, grouped in large alveoli, sheets, or trabeculae separated by a fine fibrovascular stroma. In some cases, viable tumor cells form cords or tubules, which tend to surround blood vessels (see Figures 2 and 3). Extensive areas of necrosis may be present. Characteristically, there is cellular and nuclear pleomorphism with marked nuclear vesicularity and one or more prominent nucleoli. Bizarre and giant cellular forms may be present in some lesions; others exhibit little
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS
145
Metastases, when they evolve are generally detected in the regional and mediastinal lymph nodes, bones, contralateral adrenal gland and, particularly, the lungs and liver. Prognosis
Adenomas are cured by surgery. A few examples of spontaneous remissions in association with adenomas have been recorded.5,8 Survival in patients with Cushing’s syndrome caused by adrenocortical carcinoma is seldom prolonged beyond 3 years. Current clinical investigation is addressing the concept that adjuvant drug therapy may help improve the prognosis, but at present the role of adjuvant therapy is unproven. 0.1 mm
Figure 2 Cushing’s syndrome – adrenocortical carcinoma (70 g). A low-power view of a typical carcinoma, in which there are large sheets of compact eosinophilic cells of the zona reticularis type forming large trabeculae punctuated by vascular sinusoids. This kind of appearance is typical of the large carcinomas of Cushing’s syndrome and the adrenogenital syndrome, as well as ‘nonfunctioning’ carcinomas (H&E, original magnification:×50).
Adrenogenital Syndrome The adrenogenital syndrome is associated with abnormal levels of circulating sex steroid hormones and refers to all cases of sexual precocity and heterosexual abnormalities due to adrenocortical dysfunction. Neoplasms, both benign and malignant, can give rise to the syndrome and are most often detected in children between the ages of 2 and 7 years rather than in adults. Females are more affected than males, with a female to male ratio of 3 : 1, and the left gland is involved more frequently.5,9 Clinical Presentation
0.1 mm
Figure 3 Cushing’s syndrome – adrenocortical carcinoma (386 g). The cellular pattern of a typical carcinoma is shown at higher power, where the compact cells form trabeculae punctuated by vascular sinusoids. The cells are compact in type with lipid-sparse eosinophilic cytoplasm. The nuclei show pleomorphism with increased vesicularity and one or more prominent nucleoli (H&E, original magnification:×240).
nuclear abnormality. Mitotic figures may or may not be seen. Although important in assisting with reaching a diagnosis of malignancy in the absence of proved metastases, they are not necessarily a wholly reliable guide to malignancy. Vascular invasion through the walls of blood vessels is a guide to malignancy, but is uncommon. Tumor cells may be observed within vessels or thrombi containing tumor cells may be present in the tumor sinusoids or the main adrenal vein. The use of these and other features to assist in the distinction between benign and malignant tumors (as also are the functional changes that result in cortisol overproduction) are discussed subsequently (see below).
In female children, virilization presents with clitoromegaly, hirsutism, and enhanced growth. In adult females, oligomenorrhea followed by amenorrhea, ‘masculine’ baldness, hirsutism, deepening voice, breast atrophy, decreased libido, and increased musculature occur. In male children, virilizing tumors are associated with precocious puberty, reduced gonadotropin levels and often spermatogenic arrest; lesions with corresponding functional properties in adult men usually cause no overt endocrine symptoms. In some cases, signs of Cushing’s syndrome may be present; hypertension is also a frequent finding in conjunction with virilizing tumors. Problems in differential diagnosis include late-onset congenital adrenal hyperplasia, gonadal tumors, polycystic ovarian syndrome, and central nervous system tumors causing precocious puberty as well as idiopathic sexual precocity. Idiopathic hirsutism, and ectopic human chorionic gonadotropin (HCG) production by lung cancers in particular are further causes. Feminization is rare in this context; carcinomas are the more common cause (see Table 2). Most are diagnosed in men aged 20 to 30 years, but have occasionally been reported in female children and postmenopausal women. The syndrome in adult males presents as bilateral gynecomastia with testicular atrophy. In prepubertal females, the tumors cause isosexual precocity. Pathology
Unlike tumors associated with Cushing’s syndrome, there is a considerable overlap in the size and weight of virilizing adenomas and carcinomas.5 Although large tumors are again more likely to be malignant, adenomas have been known to weight as much as 1.5 kg. All tumors are generally
146
ENDOCRINE TUMORS
Table 2 Age incidence of 80 adrenocortical tumors causing feminization.5
Age (years) 0 – 10 11 – 20 21 – 30 31 – 50 51 – 60 >60 Total
Adenomas
Carcinomas
7 – 2 2 2 – 13(16%)
4 3 14 29 14 3 67(84%)
Reproduced from Neville.5 With kind permission of Springer Science and Business Media.
encapsulated and have a reddish-brown cut surface and, with increasing size, areas of necrosis, hemorrhage, cystic change, and foci of calcification can often be observed. Although testosterone synthesis has been recorded in some cases even in prepubertal males, most tumors produce greater amounts of its precursors, such as dehydroisoandrosterone (DHA) and/or its sulfate and androstenedione.10,11 Compact cells, observable in conventionally stained histology sections, are generally the sole cell type. In adenomas, the cells, which are arranged in short cords or acini, have single, uniform, vesicular nuclei and may be of normal size (see Figure 4). Individual cell necrosis may be present. Apoptotic features are frequently noted. As tumors increase in weight, the component cells and their histological classification as benign or malignant is extremely difficult. The relatively clear-cut distinction that can be noted in Cushing’s syndrome is not apparent in virilizing lesions (see Figure 5). In proven carcinomas, the compact cells may have a syncytial arrangement or form large alveoli with prominent vascular sinusoids. In addition, there may be large areas of necrosis, and vesicular, pleomorphic and/or enlarged nuclei with prominent nucleoli are usually found. Bizarre and giant
0.1 mm
Figure 5 Virilization-adrenocortical adenoma (225 g). This is a compact cell lesion with cells containing eosinophilic lipid-sparse cytoplasm forming large trabeculae. The cells show some pleomorphism, but the nuclei are not vesicular, nor do they have prominent nucleoli. This particular lesion is one of those which is difficult to diagnose as either benign or malignant. The young man from whom this tumor was removed was alive and well some 20 years after surgery (H&E original magnification:×240).
forms of cells may occur; mitoses, however, are seldom prominent. The appearances are, thus, similar to carcinomas causing Cushing’s syndrome. Tumors associated with feminization have varied in weight from 10 g to more than 2 kg. Their gross and histological features are indistinguishable from large virilizing or cortisol-secreting tumors. Long periods of reappraisal are necessary, as the development of metastases can be delayed for up to 8 years. Consequently, all feminizing tumors with the possible exception of small prepubertal tumors should be regarded as carcinomas, irrespective of histology, and managed accordingly. In functional terms, the differences between tumors causing virilism and those causing feminization are minimal and are discussed in further detail later in the chapter. The steroid-secretion patterns of feminizing tumors reflect various alterations that channel steroidogenesis toward sex steroid formation. Estrogen-excretion rates rise to adult female levels or higher. C-19 steroid levels are also elevated in most patients, with DHA accounting for up to 50% of the total urinary 17-ketosteroids. Occasional feminizing tumors also produce cortisol and may be additionally associated with Cushing’s syndrome.12 In one case, an adrenal carcinoma that caused virilism in a child as its primary presenting feature recurred with feminization after treatment.13 Prognosis
0.1 mm Figure 4 Virilization-adrenocortical adenoma (25 g). A characteristic benign cortical adenoma associated with virilism is shown. It contain compact cells similar to those of the normal zona reticularis arranged in short strands and nests. Nuclear pleomorphism is minimal; most of the nuclei are regular, uniform, and hyperchromatic (H&E original magnification: ×350).
The signs of virilization abate over several months following the curative surgical removal of an adenoma. While excellent results following surgery alone in children have been recorded in some reports, others express guarded optimism for the outcome even of small (∼5 cm) lesions.9 Urinary steroid profiles in children are not reliable guides to the future outcome.14
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS
Recurrences in malignant cases are fatal. However, as a group such virilizing carcinomas tend to pursue a slower course than their counterparts causing Cushing’s syndrome. Metastases from feminizing lesions may also be slow to emerge but with the exception of the small prepubertal lesions most of which seem to be benign, all other feminizing tumors should be regarded as carcinomas.
Hyperaldosteronism with Low Plasma Renin (Conn’s Syndrome) The syndrome of hyperaldosteronism with low plasma renin (so-called ‘primary’ aldosteronism) was recognized as a specific clinicopathological entity in 1955.15 It is typified by hypokalemic alkalosis, hypertension, and muscle weakness, and is usually referred to as Conn’s syndrome, particularly when associated with a solitary aldosterone-secreting adrenal adenoma. Similar symptoms also occur when steroids with mineralocorticoid activity other than aldosterone (e.g. deoxycorticosterone, corticosterone) are secreted in excess, as it sometimes occurs in association with adrenocortical tumors. The hypokalemic state, which often waxes and wanes during the course of the disease, is responsible for most of the symptoms and signs. These include muscular weakness, nocturia, persistent frontal headaches, polydipsia, paraesthesia, visual disturbances, temporary paralysis, cramps, and tetany. A volume-dependent hypertension may develop and be quite severe. However, it is usually not malignant in intensity and does not cause retinopathy. It, nevertheless, causes a major threat to life with ensuing nephrosclerois, cardiac enlargement, and an increased risk for cardiovascular and/or cerebral accidents. The physical signs and symptoms of primary hyperaldosteronism are not, however, always distinguishable from ‘essential’ hypertension, without laboratory studies. Until the 1990s, the consensus was that 1–2% of unselected hypertensive patients had demonstrable primary aldosteronism.16 With the ready ability to measure both plasma aldosterone and renin levels, figures of between 5 and 40% have been cited more recently as its incidence in a population of hypertensive patients.17
147
Table 3 Primary aldosteronism: analysis of 240 patients with benign tumors of the adrenal cortex.5
Sex incidence(%) Modal age incidence Site of tumors (left : right) Single Multiple Weight of tumors <2 g (%) <4 g (%)
Male
Female
30 30 – 50
70 30 – 50
1:1 1:4
7:4 4:1 34 58
Reproduced from Neville.5 With kind permission of Springer Science and Business Media.
enzymes – the CYP11B2 (aldosterone synthase), CYP17 (17 α-hydroxylase) and CYP 11B1 (11-β-hydroxylase) helps make this distinction and enables a clearer and surer diagnosis of the lesion under study. Transcription factors such as Nurr-1 are also elevated in such lesions and are associated with raised CYP11B2 levels.19 The typical small adenoma associated with this syndrome is a circumscribed, encapsulated lesion with a distinctive golden-yellow cut surface. Carcinomas, on the other hand, present gross appearances indistinguishable from those causing other forms of hypercorticalism.5,18 As such in tumors that produce aldosterone, one would anticipate that their component cells would be of the zona glomerulosa type i.e. the cell type normally responsible for aldosterone production. This is most often not the case. Adenomas have a characteristic histological appearance typified by their protean cellular morphology. Four cell types occur in such lesions; large and small clear lipid-laden cells, zona glomerulosa type cells, and zona reticularis type (compact) cells (see Figure 6). Very few adenomas consist of a single cell type; indeed, all four types may be found in one tumor.
Pathology
This syndrome is associated with three types of adrenal change: tumor, hyperplasia of the zona glomerulosa, and nodules which may or may not accompany the hyperplastic and/or neoplastic changes.5,18 Approximately 65% of patients with this disorder have adrenal tumors.5 Nontumorous hyperaldosteronism, which accounts for the remainder and which tends to affect an older age-group, has been discussed elsewhere in detail.5 Almost all adrenal tumors causing hyperaldosteronism are benign. Ninety-two percent are unilateral and single. When multiple (8%), they are still usually unilateral. Many weigh less than 2 g (see Table 3). Nevertheless, a few adenomas can weigh as much as 75 g when they overlap with the weights of carcinomas associated with hyperaldosteronism.18 When multiple it is important to distinguish between an adenoma and a nodule; the former is associated with the production of aldosterone, the latter does not do so. The ability to demonstrate by in situ hybridization of the genes for the following
0.1 mm
Figure 6 Hyperaldosteronism with low plasma renin – adrenocortical adenoma (1 g). Cells with clear lipid-laden cytoplasm comprise the dominant cell type in this tumor. Large and intermediate clear cell types are both seen together with a small focus of zona glomerulosa type cells in relation to the fibrovascular trabeculae in the lower aspect of the photograph (H&E, original magnification:×200).
148
ENDOCRINE TUMORS
The most common pattern consists of large, lipid-laden clear cells similar to those of the normal zona fasciculata in size and nuclear/cytoplasmic ratio; they are arranged in small cords or alveoli separated by fine, fibrovascular trabeculae. The nuclei are vesicular, often with inclusions, and pleomorphism may or may not be seen. Such cells may occur alone but are more commonly found in association with smaller lipid-rich cells, which possess a vesicular nucleus and a nuclear: cytoplasmic ratio similar to that of zona glomerulosa cells. This cell type, referred to as an intermediate (hybrid cell), seems to have the cytological characteristics of both zona glomerulosa and zona fasciculata cells. In many tumors, zona glomerulosa type cells are also present. Rarely, they may be the sole component. Generally, such cells are seen in nests or short cords around the periphery of the lesion dipping in a tonguelike manner into the body of the tumor. Groups of compact cells are occasionally noted in association with the other cell types. Cellular heterogeneity and occasional foci of nuclear pleomorphism should not mislead one into a diagnosis of malignancy, as these appearances are entirely characteristic of benign lesions causing Conn’s syndrome. Proven carcinomas associated with hyperaldosteronism are rare. Most are large, weighing more than 500 g although lesions weighing between 30 and 100 g have subsequently metastasized. Characteristically, the cells are zona glomerulosa in type or are similar to the intermediate (hybrid) cell type found in adenomas, and are arranged in large alveoli or trabeculae separated by prominent vascular sinusoids (see Figure 7). Necrosis may be marked, and pleomorphism, mitotic activity, and hemorrhage may or may not be present. In some tumors, the cellular appearances are indistinguishable from that of carcinomas associated with other forms of hypercorticalism. This latter appearance tends to be associated with excess production of deoxycorticosterone and corticosterone to account for the syndrome. These steroids
0.1 mm
Figure 7 Hyperaldosteronism with low plasma renin – adrenocortical carcinoma (2000 g). The features of a typical adrenocortical carcinoma associated with hyperaldosteronism are shown. The cells are of the hybrid type and form large trabeculae punctuated by thick-walled fibrovascular trabeculae (H&E, original magnification:×160).
are formed by the inner zone cells of the normal cortex. Tumors secreting them are, therefore, probably not of glomerulosa origin but are akin to those associated with Cushing’s syndrome and the adrenogenital syndrome. Hence, they show typical compact cell carcinoma-like appearances (see Figures 2 and 3). Nonetheless, carcinomas producing aldosterone and cortisol have been recorded.20 Prognosis
Although there are rare exceptions,21 carcinomas causing this syndrome pursue a most aggressive course. Our personal experience has shown that survival seldom exceeds 1 year. Adenomas are cured by surgery. However, the hypertensive state may persist. Interpretation of the Structural and Functional Properties of Tumors in Primary Aldosteronism
The morphology of the typical adenoma of this syndrome appears at first glance to be an enigma –composed of zona fasciculata type (cortisol-producing) cells mainly and yet associated with increased aldosterone production. There is no doubt about the tumor being the main source of the raised aldosterone levels, although it also contains cortisol and corticosterone. Recent in situ hybridization techniques have shown that the enzymes necessary for both types of steroid production are present in such tumors. Nodules do not contain the enzyme complement needed to form aldosterone.22 In the normal gland, zona glomerulosa cells are found subcapsularly and aldosterone is produced only at this peripheral site. In benign tumors, when they do possess recognizable zona glomerulosa cells, these tumor cells are also found subcapsularly or in relation to the penetrating fibrovascular trabeculae. It would appear that this anatomical situation is needed for aldosterone production. Normal zona glomerulosa cells in tissue culture initially produce aldosterone but then rapidly modulate to secrete only glucocorticoids, such as cortisol. Ultrastructurally, they also undergo a morphological transition to cells of the zona fasciculata type in tissue culture.23 Benign aldosterone-producing tumors, irrespective of morphology, behave functionally in a similar manner when introduced into culture, i.e. aldosterone production is not sustained and cortisol becomes the main product. This change, from glomerulosa to fasciculata type cells, explains the structural and functional heterogeneity of the aldosterone-producing adenoma. This explanation for the cellular heterogeneity of such tumors also raises the intriguing, but probably unprovable, possibility that some ‘nonfunctioning’ nodules so frequently associated with essential hypertension may have been aldosterone-producing adenomas initially. Such lesions, as they progress, eventually become converted entirely to zona fasciculata type cells without aldosterone formation but the hypertension persists.
‘Nonhormonal’ Adrenocortical Carcinomas Adrenocortical carcinomas are considered ‘nonhormonal’ if there is no evidence of endocrine signs or symptoms. Such tumors are not steroidogenically inert; rather, they fail to
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS
form biologically active hormones, releasing instead inactive precursor steroids and/or their metabolites. Such lesions are uncommon, usually diagnosed during the fifth to seventh decades of life, although children are not exempt. They occur in males twice as often as in females.6 This type of tumor has gross features similar to those of functioning adrenocortical carcinomas although they tend to be larger, probably because they fail to produce clinical signs and symptoms of hormonal excess at an early stage. Tumors weighing in excess of 1 kg are common. Microscopically, the predominant cell is of the compact type; although occasional tumors contain more clear cells than compact cells, possibly because of defective cholesterol utilization. Occasionally, especially in adult females, an oncocytic variant is found.24,25 Such lesions are highly characteristic. The tumor cells (oncocytes) possess dense eosinophilic granular cytoplasm containing abundant mitochondria readily demonstrated with suitable antisera that detect the mitochondrial MES-13 antigens.26 Such lesions do not demonstrate epithelial antigens such as EMA. Occasionally, they may be mistaken for pheochromocytomas but staining techniques (see below) will readily provide the correct diagnosis. Prognosis
The prognosis for ‘nonfunctioning’ carcinomas, as a group, is poor. While about one-third of patients will survive beyond 3 years, most die within the first year of diagnosis.6 Most oncocytic tumors appear to be benign or of low-grade malignancy.25 In the follow-up of patients with carcinomas without overt function, it is essential to characterize the nature of the precursors being released (see below) and to use them as index substances to monitor the clinical course.
Etiological and Growth-Promoting Factors of Adrenocortical Tumors While smoking in men and the use of the contraceptive pill in women have been postulated as causative of adrenocortical tumors, the evolution of our fundamental understanding of the genesis of such lesions and the molecules involved in promoting their progression has been rendered difficult by the paucity of adrenocortical tumors. In addition, as pointed out by Koch and colleagues,27 many of the studies have been carried out using tumors where the subsequent follow-up has been relatively short so that the diagnosis of malignancy or benignity based on histology alone has to be questioned. Hence, the conclusions as to which factors may be involved at this time must be viewed as speculative. Nevertheless, clues have been obtained from the study of familial syndromes. Adrenal tumors are known to occur in several inherited disorders, particularly as part of the Li-Fraumeni and Beckwith-Weideman syndromes, where germline mutations of p53 and modifications at chromosome 11p15.5 may be found.27 – 29 Information remains scant about the genetic changes which initiate and subsequently modulate the behavior of sporadic adrenal tumors. This is despite the fact that such tumors have been subjected to diverse gene screening techniques, in addition to the assessment of whether known specific
149
oncogenes or suppressor genes involved in other tumor systems are part of the genesis of adrenal cortical tumors. Current thinking views carcinomas as arising from preexisting adenomas. An early initiating event appears to be activation of a proto-oncogene(s) on chromosome 4.27,30 Progression thereafter may involve the subsequent activation of oncogenes on chromosomes 5 and 12 and inactivation of tumor suppressor genes on chromosomes 1 and 17. Other genetic abnormalities may involve loss of material on chromosomes 2 and 11 with gains on chromosomes 4 and 19.27,31 Various analyses have concluded that the majority of adenomas exhibit polyclonality, whereas carcinomas are monoclonal.27 However, such a conclusion must be viewed with caution as the distinction of an adenoma from an adrenal nodule has not been definitively achieved always. One of the most important derangements that may provide growth advantage involves changes at chromosome 11p15.5.29,31,32 This area is intimately involved with the insulin-like growth factor (IGF) system. The IGF-2 gene and its growth factor product, generally in its precursor form, are overexpressed in the majority of sporadic adrenocortical carcinomas, most often due to paternal isodisomy. This chromosome region also harbors two putative growth suppressor genes (H 19 and p57/KIP2 ), whose loss may be contributory.27 Some reports state that adrenal adenomas do not exhibit these alterations of IGF-2 and 11p15.5, but p57 loss has been reported in them. Hence, such changes are not cancer specific. The IGF-1 receptor, the partner for IGF-2, may also be raised in carcinomas thereby providing a local autocrine/paracrine feedback loop. The binding protein, IGFBP-2 may also be increased but it may enhance cell proliferation by an IGF-2 independent mechanism. Taken together these molecular markers may permit a more accurate diagnosis of malignancy and provide a useful prognostic assessment when present in elevated quantities and their levels or presence can be measured in a particular tumor. Other growth-promoting factors may also be detected in carcinomas, including the EGF receptor but not EGF itself. However, an alternate ligand, TGF-α, is frequently present in raised quantities.33 Her-2/η protein is not a feature of adrenal lesions.34 The vascular endothelial growth factor (VEGF) system may also contribute to growth through its effects by promoting angiogenesis. CXC chemokines and the orexin system are further examples of growth-promoting factors reported as occurring in such tumors.35,36 Down-regulation of p21 and p16 may be involved from time to time37 but oncogene mutations or overexpression of genes such as RET and ras are absent or are rare occurrences in carcinoma.38 The gene product p53 is frequently involved. Loss of its chromosomal locus or somatic mutations may occur in as many as 70% of carcinomas. Current opinion views p53 involvement in progression rather than tumor initiation.27,39 A particular germline mutation of p53 is found with high frequency in sporadic adrenocortical tumors in southern Brazil and carries a poor prognosis for adult but not pediatric tumors.40,41
150
ENDOCRINE TUMORS
Hence, while little is known about the early events in the tumorigenic process, altered 11p15.5 function would seem to play a key role and should form part of all future routine investigations of adrenocortical tumors. Although further data are required, this workup may lead to the derivation of rational therapeutic strategies predicated on gene targeting.
Differentiation of Benign from Malignant Adrenocortical Tumors While the molecular alterations discussed above enable our appreciation and understanding of some of the differences between benign and malignant tumors, not all carcinomas (proven by their ability to metastasize) contain distinguishing genetic anomalies. Hence, in many instances, the distinction between benign and malignant tumors and the attribution of prognosis is the role of the histopathologist. Small benign tumors found in each case of hypercorticalism and nonfunctioning nodules (sometimes referred to as adenomas and which become rather frequent with aging and hypertension) are quite typical and their diagnosis presents no difficulty. Additionally, the classical carcinoma found in many cases of hypercorticalism is also readily identified. It is in the diagnosis of the tumors between 5 and 10 cm in diameter and between 100 and 500 g in weight where the greatest difficulty is encountered. Prominent nuclear pleomorphism, a high nuclear/cytoplasmic ratio and enlarged vesicular nuclei with one or more prominent nucleoli, atypical mitoses, together with distinct areas of necrosis in contrast with single cell apoptosis, are valuable but not absolute criteria. The presence of true capsular penetration and/or vascular wall invasion in contrast with tumor cells just present in the blood vessels is also helpful. Numerous attempts have been made to improve the histological assessment of the prognosis of such lesions. These, while helpful, remain subjective and often do not consider the existence of heterogeneity in the different areas of any one tumor. One of the best proposed schemes and one used by numerous workers since its first description is the method proposed by Weiss and his colleagues.42,43 This scheme measures nine different histological parameters along with the tumor size and weight (see Table 4). In the final analysis, only one of the chosen variables, mitotic rate, possessed a statistically significant association in predicting the outcome from proven examples of carcinoma. Several workers, using the same and other closely related Table 4 Histological criteria to assist in distinguishing benign from malignant adrenocortical carcinomas.a
High nuclear grade Mitotic rate >5 per 50 HP fields Atypical mitotic figures Eosinophilic tumor cell cytoplasm (>75% of tumor cells) Diffuse architecture (>33% of tumor) Necrosis Venous invasion Sinusoidal invasion Capsular incasion a
The presence of three or more features highly correlates with subsequent malignant behavior.
criteria and assessing the mitotic rate by conventional histology or through the use of a Ki67 index, have reached similar conclusions.44 – 49 While attempting to improve prognostication from histological material, several groups have added molecularly defined indices to a morphological scoring system. One report in particular examined a series of cell cycle proteins or indices–Ki67, p53, mdm2, Bcl-2, cyclin D1, p21, and p27–and added the results to a morphological score. Mitotic activity (>5 per HP field) and evidence of adjacent organ invasion were the only prognostic indices for tumors that later exhibited distant metastases.50 p53 up-regulation has been noted in many (but not all) carcinomas, especially in those with overt function. This has not been shown, to date, in adenomas. This up-regulation appears to be a late factor, more involved in tumor progression.49,51 The overproduction of cyclin E,52 the up-regulated IGF-2 gene and 17p13 loss of heterozygosity (LOH, where the loss of a suppressor gene is considered to reside) are all currently strong prognostic predictors of malignancy. Other approaches have involved the use of cDNA arrays when several gene clusters have been identified as having prognostic value.53,54 One was closely related to an IGF-2 cluster. However, such studies do not yet have practical utility for clinical application. At present, in addition to the morphological approaches it may be useful to search for LOH at 17p13 and 11p15.5 together with demonstrating whether IFG-2 is up-regulated, in an attempt to be more precise in the prognostication of a particular tumor.55 Further indices are needed and with better understanding of molecular changes in these lesions and the increasing ability to demonstrate molecules in archival material it is hoped that improved prognostic indices will be forthcoming.
Steroidogenesis, its Diversity in Adrenocortical Tumors and Relation to the Clinical Syndromes The functional zoning of the adrenal cortex remains a fascinating enigma with regard to the various types of hormones that are produced by the various cell types in the several discrete zones and how the processes are controlled. This is also true for the tumors of this region, many of which have an identical cellular morphology and yet produce different steroid hormones and clinical effects. Fundamental molecular and biochemical studies are now providing some insights regarding the changes in tumors and how the different pathways are controlled, both in the normal gland and its neoplasms.56 – 60 Tumors are frequently stated to be ACTH-independent or autonomous, yet no activating mutation or increased gene expression of the ACTH receptor has been recorded.61 – 63 In fact, down-regulation of this receptor is seen. It now appears that one explanation for ACTH independence may be that some tumors express one or more ‘ectopic’ or overactive ectopic receptors whose function appears to be to recapitulate the effects of ACTH.64 – 69 Such receptors have been found in adrenal adenomas and in a particular form of bilateral macronodular hyperplasia causing Cushing’s syndrome.70,71 The receptors include those for gastric inhibitory polypeptide
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS
(GIP), vasoactive intestinal peptide (VIP), serotonin, luteinizing hormone-human chorionic gonadotropin (LH-HCG), leptin, IL-1, angiotensin II, and those of the orexin system. Whether this situation pertains to carcinomas remains to be reported but the differential effects of the LH-HCG system might be operational in cases of sex hormone excess. Recently, CXC chemokines have been recorded as being produced in an adrenal carcinoma causing a preclinical form of Cushing’s syndrome.35 They seem to influence hormone production and also tumor growth. Taken together, these findings have important therapeutic implications.69 No mutations of the gene coding for the P-450 cytochrome series of adrenal enzymes responsible for the conversion of cholesterol into steroid hormones have been found in adrenal tumors (including carcinomas). Thus, there is no explanation for the predominance of certain types of products in different syndromes.72 This has led to the search for other mechanisms to account for the steroid miscellany. The last few years have seen the recognition of a series of transcription factors that are involved in the control of steroid production. Their alterations in tumors go some way to begin to explain the different syndromes. Such factors include COUP-TP, DAX-1, the steroidogenic factor-1 (SF-1), and several members of the orphan nuclear hormone receptor super-family.73 – 76 For example, in cortisol-producing adenomas, DAX-1 is found to be down-regulated. This results in the up-regulation of the enzyme 17-α-hydroxylase (CYP17) leading to increased cortisol production. In other situations where synthesis is channeled in a direction other than toward cortisol, DAX-1 is up-regulated and the level of the CYP17 is lowered. Additionally, adenomas associated with Cushing’s syndrome have been reported to contain suppressed levels of the type 2 11-β-hydroxysteroid dehydrogenase, thereby creating a higher cortisol output through inhibition of its conversion to its inactive product cortisone.77,78 With respect to tumors associated with the adrenogenital syndrome the activity of the side chain-cleaving enzyme, 17.20-lyase necessary to produce androgens, has been shown to be elevated thereby channeling production toward androgens and hence virilism.79 Other transcriptional factors such as GATA-6 are involved in androgen production.80,81 In virilizing lesions, its up-regulation may account for the overproduction of C-19 steroids such as DHA sulfate.82 The initial steps in steroid hormone production from cholesterol are also influenced by these transcription factors and other orphan nuclear receptors most importantly nerve growth factor-induced clone B (NGFI-B), and are possible sites of derangement in ‘nonfunctioning’ tumors leading to the higher production of functionally inactive precursors, such as 17-hydroxyprogesterone and 11-deoxycortisol60,83 Further understanding of these and, as yet, undiscovered factors may in time help reveal the basis where by diversity of hormone production is a feature not only of tumors but also of normal adrenocortical cells. The tumor changes found in hyperaldosteronism have already been outlined.
Diagnostic Investigative Tests and Staging Biochemical investigations and imaging techniques form the cornerstone for the differential diagnosis of all adrenal
151
Table 5 Primary investigations in patients suspected of harboring an adrenal tumor including those detected initially by ultrasonography or CT scans (incidentalomas).
Urinary 24-hour free cortisol level Urinary catecholamines, HMA and VMA levels Plasma-free metanephrines Plasma aldosterone, testosterone Serum dehydroisoandrosterone sulfate (DHAS) Serum electrolytes Short overnight dexamethasone test If tumor > 4 cm Urinary steroid profile Guided biopsy using ultrasonography Imaging (MRI, positron emission tomography [PET])
lesions, including all types of incidentalomas (see p. 18), which are being increasingly recognized as a result of magnetic resonance imaging (MRI) and /or CT scans for other unrelated disorders. Table 5 illustrates a series of tests of value in the initial investigation of any subject suspected of harboring an adrenal tumor. Biochemical Diagnosis
Urinary free cortisol is a useful screening test for Cushing’s syndrome, but results may be normal in 8 to 15% of patients. Minimally elevated results should always be confirmed by further testing before making a diagnosis of Cushing’s syndrome. The overnight dexamethasone suppression test is a useful outpatient-screening method. Various doses have been used; but 1 mg of dexamethasone is usually given at midnight. A normal response is a plasma cortisol <140 nmol L−1 (5 µg dL−1 ) between 8 and 9 a.m. the following morning. False-positive results with this dose occur in about 12% of patients; the false-negative rate is <2%. With this test, a cortisol value of <15 nmol L−1 (2 µg dL−1 ) effectively excludes Cushing’s syndrome. The high-dose dexamethasone suppression test should be used in patients in whom other causes of Cushing’s disease need to be ruled out. Currently this test involves giving 2 mg dexamethasone every 6 hours for 48 hours; urinary free cortisol is measured at 48 hours and a greater than 50% suppression of urinary and plasma cortisol is a positive response.84,85 The corticotrophin releasing hormone (CRH) test involves giving this 41 amino acid peptide intravenously in a dose of 1 µg kg−1 body weight or a single dose of 100 µg, usually in the morning and blood samples for ACTH and cortisol are taken every 15 minutes for 1 to 2 hours following its administration. Patients with Cushing’s disease show an exaggerated response with ACTH increase of 50% and a cortisol rise greater than 20% over baseline values. No response is seen in the ectopic ACTH syndrome and a blunted response is noted with adrenal tumors.86,87 Biochemical investigations for patients with virilizing symptoms and/or signs and suspected of having an adrenal lesion should include assays for serum testosterone, sex hormone binding globulin (SHBG) (to obviate spurious results due to variations in binding capacity – e.g., oral contraceptives, liver disease), DHAS and 17α-hydroxyprogesterone.
152
ENDOCRINE TUMORS
The steroid-secretion patterns of feminizing tumors are indicative of enzyme defects, which channel steroidogenesis toward sex steroid formation. Estrogen-excretion rates may rise in men to adult female levels, or above. When primary aldosteronism is suspected, any diuretic should be discontinued for at least 4 weeks prior to biochemical investigations being started. Hypokalemia should be corrected and the patient kept on a liberal sodium intake (100 mmol day−1 ). When the serum potassium is normal, a 24-hour collection is taken to measure aldosterone and renin. Plasma aldosterone, renin, and 18-hydroxycorticosterone levels should also be measured, with the samples being drawn in the supine (08.00) and after 4 hours in the erect positions (12.00). In patients with primary aldosteronism caused by adrenocortical adenomas, the aldosterone and 18-hydroxycorticosterone levels are significantly lowered after 4 hours in the erect posture. Plasma renin activity remains suppressed when erect instead of rising as in normal subjects.88 The so-called nonfunctioning adrenocortical carcinomas often excrete biologically inactive steroid hormone precursors, such as 3α-hydroxy-5-ene steroids and tetrahydro-11deoxycortisol. Determination of such steroid profiles can serve as an important adjunct in the assessment of whether a particular adrenal tumor is malignant. Such profiles also indicate the steroids that are most useful to measure sequentially in the follow-up phase for monitoring the efficacy of therapy and for earlier detection of recurrences and/or metastases.89
Figure 8 Computed tomography of a patient with adrenocortical cancer. Notice the large tumor in the left adrenal.
Imaging Techniques Computed Tomography CT is the most commonly used modality to detect adrenal masses.90 Depending on the patients’ age, up to 10% of abdominal CT scans obtained for unrelated reasons will demonstrate an unsuspected adrenal mass. Most lesions smaller than 4 cm are benign but there have been exceptions; those above 6 cm in size are not infrequently found to be adrenocortical carcinomas.89 – 91 However, size alone cannot be used to exclude malignancy. Attenuation measurements or CT densitometry using unenhanced CT (prior to intravenous administration of contrast medium) may help differentiate benign from malignant adrenocortical lesions. The majority of adrenal adenomas contain large amounts of intracytoplasmic lipid that account for their low attenuation on unenhanced CT scan. A threshold of 10 Houndsfield Units (HU) or less corresponds to a sensitivity of 71% and specificity of 98% for the diagnosis of benign lipid-rich adrenal adenomas. If the density is above 10 HU, intravenous contrast is administered, and both venous phase and delayed images are required. This can further characterize the adrenal lesions. Such new contrast enhanced CT scans are capable of visualizing lesions less than 5 mm90 – 92 (see Figure 8) Magnetic Resonance Imaging The high contrast resolution and the multiplanar imaging capability of MRI allow the detection and characterization of many adrenal masses. Both T1 and T2 relaxation times have been used to assist in the differentiation between adenomas, adrenal metastases, and pheochromocytomas. In general, malignant masses
Figure 9 Magnetic Resonance Imaging (MRI) of a patient with adrenocortical cancer in the right adrenal (notice the arrow).
have a higher fluid content and therefore appear brighter on T2-weighted images. Avid enhancement with delayed washout and high signal intensity on T2-weighted images are features often shared by adrenocortical carcinomas and adrenal metastases, which usually contain fewer lipids than adenomas. Chemical-shift MRI is used to detect intracellular lipid content within structures by showing a loss of signal intensity between in-phase and out-of-phase gradient-eco sequences. This principle has been used to distinguish benign lipid-containing adenomas from nonadenomas. Most adenomas, therefore, can be distinguished from malignant masses as the latter in most instances do not contain lipids90 – 92 (see Figure 9). Positron Emission Tomography 18 F –Fluorodeoxyglucose PET (FDG-PET) has been suggested for the characterization of adrenal masses in patients with clinically silent adrenal masses or for the workup of a patient with a nonprimary malignancy.92,93 A recent development in the identification
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS
Figure 10 Positron emission tomography (PET) with
11
153
C-metomidate (MTO) in a patient with adrenocortical cancer in the right adrenal.
of adrenocortical tissue is the 11-β hydroxylase radiotracer 11 C-metomidate, which distinguishes masses of adrenocortical origin from others. However, one of its drawbacks is that it does not distinguish between adrenocortical adenomas and carcinomas.94,95 This distinction can be done better by one of the methods already mentioned or by FDG-PET, which has a 95% overall accuracy. 11 C-metomidate PET, however, seems to be useful for staging of adrenocortical cancers as well as for posttreatment follow-up95 (see Figure 10).
reducing the amount of hormone secreting tissue. Despite operative intervention, most patients with adrenal carcinomas have a poor prognosis. The disease-free and survival times have already been mentioned in relation to the different syndromes (see previous text). Recurrent local and metastatic disease is common and reoperation should be attempted. Metastases occurs most often in the lymph nodes, liver, lung, and bone.85,96 – 99 Therefore, cytotoxic treatment is necessary for the majority of patients. Medical
Selective Venous Sampling
This method was previously one of the standard procedures for the diagnosis of adrenal lesions. Nowadays, this method has been abandoned due to improved scintigraphic and standard radiological techniques (CT, MRI, ultrasound). Staging Procedures
Staging includes seeking of metastases as well as assessing other organ functions. In a search for metastatic lesions, chest x-ray, MRI, CT scan, and PET scan are performed and an assessment of cardiovascular, renal, and hepatic function should also be carried out. The following stages are recognized: Stage I Tumor ≤5 cm, negative nodes, no local invasion, no metastases Stage II Tumor >5 cm, negative nodes, no local invasion, no metastases Stage III Tumor of any size, positive nodes, and/or local invasion, no metastases Stage IV Tumor of any size with distant metastases
Treatment Surgical
Surgery is the therapy of choice for all tumors. Increasingly laparoscopic approaches are being used with success.96 Surgery can be curative in benign disease and in some stage I –III adrenocortical cancers. Subtotal resection (debulking) of advanced adrenocortical cancers may also be helpful by
(1,1-dichloro-diphenyl-dichloroo,p’-DDD o,p -DDD ethane) also called Mitotane is an isomer of the insecticide dichloro-diphenyl-trichloroethane (DDT) and is the only drug so far known to have adrenolytic action. It alters mitochondrial function, inhibits cholesterol side-chain cleavage and 11β-hydroxylation, leading to suppression of steroid production. Plasma as well as urinary steroid levels decline. Mitotane increases the extra-adrenal metabolism of cortisol leading to a reduction in urinary 17-hydroxycorticosteriods but with increased formation of 6β-hydroxicortisol so that the apparent plasma levels of corticosteroids remain unchanged. The drug is usually given orally 2–6 g day−1 , gradually increasing to 9–10 g day−1 subject to tolerance. The maximum tolerated dose varies from 2 to 16 g day−1 . Small proportions are metabolized to inactive metabolites by both the liver and kidney. About 60% of the drug is excreted unchanged in the feces. It takes weeks of treatment for the therapeutic effects of mitotane to become apparent.100 – 102 Monitoring serum levels may be advantageous as some authors have reported that tumor responses correlate with serum levels, although there is far from unanimity in this context.101 Therapy may be required for at least 3 months before deciding whether the drug is influencing patient management.103 The higher serum levels may produce severe toxic side effects, underscoring the need for controlled studies to confirm the appropriate therapeutic levels.100,102 – 104 At higher doses, almost all patients experience side effects, which are both gastrointestinal (anorexia, diarrhea, vomiting) and/or neuromuscular (lethargy, somnolence, dizziness).
154
ENDOCRINE TUMORS
Table 6 Responses to o,p -DDD treatment in different studies.
Study 105
Van Slooten et al. Luton et al.106 Decker et al.107 Pommier et al.108 Wooten and King et al.109 Haak et al.104 Barzon et al.110 Baudin et al.103
Year
No. of patients
1984 1990 1991 1992 1992 1994 1997 2001
34 59 36 29 551 55 11 13
Responses (%) PR, 8 (23.5%) PR, 8 (13.5%) CR, 2, PR, 6 (22%) PR, 7 (24%) CR, PR (35%) CR, 8, PR 7 (27%) PR, 2 (18%) CR, 1, PR, 3 (31%)
CR, complete response; PR, partial response.
Table 7 Responses to cisplatin-based chemotherapy with or without o,p -DDD in different studies including more than 10 patients.
Study
Year
Drugs
No. of patients
van Slooten et al.119 Schlumberger et al.120 Bukowski et al.121 Berruti et al.122 Bonacci et al.123 Williamson et al.124 Khan et al.125
1983 1991 1993 1998 1998 2000 2004
CAP FDP P EDP EP EP OPEC
11 13 37 28 18 45 11
Responses (%) PR, 2 (18%) CR, 1, PR, 2 (23%) CR, 1, PR, 10 (30%) CR, 2, PR, 13 (53.5%) CR, 3, PR, 3 (33%) PR, 5 (11%) PR, 2 (18%)
CAP, cyclophosphamide, doxorubicin, and cisplatin; FDP, 5-fluorouracil, doxorubicin, and cisplatin; P, cisplatin; EDP, etoposide (VP16), doxorubicin, and cisplatin; EP, etoposide and cisplatin; OPEC, vincristine, cisplatin, teniposide, and cyclophosphamide; CR, complete response; PR, partial response.
The half-life of the parent compound ranges from 18 to 159 days. Blood levels generally become undetectable after 6 to 9 weeks following discontinuation of therapy in most patients.103 In a series of studies of inoperable adrenocortical cancers (see Table 6), 13.5–35% of patients have shown regression of both primary tumor and metastases in response to mitotane although there is no evidence that this improves survival.103 – 110 It is also interesting to note that no improvement in outcome has been achieved over this period (see Table 6). The role of mitotane as an adjuvant after surgical resection is still not known, although its use has been recommended.111,112 Moreover, multidrug resistance (MDR) mediated by MDR1 gene/P-glycoprotein 7 (Pgp) can be reversed by mitotane due to its interference with Pgp function. As high levels of Pgp have been found in adrenocortical cancer,113 this opens a possible new use for this drug in combination with other forms of cytotoxic therapy (see Table 7). All patients treated with mitotane should receive longterm glucocorticoid maintenance therapy, as there is a risk of adrenal insufficiency. Some patients may in addition need mineralocorticoid replacement. Streptozotocin Streptozotocin is a member of a group of alkylating antineoplastic agents known as alkylnitrosoureas, which is neither cell cycle phase specific nor cross-resistant with other nitrosoureas. It acts by inhibiting DNA synthesis and RNA transcription, leading to cell death by apoptosis or necrosis.114 Its plasma half-life is only 35–40 minutes, and <10% of the drug is excreted by the kidneys. Nausea and
vomiting are common side effects. Elevated liver enzymes and renal toxicity may occur with prolonged treatment especially with higher doses.115 Streptozotocin has been shown to concentrate in the adrenal cortex in mice.116 This observation led to a pilot clinical trial with the combination of streptozotocin and mitotane that has shown a beneficial effect in patients with advanced adrenocortical cancer.117 By using both drugs in combination, their individual doses can be decreased to more tolerable levels. In a recent phase II study, the adjuvant use of both streptozotocin and mitotane in 40 patients with adrenocortical cancer after curative resection was found to have an impact on the disease-free interval as well as on survival. Complete or partial responses were obtained in 36% of patients with measurable disease.118 Cisplatin-Based Chemotherapy Various other chemotherapeutic approaches have been tried in patients with adrenocortical cancer. Single-agent chemotherapy has not been effective. The preferred second-line approach in locally recurrent or metastatic adrenocortical cancer is platinumbased therapy. The responses obtained in different studies using most commonly cisplatin-based chemotherapy regimens with or without o,p -DDD are summarized in Table 7.119 – 125 Other combinations including vincristine, etoposide, doxorubicin, or vincristine, as well as cyclophosphamide regimens have also shown partial responses. Doxorubicin is said to be ineffective as a second-line chemotherapy in patients with well-differentiated or functioning tumors and in whom mitotane has been ineffective. In a recent study, the efficacy and tolerance of combination therapy of vincristine, cis-platinum, etoposide, and cyclophosphamide was evaluated in 11 patients with advanced adrenocortical cancer after failure of streptozotocin and o,p -DDD therapy. The median survival was 21 months from the onset of therapy; a partial response was achieved in only two patients.125 The side effects of these different cytotoxic chemotherapies are well known and include anemia, loss of appetite, nausea, vomiting, diarrhea, hemorrhage, infection, and hair loss. Symptomatic Treatment Drugs that inhibit adrenal steroidogenesis but have no antitumor effects may be indicated for symptomatic relief of functioning metastatic or inoperable disease. Steroid inhibition in severely Cushingoid subjects before surgical intervention (using steroid synthesis blockers such as ketoconazole, aminoglutethimide, metyrapone, etomidate, and/or o,p -DDD) may be effective. Ketoconazole is an imidazole-derivative antifungal agent that blocks 11β-hydroxylase and other enzymes in the biosynthetic pathway of corticosteroid production. Metyrapone also inhibits cortisol production by inhibiting the 11β-hydroxylase. Aminoglutethimide blocks steroidogenesis by preventing the conversion of cholesterol to pregnenolone. Adrenal insufficiency is a risk when using all these agents, and replacement steroids may be required. Radiotherapy There is no evidence to suggest that radiation therapy has any role in the management of primary adrenocortical cancer. The one major exception is that local
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS
radiotherapy may be helpful in palliative treatment of bone metastases.
ADRENAL MEDULLARY TUMORS Pheochromocytoma Pheochromocytomas are tumors of neuroectodermal origin arising from chromaffin tissues, which are widespread in their association with sympathetic ganglia during fetal life. Postnatally, most chromaffin tissue degenerates with the predominant exception of the adrenal medulla. Pheochromocytomas occur most frequently between the ages of 29 and 50 years, although no age is exempt. Congenital lesions have been recorded and, in children, the peak age incidence is between 9 and 14 years. In children, males are affected more often; the tumors are often extra-adrenal, multifocal, and associated with hereditary syndromes. The incidence in adults shows no sex difference. Pheochromocytomas have been recorded during pregnancy.126 The majority (90%) of pheochromocytomas are adrenal in origin with the right gland being more often involved. The remainder are extra-adrenal, now referred to as paragangliomas, and are found in association with the organs of Zuckerkandl and paravertebral sympathetic ganglia although tumors involving the bladder, gallbladder, thorax, heart, neck, and brain have also been recorded.127,128 Paragangliomas (formerly chemodectomas, glomus jugulare tumors) arising from the parasympathetic system also occur in the head and neck. Generally, they lack endocrine activity.129,130 Almost all pheochromocytomas are benign; occasionally (up to 20%) multiple benign pheochromocytomas/paragangliomas are encountered (e.g., one in an adrenal gland and another in the organ of Zuckerkandl). Such lesions must not be confused with malignancy. However, paragangliomas are often more likely to be malignant. It is now considered that around 20% of pheochromocytomas and paragangliomas have a familial basis.129 – 131 Such tumors are found in association with the multiple endocrine neoplasia (MEN) type 2 syndrome, the von HippelLandua (VHL) syndrome, neurofibromatosis 1 (NF 1) and germline mutations of the succinate dehydrogenase (SDH) complex. Additionally, some family members present only with pheochromocytomas. However, follow-up studies will frequently show the later evolution of manifestations of one of the above syndromes. Within any one family with the relevant syndrome, despite each member having identical mutations, the tumors develop at widely different ages. In the MEN 2 autosomal-dominant syndrome, characterized by medullary thyroid carcinoma, primary hyperparathyroidism and pheochromocytoma (type 2a) and pheochromocytoma, medullary thyroid carcinoma and multiple mucosal neuromas (type 2b), such pheochromocytomas are often bilateral. If they are unilateral on initial presentation, a contralateral lesion often develops within the following 10 years. Extra-adrenal disease and malignancy are very rare. Germline mutations of the RET proto-oncogene coding for a receptor tyrosine kinase are found. Occasional somatic mutations at the VHL locus have been noted. Less than 5% of such pheochromocytomas are malignant. Interestingly, unilateral
155
pheochromocytomas have been recorded in the autosomaldominant MEN 1 syndrome. Inactivation of the MEN 1 (menin) locus has also been noted. Longer follow-up, however, is needed to exclude development of a contralateral tumor as might be expected in a genetic syndrome. The type 2 VHL syndrome is associated with pheochromocytoma in 20–40% of patients. The tumors are bilateral in about 50% of cases and/or multifocal. Most are benign. Somatic RET changes do not occur in such VHL-related tumors.131 NF1 encodes for the protein neurofibromin, which has a GTPase-activating protein (GAP) family domain that down-regulates ras activity. In this syndrome, numerous disregulating mutations have been found. Around 3–13% of patients develop a pheochromocytoma, of which around 10% are bilateral, occasionally ectopic, but rarely (12%) malignant.131 Such lesions tend to be found at a later age than those of the MEN 2 syndromes. Families with pheochromocytomas/paragangliomas have now been detected where there are germline mutations of SDHB, SDHD and less commonly SDHC.132 – 136 They account for 70% of all familial head and neck paragangliomas and around 7–8% of ‘sporadic’ pheochromocytomas and head and neck paragangliomas. Deletions of the SDHB/D genes have also been recorded. SDHB mutations most commonly associated with pheochromocytomas are said to carry an increased tendency to malignancy. The SDHD changes in head and neck paragangliomas carry an increased incidence of multifocality. SDH mutations may induce hypoxiaresponsive genes including those of the VEGF family and account for the prominent tumor vascularity. Raised VEGF levels also occur in sporadic examples.137 The relative absence of somatic mutations in RET, VHL, and the SDH complex genes in sporadic forms of pheochromocytomas or paragangliomas has drawn attention to the need to ascertain the other mechanisms that account for the basis of the evolution of this tumor in general. Clinical Presentation
The clinical manifestations of pheochromocytomas are due to the effects of the released catecholamines and seldom due to the tumor mass per se. Common signs and symptoms include hypertension, headache, palpitation and anxiety, with weakness, visual disturbances, and abdominal and chest pain. Gastrointestinal upsets, such as diarrhea, are less frequent. All symptoms are typically paroxysmal, lasting for minutes to hours. In many cases, the hypertension is sustained but exhibits marked fluctuations with peak blood pressures occurring during symptomatic episodes. Rarely, hypotension may be the presenting sign. There should be a high degree of clinical suspicion of pheochromocytoma in patients with accelerated hypertension, unusual blood pressure lability, paroxysmal tachyarrhythmias, hypermetabolism, abnormal carbohydrate metabolism, and abnormal pressure responses to the induction of anesthesia or antihypertensive drugs.138,139 However, pheochromocytomas are still detected as incidental findings at MRI done for other reasons, at autopsy, or as a result of cardiovascular collapse at the time of an unrelated surgery.140
156
ENDOCRINE TUMORS
The majority of pheochromocytomas release norepinephrine and/or epinephrine. The production of the latter is almost always the prerogative of an adrenal gland-sited lesion as opposed to an extra-adrenal tumor.141 Pheochromocytes often contain a variety of peptides, including chromogranin A, B, and C, adrenomedullin, 3,4-dihydroxyphenylalanine (DOPA), neuropeptide Y, the enkephalins, VIP, gastrin, C-type natriuretic peptide, somatostatin, growth hormonereleasing factor, calcitonin, vasopressin, ACTH, and VEGF, and its receptors.142 – 156 The clinical relevance of many of these peptides remains to be elucidated. Neuropeptide Y may be involved in the regulation of myocardial perfusion. The presence of high circulating levels of neuropeptide Y may contribute to the cardiovascular features of the disorders. Raised plasma levels of VIP are mainly found in extraadrenal tumors and may be responsible for the diarrhea and hypokalemia in these patients. Chromogranin A is the major soluble protein stored in the vesicles of the adrenal medulla and sympathetic nerves and is secreted along with catecholamines. Elevated plasma levels can be found in patients with pheochromocytoma. The release of ACTH may result in Cushing’s syndrome.157 Pathology
The precise size and weight of a pheochromocytoma is determined by its functional activity. The more hormonally active lesions tend to be found earlier and are, therefore, smaller. While weights of 1.4 g have been noted, most tumors weigh around 100 g, and lesions of 3.5 kg have been recorded. Their size varies from being a small nodule to 14 cm in diameter.158 Most tumors are encapsulated; some remain as small intraadrenal lesions. The cut surface is pearly gray or light brown with areas of hemorrhage, necrosis, myxomatous change, and cysts in larger lesions. Microscopically, two dominant morphological patterns, the large and small alveolar types, are found, both frequently occurring in different parts of
0.1 mm
Figure 11 Benign pheochromocytoma. The large alveolar pattern is illustrated with the tumor cells forming solid trabeculae and sheets punctuated by thin-walled vascular sinusoids (H&E, original magnification:×200).
0.1 mm
Figure 12 Benign pheochromocytoma. The small alveolar pattern is shown with the tumor cells arranged in small groups separated by prominent fibrovascular trabeculae. Note the marked nuclear and cellular pleomorphism of this benign lesion (H&E, original magnification: × 200.
a single tumor. Both patterns consist of mature pheochromocytes (see Figures 11 and 12). The cell cytoplasm is granular, basophilic, and sometimes vacuolated. While the single nuclei are generally round and vesicular, nuclear and cellular pleomorphism may be marked. Giant and bizarre forms may be seen, although these features cannot be equated with malignancy. It is stated that the lesions found in the VHL syndrome are morphologically distinct from those of the MEN syndrome.159 It is imperative in all suspected lesions to verify the diagnosis of pheochromocytoma by histochemical and/or immunocytochemical techniques. Formerly, the chromaffin reaction was valuable. However, with the demonstration of their specific localization in pheochromocytes, the delineation of neurofilaments, neurone-specific enolase (NSE), tyrosine hydroxylase, and/or the peptides discussed above is a better and more meaningful approach, while also providing valuable functional data. The Epo receptor and Epo may be found in VHL lesions, whereas only Epo is present in MEN cases.159 The presence or absence of these properties, however, does not appear to help with the histologic diagnosis of malignancy. As with other endocrine tumors, a definitive diagnosis of malignancy cannot be made by morphological criteria alone. Benign pheochromocytomas can display more pleomorphism and more frequent mitoses than malignant lesions. Minimal capsular invasion and tumor cells in blood vessels are features of benign lesions. Malignant lesions have been stated to be larger, have extensive areas of necrosis, and are composed of small cells. However, such features are not absolute criteria for the diagnosis of malignancy.160 In view of the lack of accepted criteria of malignancy, it is not surprising that there are such conflicting reports regarding its incidence. Morphological criteria have suggested that 10% of all pheochromocytomas are malignant. If the proved ability to metastasize is considered as the sole criterion, only around 1% of all pheochromocytomas are malignant.158
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS
A diagnosis of malignancy may be achieved through integrated histological, functional, and genetic studies of the tumor and its products. Lesions less than 100 g in weight are more likely to be benign. Potentially valuable morphologic indices have recently been proposed to distinguish benign from malignant.161 When these indices are used together with a Ki67 score of over 5%, and in the presence of raised high telomerase reverse transcriptase (hTERT) mRNA levels, they have proved to be valuable and highly associated with malignancy but they are not infallible.162,163 The levels of neuropeptides and S-100 positive cells have been reported as being lowered in malignancy. While most tumors will produce catecholamines, metastases may be functionally active or inactive. The tumor content of dopamine/dopa and their presence together with their metabolites in increased quantities in the urine provides a further indicator of malignancy.164,165 The molecular pathogenetic analyses of pheochromocytomas, as with adrenal tumors in general, are at an early stage but interesting leads have been attained. Pheochromocytomas exhibit a significant degree of genetic instability.166 Down-regulation of several tumor suppressor genes has been recorded167 and increased copies of a number of genes, identified through the use of comparative genomic hybridization (CGH) analyses, have been noted.162 p53 alterations are found in both benign and malignant chromaffin tumors. The order in which they appear may vary. A recent excellent review provides a basis for understanding the origin and development of pheochromocytomas and paragangliomas.168 It appears that spontaneous pheochromocytomas and some examples of familial pheochromocytomas may evolve along different pathways initially. Adrenomedullary hyperplasia may be the initial step followed by several independent gene alterations that drive the lesion toward malignancy, but this is not a ubiquitous phenomenon. A considerable subset of malignant pheochromocytomas exhibit a combination of 5p, 7p and 12q gains. Paragangliomas, by comparison, show loss of 11q, the site of the SDHD mutation. The most common sites of metastases are the regional lymph nodes, liver, lung, brain, and bones. It is important to distinguish malignancy and metastases from multiple ectopic pheochromocytomas/paragangliomas. Metastases occur at sites where chromaffin tissue is not normally found. Prognosis
Generally, recurrences are apparent within 1 year and survival beyond 3 years is exceptional. Survival following removal of a malignant pheochromocytoma is variable and depends on the degree of spread noted at surgery and the success of any debulking procedure. It is generally difficult to predict the course that malignant lesions will pursue. There appear to be two subsets. One follows an aggressive course with lung, liver, and brain metastases and a median survival of around 4–5 years. The other is more indolent with metastases in lymph nodes and bones in particular. Survival in this group can be for many years. It is not possible, at this time, to predict which course a particular lesion will follow.
157
Diagnostic Investigative Tests and Staging Differential Diagnosis Pheochromocytoma may be suspected when a crisis, the physiologic consequence of abrupt catecholamine release, is precipitated by factors such as trauma, certain drugs, anesthesia, or surgery. The use of β-blockers may cause a paradoxical rise in blood pressure in some patients. Several disorders may mimic the symptoms of pheochromocytomas by inducing elevations in catecholamine levels. This includes abrupt withdrawal of medications such as clonidine. Alcohol withdrawal may produce a similar picture, as well as cerebral events such as cerebral vasculitis, subarachnoid hemorrhage, or migraine. On the other hand, the symptoms of pheochromocytomas may be mistaken for those of panic attacks, hypoglycemic episodes, or accelerated hypertension due to other etiologies. Disorders such as mastocytosis and the carcinoid syndrome, which are characterized by episodic spells and symptoms, may also mimic pheochromocytomas. However, hypertensive crises, which often occur with pheochromocytoma, are notably absent in these disorders.169 Indications for Screening Because pheochromocytomas are rare tumors, screening is recommended in the following conditions.
1. Hypertension with episodic features, suggesting pheochromocytomas (the classic triad of headaches, palpitations, and diaphoresis) 2. Refractory hypertension 3. Prominent blood pressure lability 4. Severe pressor responses during anesthesia or surgery 5. Unexplained hypotension during anesthesia, surgery, or pregnancy 6. A family history of pheochromocytomas or a familial neuroendocrine disorder such as MEN 2, VHL disease, NF, or glomus tumors 7. Incidentally discovered adrenal masses 8. Idiopathic dilated cardiomyopathy Biochemical Diagnosis The diagnosis of pheochromocytoma is made by demonstrating the existence of elevated circulating or urinary catecholamines and/or their metabolites. Screening methods include 24-hour urine collections for epinephrine and norepinephrine and/or their metabolites, metanephrine, normetanephrine, and vanillylmandelic acid (VMA). Determination of plasma catecholamines and metanephrines has recently become more popular.170 – 174 Pheochromocytomas contain, produce, and release a wide variety of catecholamines, their precursors, and metabolites. Hence, there is neither an optimal screening nor preferred diagnostic test. Measurement of plasma-free metanephrines has been recently suggested as the best screening method.171 – 174 However, catecholamines have short half-lives and are secreted episodically. A random plasma measurement may miss the peak catecholamine levels. Although a 24-hour urine collection has the advantage of integrating catecholamine secretion over a period, it is more cumbersome for patients when they are being managed as outpatients and may yield a false-negative result if the
158
ENDOCRINE TUMORS
Figure 13 Pheochromocytoma Positron emission tomography with tracer
collection is performed incompletely by the patient or in the absence of symptoms or hypertension associated with episodic catecholamine secretion. In the initial evaluation of a patient suspected of having a pheochromocytoma it is recommended to take a 24-hour urine collection for free or unmetabolized catecholamines (epinephrine and norepinephrine) and total metanephrines and creatinine, this last test to verify the adequacy of collection.174 Urinary catecholamine levels or metabolites two or three times above the upper limit of the normal levels are generally found in patients with confirmed tumors. Of the different metabolites that can be detected in a 24-hour urine collection, the metanephrines are the most sensitive and specific. Plasma measurements of metanephrines, particularly the levels of free metanephrines, are also considered to be a highly sensitive method for biochemical diagnosis, especially for hereditary pheochromocytomas (sensitivity 97%, specificity 96%).47,48 Care is needed though as false-positive metanephrine levels may occur in older patients.173 Plasma for catecholamine assays is drawn after overnight fasting with the patient in a supine position and with a heparin lock inserted 20 to 30 minutes before withdrawing blood. Tricyclic antidepressants, clonidine, clofibrate, and α/β-blockers that may interfere with urinary catecholamines and metabolite measurements should be discontinued, preferably 2 weeks before collection. Blood pressure should be controlled with agents such as calcium channel blockers that do not interfere with the assays. Stress associated with serious illnesses such as myocardial infarction, cerebrovascular accidents, or congestive heart failure, may cause elevated catecholamine levels. In renal insufficiency, plasma and urinary levels may be falsely elevated.173 Chromogranin A, a soluble protein, stored and secreted by chromaffin tissue is an important marker for pheochromocytomas and is elevated in plasma in more than 80% of patients. Chronic renal failure can yield a falsely elevated chromogranin A level.175,176 Imaging Techniques Once biochemical results indicate a pheochromocytoma, imaging techniques such as MRI, CT, metaiodobenzylguanidine (MIBG) or octreotide scintigraphy,
11
C-hydroxyephedrine (HED). Notice the involvement of left adrenal (arrow).
and most recently PET-scanning are employed for tumor localization.177 – 179 CT and MRI are usually the initial localizing procedures for adrenal and extra-adrenal tumors; their sensitivity varies between 75 and 100% depending on location and whether the tumor is primary, recurrent, or metastatic. However, the specificity of both methods is low. Radiotracer techniques have been developed to image the adrenal medulla and its lesions based on specific catecholamine transport and storage mechanisms possessed by pheochromocytomas. These include scintigraphy with I131 and I123-MIBG.180 – 183 New tracers for PET have the advantages of these same biological properties and include 11 C-hydroxyephedrine (HED), C11-epinephrine and 6-18F-fluorodopamine and 18F-DOPA179,184 – 186 (see Figure 13). Somatostatin receptor (octreotide) scintigraphy has been developed because around 25% of benign pheochromocytomas possess somatostatin type 2 receptors.187,188 In a large retrospective study of patients who underwent surgery for mostly benign pheochromocytomas, the overall detection rate for tumors larger than 1 cm in diameter was 90% for 123I-MIBG but only 25% for octreotide scintigraphy.180 MIBG-scanning requires preparation including the withdrawal of drugs before examination. In addition, imaging requires a delay of 48 hours after MIBG injection, because of the slow clearance of radioactivity from normal organs. Moreover, malignant pheochromocytomas may show a lower uptake of the tracer because of a loss of expression of catecholamine transporters. However, in patients with metastases, octreotide scintigraphy detected lesions in 70–80% of patients, including I123-MIBG negative cases. For extra-adrenal tumors (paragangliomas), octreotide scintigraphy was reported to be positive in almost 100% of cases.188 The sensitivity for FDG-PET has been reported to be 70% for solitary malignant pheochromocytomas.189,190 6-18F-fluorodopamine PET was reported to yield a 100% sensitivity in recent studies by Pacak and coworkers.186 In another recent study, HED scans gave a sensitivity and specificity of 91.6% and 100% respectively.185 These two new imaging PET-tracers have the potential to improve
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS
the limited sensitivity of MIBG and octreotide scintigraphy. However, their value in malignant pheochromocytomas or in paragangliomas remains to be ascertained. One problem with these tracers is their physiological uptake in other organs, such as salivary glands, heart, liver, pancreas, spleen, and kidney that may appear as false-positive uptakes (see Figure 4). Staging Procedures The following staging procedures are recommended:
1. An assessment of hypertension, cardiovascular, and renal functions 2. Localization of the neoplastic lesions 3. A search for extra-adrenal sites of involvement For the latter two staging procedures the following localizing procedures are suggested: 1. Primarily, CT scan or MRI 2. 131I/123I-MIBG scan or PET with 6-18F-fluorodopamine or 11 C-hydroxyephedrine Treatment Surgical Care While surgical excision of pheochromocytomas is the treatment of choice, it has a morbidity as high as 40% and a mortality of 2–4%, although outcomes have improved with preoperative treatment such as α-receptor blockade and volume expansion.191 The recent introduction of laparoscopic techniques is also opening new horizons for surgical approaches with favorable outcomes.96 Phenoxybenzamine, a noncompetitive α-blocker, has traditionally been used for preoperative preparation and is titrated to reduce blood pressure to normal levels or orthostasis, or both. The starting oral dose is usually 10 mg/day, but most patients require 80 to 100 mg daily for control. When α-blockade is established, β-blockade may be started if the patient is tachycardic or has arrhythmias. Without prior α-blockade, β-blockade alone can lead to unopposed α-receptor stimulation and further elevation of the blood pressure. Because phenoxybenzamine blocks catecholamines binding to their receptors, it minimizes the risk of a hypertensive crisis during intubation, induction of anesthesia, exploration and tumor manipulation. A noncompetitive α-blocker would theoretically be preferred. However, complete α-blockade can mask the dramatic fall in blood pressure seen after tumor resection. Phenoxybenzamine may also lead to postoperative hypotension because of its prolonged halflife of 24 hours. Calcium channel blockers have generated increased attention as these agents may improve intraoperative systemic vascular resistance by blunting catecholamine-mediated arterial vasoconstriction during tumor manipulation. They also have fewer side effects. The oral formulation of labetalol, with an α/β-blocking ratio of 1 : 3, may not be ideal for surgery preparation because the α-blockade is weaker than that of phenoxybenzamine.192 In current practice, tradition and experience still guide the length of preoperative therapy with volume expansion and α/β-blockade, which typically involves treatment for at least 10 to 14 days. There are no
159
reliable features that predict a good surgical outcome. Each patient must be evaluated individually when selecting antihypertensive medicines. The surgical approach depends on the clinical situation. In patients with familial pheochromocytomas, a trans-abdominal incision allows adequate visualization and bilateral adrenalectomy if required. The flank approach offers better exposure and reduces blood loss for the patient with a solitary tumor. To an increasing extent, laparoscopic adrenalectomy is gaining favor.96,193 Intraoperative hypotension is managed initially with volume expansion and then with intravenous pressor agents if necessary. Postoperative hypoglycemia, which may be due to reactive hyperinsulinemia, should be anticipated and warrants routine screening of glucose monitoring in the early postoperative hours. After surgery, catecholamine levels usually return to normal in approximately 2 weeks. If hypertension persists despite normal catecholamine levels, essential hypertension, or hypertension secondary to renal damage may be the cause. Medical Management While some reports indicate a malignancy rate for pheochromocytomas of between 3% to 13%, if the proved ability to metastasize is taken as the sole criterion, only around 1% of all pheochromocytomas are malignant).178 Long-term follow-up of one study in particular has yielded an interesting conundrum. Pheochromocytoma patients, as a group, seem to have an increased risk of death from other cancers, and/or cardiovascular disorders. In one recent series, six patients with hereditary disease developed other neuroectodermal tumors.194 For malignant tumors, surgery is the only truly hope of cure. Debulking operations also have a role to play. The response to chemotherapeutic agents such as vincristine, cyclophosphamide, and dacarbazine used in combination,195,196 has been disappointing, but they may be tried together with antihypertensive treatment. An alternative might be to use 131I-MIBG as a radiotherapeutic with or without chemotherapy. Since a large number of these tumors possess somatostatin receptors, another similar approach might involve 90 Yttrium-DOTA-octreotide or 177 Lutetium-DTPA-DOTA-octreotate. Although there are anecdotal reports of other agents, such as cisplatin, the taxanes, and gemcitabine, having some activity their use must be viewed as investigational in the absence of wellcharacterized phase II clinical trial data. External beam radiation has been used for bone metastases palliation, but has limited utility in the management of the primary tumor. Pregnancy
Management of pheochromocytomas in pregnancy is especially challenging.197 The mortality rate for mother and fetus is reported to be approximately 50%. Pheochromocytomas may also be easily misdiagnosed as preclampsia, especially later in pregnancy. The diagnosis is typically made by evaluation of the urinary collection of catecholamines and metanephrines. MRI is the preferred imaging modality because there is no ionizing radiation. MIBG is contraindicated in pregnancy. Surgery is typically performed
160
ENDOCRINE TUMORS
before 20 to 24 weeks of gestation. Thereafter, medical treatment is attempted depending on the maternal status and caesarean section is planned followed by tumor resection. Calcium channel blockers are to be preferred to control blood pressure.
OTHER ANOMALIES INCLUDING SO-CALLED ‘INCIDENTALOMAS’ While this chapter has focused upon malignant functioning tumors of the adrenal gland, it is important to realize that other tumors, benign and malignant, but not unique to the adrenal gland may be detected from time to time, more so since the introduction of CT and MRI scanning, the so-called incidentalomas.1 – 4 The most common lesion will be the so-called nonfunctioning nodule, referred to by some authors inappropriately as adenomas.5 Such nodules are common, increasing with age, and prone to be found in hypertensive and diabetic populations. Their etiology and genesis have been discussed previously. Nonetheless, with the passage of time and our increasing understanding of subclinical forms of hypercorticalism and its earlier detection as a result of CT scanning followed by tests of adrenal function, such lesions now merit reappraisal. This applies particularly with respect to cortisol and aldosterone synthesis as some will prove to be true adenomas rather than simple nodules.2 Other lesions of the adrenals include myelolipomas, cysts, pseudocysts, adenomatoid tumors, tuberculosis, histoplasmosis, amyloid deposits, and hemorrhage.198 – 200 Neoplasms of the adrenal stroma, and adipose, neural, and vascular tissues will occur from time to time. Primary adrenal lymphomas occur but are extremely rare. Clinical acumen and histological examination will serve to characterize such lesions whose treatment is the same as when they occur at more common sites.201 These various pathologies as incidental findings are said to occur in approximately 0.5 to 2% of all CT studies.
11C-MTO
CT
11C-HED Figure 14 Computerized tomography (CT) and positron emission tomography (PET) with 11 C-metomidate (MTO) and 11 C-hydroxyephedrine (HED) in a patient with a 2 cm incidentaloma in the right adrenal. The metomidate PET is negative, whereas the HED PET is positive indicating a pheochromocytoma.
However, with the increasing age of the population, the figure is nearer to 10%. Around 10–15% of the lesions will be bilateral, especially if they are nodules. At autopsy, 2–3% of all subjects will have such unanticipated adrenal pathology. Very rarely primary tumors with functional properties are found,200 although it has been stated that over one-third of pheochromocytomas are detected as incidentalomas202 (see Figure 14). It should be remembered that the adrenal is a common site for metastases particularly from lung and breast carcinomas. Not infrequently such metastases are bilateral and involve adrenal nodules.
CONCLUSIONS Adrenocortical and medullary tumors are uncommon lesions. The majority are benign, cured by surgery. If overt metastases are not detected, the histological diagnosis of malignancy may be rendered difficult. Functional biochemical parameters and molecular and genetic analysis can be helpful. Assay of tumor products may also provide useful indices to monitor therapy. For the localized tumor, surgery remains the mainstay of curative treatment. Too few malignancies have been seen at any one center to enable useful therapeutic recommendations to be achieved with the aim of improving outcome for patients with recurrent and metastatic disease. A multicenter international study group may be one of the best ways to proceed to overcome this problem and to derive regimes of value for patients in the future.
REFERENCES 1. Geelhoed GW, Drury EM. Management of the adrenal ‘incidentaloma’. Surgery 1982; 92: 866 – 74. 2. Barzon L, et al. Prevalence and natural history of adrenal incidentalomas. Eur J Endocrinol 2003; 149: 273 – 85. 3. Grossrubatscher E, et al. The natural history of incidentally discovered adrenocortical adenomas; a retrospective evaluation. J Endocrinol Invest 2001; 24: 846 – 55. 4. Mansmann G, et al. The clinically inapparent adrenal mass: update in diagnosis and management. Endocr Regul 2004; 25: 309 – 40. 5. Neville AM, O’Hare MJ. The Human Adrenal Cortex. Berlin, Germany: Springer-Verlag, 1982. 6. Lewinsky BS, et al. The clinical and pathological features of ‘nonhormonal adrenocortical tumors. Cancer 1974; 33: 778 – 90. 7. Neville AM, Symington T. The pathology of the adrenal gland in Cushing’s syndrome. J Pathol Bacteriol 1967; 93: 19 – 35. 8. Margulies PL, et al. Remission of Cushing’s syndrome during pregnancy. Int J Gynaecol Obstet 1983; 21: 77 – 83. 9. Mayer SK, et al. Childhood adrenocortical tumors: case series and reevaluation of prognosis-a 24 year experience. J Pediatr Surg 1997; 144: 61 – 5. 10. Del Gaudio AD, Del Gaudio GA. Virilizing adrenocortical tumors in adult women. Cancer 1993; 72: 1997 – 2003. 11. Muensterer OJ, et al. Testosterone-producing adrenocortical neoplasm in a 6-year old boy. Eur J Pediatr Surg 2001; 11: 354 – 7. 12. De Asis DN, Samaan NA. Feminizing adrenocortical carcinoma with Cushing’s syndrome and pseudohyperparathyroidism. Arch Intern Med 1978; 138: 301 – 3. 13. Halmi KA, Lascari AD. Conversion of virilization to feminization in a young girl with adrenal cortical carcinoma. Cancer 1971; 27: 931 – 5. 14. Malunowicz EM, et al. Heterogeneity of urinary steroid profiles in children with adrenocortical tumors. Horm Res 1995; 44: 182 – 8.
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS 15. Conn JW. Presidential address. Part I: painting the background. Part II: primary aldosteronism: a new clinical syndrome. J Clin Lab Med 1995; 45: 3 – 6. 16. Brown JJ, et al. Aldosterone: physiological and pathophysiological variations in men. J Clin Endocrinol Metab 1972; 1: 397 – 449. 17. Kaplan NM. The current epidemic of primary aldosteronism: causes and consequences. Review 2004; 22: 863 – 9. 18. Neville AM, Symington T. Pathology of primary aldosteronism. Cancer 1966; 12: 1854 – 68. 19. Lu L, et al. Nur-related factor 1 and nerve growth factor-induced clone B in human adrenal cortex and its disorders. J Clin Endocrinol Metab 2004; 89: 4113 – 8. 20. Kurtulmus N, et al. Co-secretion of aldosterone and cortisol by an adrenocortical carcinoma. Horm Res 2004; 62: 67 – 70. 21. Sweeney AT, et al. A malignant aldosteronoma. Endocr Pract 2002; 8: 373 – 7. 22. Enberg U, et al. Postoperative differentiation between unilateral adrenal adenoma and bilateral adrenal hyperplasia in primary aldosteronism by mRNA expression of the gene CYP11B2. Eur J Endocrinol 2004; 151: 73 – 85. 23. O’Hare MJ, Monoghan P, Neville AM. The pathology of adrenocortical neoplasia; a correlated structural and functional approach to the diagnosis of malignant disease. Hum Pathol 1979; 10: 137 – 54. 24. Tanaka K, et al. Oncocytic adrenocortical carcinoma. J Urol 2004; 64: 376 – 7. 25. Song SY, et al. Oncocytic adrenocortical carcinomas; a pathological and immunohistochemical study of four cases in comparison with conventional adrenocortical carcinomas. Pathol Int 2004; 54: 603 – 10. 26. Bisceglia M, et al. Adrenocortical oncocytic tumors; report of 10 cases and review of the literature. Int J Surg Pathol 2004; 12: 231 – 43. 27. Koch CA, Pacak K, Chrousos GP. The molecular pathogenesis of hereditary and sporadic adrenocortical and adrenomedullary tumors. J Clin Endocrinol Metab 2002; 87: 5367 – 84. 28. Darracott Vaghan E Jr. Diseases of the adrenal gland. Med Clin North Am 2004; 88: 443 – 66. 29. Sidhu S, et al. Clinical and molecular aspects of adrenocortical tumourigenesis. ANZ J Surg 2003; 73: 727 – 38. 30. Kjellman M, et al. Genetic aberrations in adrenocortical tumors detected using comparative genomic hybridization correlate with tumor size and malignancy. Cancer Res 1996; 56: 4219 – 23. 31. Sidhu S, et al. Adrenocortical cancer: recent clinical and molecular advances. Curr Opin Oncol 2003; 16: 13 – 8. 32. Fottner CH, et al. Role of insulin-like growth factor system in adrenocortical growth control and carcinogenesis. Horm Metab Res 2004; 36: 397 – 405. 33. Sasano H, et al. Transforming growth factor α, epidermal growth factor, and epidermal growth factor receptor expression in normal and diseased human adrenal cortex by immunohistochemistry and in situ hybridization. Mod Pathol 1994; 7: 741 – 6. 34. Saeger W, Fassnacht M, Reincke M. Allolio. Expression of HER2/neu receptor protein in adrenal tumors. Pathol Res Pract 2002; 198: 445 – 8. 35. Schteingart DE, et al. Overexpression of CXC chemokines by an adrenocortical carcinoma: a novel clinical syndrome. J Clin Endocrinol Metab 2001; 86: 3968 – 74. 36. Blanco M, et al. Cellular localization of orexin receptors in human adrenal gland, adrenocortical adenomas and pheochromocytomas. Regul Pept 2002; 104: 161 – 5. 37. Pilon C, et al. Inactivation of the p16 tumor suppressor gene in adrenocortical tumors. J Clin Endocrinol Metab 1999; 84: 2776 – 80. 38. Lin SR, et al. Mutations of K-ras oncogene in human adrenal tumors in Taiwan. Br J Cancer 1998; 77: 1060 – 5. 39. Lin SR, et al. A significant decrease of the transcriptional activity of p53 mutants deriving from human functional adrenal tumors. DNA Cell Biol 1996; 10: 793 – 803. 40. Latronico AC, et al. An inherited mutation outside the highly conserved DNA-binding domain of the p53 tumor suppressor protein in children and adults with sporadic adrenocortcaltumors. J Clin Endocrinol Metab 2001; 86: 4970 – 3.
161
41. Pinto EM, et al. Founder effect for the highly prevalent R337H mutation of tumor suppressor p53 in Brazilian patients with adrenocortical tumors. Arq Bras Endocrinol Metaol 2004; 48: 647 – 50. 42. Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 1984; 8: 163 – 9. 43. Weiss LM, Medeiros LJ, Vickery AL. Jr. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989; 13: 202 – 6. 44. Wagner M, et al. Adrenocortical tumors. I Prognastic evaluation of a series of 17 cases using the Weiss criteria. Ann Pathol 1993; 13: 306 – 11. 45. Aubert S, et al. Weiss system revisited: a clinicopathalogic and immunohistochemical study of 49 adrenocortical tumors. Int J Surg Pathol 2002; 26: 161 – 209. 46. Lucon AM, et al. Adrenocortical tumors: results of treatment and study of Weiss’s score as a prognostic factor. Rev Hosp Clin Fac Med 2002; 57: 251 – 6. 47. Pohlink C, et al. Does tumor heterogeneity limit the use of Weiss criteria in the evaluation of adrenocortical tumors? J Endocrinol Invest 2004; 27: 565 – 9. 48. Terzolo M, et al. Immunohistochemical assessment of Ki-67 in the differential diagnosis of adrenocortical tumors. Urologia 2001; 57: 176 – 82. 49. Arola J, et al. p53 and Ki67 in adrenocortical tumors. Endocr Res 2000; 26: 861 – 5. 50. Stojadinovic A, et al. Adrenocortical adenoma and carcinoma: histopathological and molecular comparative analysis. Mod Pathol 2003; 16: 742 – 51. 51. Sredni ST, et al. P53 as a prognostic factor in adrenocortical tumors of adults and children. Braz J Med Biol Res 2003; 36: 23 – 7. 52. Tissier F, et al. Cyclin E correlates with malignancy and adverse prognosis in adrenocortical tumors. Eur J Endocrinol 2004; 150: 809 – 17. 53. Gicquel C, et al. Molecular markers and long-term recurrences in a large cohort of patients with sporadic adrenocortical tumors. Cancer Res 2001; 61: 6762 – 7. 54. de Fraipont F, et al. Gene expression profiling of human adrenocortical tumors using complementary deoxyribonucleic acid microarrays identifies several candidate genes as markers of malignancy. J Clin Endocrinol Metab 2005; 90: 1819 – 29. 55. Erickson LA, et al. Pathologic features and expression of insulin-like growth factor-2 in adrenocortical neoplasms. Endocr Pathol 2001; 12: 429 – 35. 56. Kirschner LS. Signalling pathways in adrenocortical cancer. Ann N Y Acad Sci 2002; 968: 222 – 39. 57. Simard J, et al. Molecular biology of the 3β-hydroxysteroid dehydrogenase 5− 4 isomerase gene family. Endocr Rev 2005; 26: 525 – 82. 58. Ing-Cherng Guo, Meng-Chun Hu, Bon-chu Chung. Transcriptional regulation of CYP11A1. J Biomed Sci 2003; 10: 593 – 8. 59. Lu L, et al. Nur-related factor 1 and nerve growth factor-induced clone B in human adrenal cortex and its disorders. J Clin Endocrinol Metab 2004; 89: 4113 – 8. 60. Bassett MH, White PC, Rainey WE. A role for the NGFI-B family in adrenal and adrenocortical disease. Endocr Res 2004; 30: 567 – 74. 61. Beuschlien F, et al. ACTH-receptor expression, regulation and role in adrenocortical tumor formation. Eur J Endocrinol 2001; 144: 199 – 206. 62. Reincke M, et al. Deletion of the adrenocorticotropin receptor gene in human adrenocortical tumors: implications for tumorigenesis. J Clin Endocrinol Metab 1997; 82: 3054 – 8. 63. Latronico AC, et al. No evidence for oncogenic mutations in the adrenocorticotropin receptor gene in human adrenocortical neoplasms. J Clin Endocrinol Metab 1995; 80: 875 – 7. 64. Takahashi K, Totsune K, Murakami O. Adrenocortical peptides: autocrine or paracrine regulators for the steroid hormone secretion or the cell proliferation? Exp Clin Endocrinol Diabetes 2002; 110: 373 – 80. 65. Lacroix A, N’Diaye N, Tremblay J. et al. Ectopic and abnormal hormone receptors in adrenal Cushing’s syndrome. Endocr Rev 2001; 22: 75 – 110.
162
ENDOCRINE TUMORS
66. Schubert B, et al. Angiotensin II type 1 receptor and ACTH receptor expression in human adrenocortical neoplasms. Clin Endocrinollin Endocrinol 2001; 54: 627 – 32. 67. Beauregard C, Dickstein G, Lacroix A. Classic and recent etiologies of Cushing’s syndrome. Diagnosis and therapy. Treat Endocrinol 2002; 1: 79 – 94. 68. Lacroix A, et al. Cushing’s syndrome variants secondary to aberrant hormone receptors. Trends Endocrinol Metab 2004; 15: 375 – 82. 69. Sonino N, Boscaro M, Fallo F. Pharmacologic management of Cushing syndrome: new targets for therapy. Treat Endocrinol 2005; 4: 87 – 94. 70. Bertherat J, et al. In vivo and in vitro screening for illegitimate receptors in adrenocorticotropin-independent macronodular adrenal hyperplasia causing Cushing’s syndrome: identification of two cases of gonadotropin/gastric inhibitory polypeptide-dependent hypercortisoloism. J Clin Endocrinol Metab 2005; 90: 1302 – 10. 71. Bourdeau I, et al. Gene array analysis of macronodular adrenal hyperplasia confirms clinical heterogeneity and identifies several candidate genes as molecular mediators. Oncogene 2004; 23: 1575 – 85. 72. Shibata H, et al. Regulation of differential COUP-TF-coregulator interactions in adrenal cortical steroidogenesis. J Steroid Biochem 2003; 85: 449 – 56. 73. Shibata H, et al. Expression profiles of COUP-TF, DAX-1, and SF1 in the human adrenal gland and adrenocortical tumors: possible implication in steroidogenesis. Mol Genet Metab 2001; 74: 206. 74. Barbosa AS, et al. Assessment of the role of transcript for GATA4 as a marker of unfavourable outcome in human adrenocortical neoplasms. Endoc Dis 2004; 4: 1 – 11. 75. Shibata H, et al. COUP-TFI expression in human adrenocortical adenomas: possible role in steroidogenesis. J Clin Endocrinol Metab 1998; 83: 4520 – 3. 76. Figueiredo Bc, et al. Amplification of the steroidogenic factor 1 gene in childhood adrenocortical tumors. J Clin Endocrinol Metab 2005; 90: 615 – 9. 77. Mazzocchi G, et al. Cortisol-secreting adrenal adenomas express 11 beta-hydroxysteroid dehydrogenase type-2 gene yet possess low 11 beta-HSD2 activity. J Investig Med 2001; 49: 191 – 4. 78. Mune T, et al. Role of ll β-hydroxysteroid dehydrogenase type 2 expression in determining the phenotype of adrenal adenomas. J Clin Endocrinol Metab 2003; 88: 864 – 70. 79. Dall’Asta C, et al. Coexistence of 21-hydroxylase and 11 betahydroxylase deficiency in adrenal incidentalomas and in subclinical Cushing’s syndrome. Horm Res 2002; 57: 192 – 6. 80. Jimenez P, et al. GATA-6 is expressed in the human adrenal and regulates transcription of genes required for adrenal androgen biosynthesis. Endocrinology 2003; 144: 4285 – 8. 81. Kiiveri S, et al. Transcription factors GATA-4 and GATA-6 in human adrenocortical tumors. Endocr Res 2004; 30: 919 – 23. 82. Saner KJ, et al. Steroid sulfotransferase 2A1 gene transcription is regulated by steroidogenic factor 1 and GATA-6 in the human adrenal. Mol Endocrinol 2005; 19: 184 – 97. 83. Havelock JC, et al. The NGFI-B family of transcription factors regulates expression of beta-hydroxysteroid dehydrogenase type 2 in the human ovary. Mol Hum Reprod 2005; 11: 79 – 85. 84. Orth DN. Cushing’s syndrome. N Engl J Med 1995; 332: 791 – 803. 85. Montwill J, Igoe D, McKenna TJ. The overnight dexamethasone test is the procedure of choice in screening for Cushing’s syndrome. Steroids 1994; 59: 296 – 8. 86. Chrousos GP, et al. The corticotropin-releasing factor stimulation test. An aid in the evaluation of patients with Cushing’s syndrome. N Engl J Med 1984; 310: 622 – 6. 87. Nieman LK, et al. A simplified morning ovine corticotropinreleasing hormone stimulation test for the differential diagnosis of adrenocorticotropin-dependent Cushing’s syndrome. J Clin Endocrinol Metab 1993; 77: 1308 – 12. 88. Melby JC. Diagnosis of hyperaldosteronism. Endocrinol Metab Clin North Am 1991; 20: 247 – 55. 89. Grondal S, et al. Steroid profile in urine: a useful tool in the diagnosis and follow up of adrenocortical carcinoma. Acta Endocrinol (Copenh) 1990; 122: 656 – 63.
90. Mayo-Smith WW, et al. State-of-the-art adrenal imaging. Radiographics 2001; 21: 995 – 1012. 91. Dunnick NR, Korobkin M. Imaging of adrenal incidentalomas: current status. AJR Am J Roentgenol 2002; 179: 559 – 68. 92. Kaltsas G, et al. Recent advances in radiological and radionuclide imaging and therapy of neuroendocrine tumours. Eur J Endocrinol 2004; 151: 15 – 27. 93. Maurea S, et al. Imaging of adrenal tumors using comparison of benign and malignant lesions. AJR Am J Roentgenol 1999; 173: 25 – 9. 94. Bergstrom M, et al. PET imaging of adrenal cortical tumors with the 11beta-hydroxylase tracer 11C-metomidate. J Nucl Med 2000; 41: 275 – 82. 95. Eriksson B, et al. The role of PET in localization of neuroendocrine and adrenocortical tumors. Ann N Y Acad Sci 2002; 970: 159 – 69. 96. Cobb WS, et al. Laparoscopic adrenalectomy for malignancy. AM Surg 2005; 189: 405 – 11. 97. Schulick RD, Brennan MF. Long-term survival after complete resection and repeat resection in patients with adrenocortical carcinoma. Ann Surg Oncol 1999; 6: 719 – 26. 98. Allolio B, et al. Management of adrenocortical carcinoma. Clin Endocrinol (Oxf) 2004; 60: 273 – 87. 99. Bulow B, Ahren B. Adrenal incidentaloma – experience of a standardized diagnostic programme in the Swedish prospective study. J Intern Med 2002; 252: 239 – 46. 100. Ahlman H, et al. Cytotoxic treatment of adrenocortical carcinoma. World J Surg 2001; 25: 927 – 33. 101. Khorram-Manesh A, et al. Adrenocortical carcinoma: surgery and mitotane for treatment and steroid profiles for follow-up. World J Surg 1998; 22: 605 – 11. 102. Bargenstal D, et al. Chemotherapy of adrenocortical cancer with o,p’DDD. Ann Intern Med 1960; 53: 672 – 82. 103. Baudin E, et al. Impact of monitoring plasma 1,1-dichlorodiphenildichloroethane (o,p’DDD) levels on the treatment of patients with adrenocortical carcinoma. Cancer 2001; 92: 1385 – 92. 104. Haak HR, et al. Optimal treatment of adrenocortical carcinoma with mitotane: results in a consecutive series of 96 patients. Br J Cancer 1994; 69: 947 – 51. 105. van Slooten H, et al. The treatment of adrenocortical carcinoma with o,p’-DDD: prognostic implications of serum level monitoring. Eur J Cancer Clin Oncol 1984; 20: 47 – 53. 106. Luton JP, et al. Clinical features of adrenocortical carcinoma, prognostic factors, and the effect of mitotane therapy. N Engl J Med 1990; 322: 1195 – 201. 107. Decker RA, et al. Eastern Cooperative Oncology Group study 1879: mitotane and adriamycin in patients with advanced adrenocortical carcinoma. Surgery 1991; 110: 1006 – 13. 108. Pommier RF, Brennan MF. An eleven-year experience with adrenocortical carcinoma. Surgery 1992; 112: 963 – 70; discussion 970 – 1. 109. Wooten MD, King DK. Adrenal cortical carcinoma. Epidemiology and treatment with mitotane and a review of the literature. Cancer 1993; 72: 3145 – 55. 110. Barzon L, et al. Adrenocortical carcinoma: experience in 45 patients. Oncology 1997; 54: 490 – 6. 111. Vassilopoulou-Sellin R, et al. Impact of adjuvant mitotane on the clinical course of patients with adrenocortical cancer. Cancer 1993; 71: 3119 – 23. 112. Barzon L, et al. Comment – is there a role for low doses of mitotane (o,p’-DDD) as adjuvant therapy in adrenocortical carcinoma? J Clin Endocrinol Metab 1999; 84: 1488 – 9. 113. Bates SE, et al. Mitotane enhances cytotoxicity of chemotherapy in cell lines expressing a multidrug resistance gene (mdr-1/Pglycoprotein) which is also expressed by adrenocortical carcinomas. J Clin Endocrinol Metab 1991; 73: 18 – 29. 114. Bolzan AD, Bianchi MS. Genotoxicity of streptozotocin. Mutat Res 2002; 512: 121 – 34. 115. Weiss RB. Streptozotocin: a review of its pharmacology, efficacy, and toxicity. Cancer Treat Rep 1982; 66: 427 – 38. 116. Tjalve H, Wilander E, Johansson EB. Distribution of labelled streptozotocin in mice: uptake and retention in pancreatic islets. J Endocrinol 1976; 69: 455 – 6.
NEOPLASTIC DISORDERS OF THE ADRENAL GLANDS 117. Eriksson B, et al. Treatment of hormone-producing adrenocortical cancer with o,p’DDD and streptozotocin. Cancer 1987; 59: 1398 – 403. 118. Khan TS, et al. Streptozotocin and o,p’DDD in the treatment of adrenocortical cancer patients: long-term survival in it’s adjuvant use. Ann Oncol 2000; 11: 1281 – 7. 119. van Slooten H, van Oosterom AT. CAP (cyclophosphamide, doxorubicin, and cisplatin) regimen in adrenal cortical carcinoma. Cancer Treat Rep 1983; 67: 377 – 9. 120. Schlumberger M, et al. 5-fluorouracil, doxorubicin, and cisplatin as treatment for adrenal cortical carcinoma. Cancer 1991; 67: 2997 – 3000. 121. Bukowski RM, et al. Phase II trial of mitotane and cisplatin in patients with adrenal carcinoma: a Southwest Oncology Group study. J Clin Oncol 1993; 11: 161 – 5. 122. Berruti A et al., Italian Group for the Study of Adrenal Cancer. Mitotane associated with etoposide, doxorubicin, and cisplatin in the treatment of advanced adrenocortical carcinoma. Cancer 1998; 83: 2194 – 200. 123. Bonacci R, et al. Cytotoxic therapy with etoposide and cisplatin in advanced adrenocortical carcinoma. Br J Cancer 1998; 78: 546 – 9. 124. Williamson SK, et al. Phase II evaluation of cisplatin and etoposide followed by mitotane at disease progression in patients with locally advanced or metastatic adrenocortical carcinoma: a Southwest Oncology Group Study. Cancer 2000; 88: 1159 – 65. 125. Khan TS, et al. Vincristine, cisplatin, teniposide, and cyclophosphamide combination in the treatment of recurrent or metastatic adrenocortical cancer. Med Oncol 2004; 21: 167 – 77. 126. Hume DM. Pheochromocytoma in the adult and in the child. Am J Surg 1960; 99: 458 – 96. 127. Nomura A, et al. Images in cardiovascular medicine. Malignant cardiac pheochromocytoma with bone metastases. Circulation 1993; 97: 1993 – 4. 128. Mehra S, Chung-Park M. Gallbladder paraganglioma; a case report with review of the literature. Arch Pathol Lab Med 2005; 129: 523 – 6. 129. Kaltsas GA, Papadogias D, Grossman AB. The clinical presentation (symptoms and signs) of sporadic and familial chromaffin cell tumours (phaeochromocytomas and paragangliomas). Front Horm Res 2004; 31: 61 – 75. 130. Dluhy RG. Screening for genetic causes of hypertension. Curr Hypertens Rep 2002; 4: 439 – 44. 131. Gimm O. The genetic basis of pheochromocytoma. In Lenhert H (ed) Pheochromocytoma. Karger Publishers, 2004, Vol. 31: 45 – 60. 132. McDonnell CM, et al. K40E: a novel succinate dehydrogenase (SDH)B mutation causing familial phaeochromocytoma and paraganglioma. Clin Endocrinol 2004; 61: 510 – 4. 133. Allibhai Z, et al. Malignant pheochromocytoma associated with germline mutation of the SDHB gene. J Urol 2004; 172: 1409 – 10. 134. Favier J, et al. Hereditary paraganglioma/pheochromocytoma and inherited succinate dehydrogenase deficiency. Horm Res 2005; 63: 171 – 9. 135. Neumann HPH, et al. Distinct clinical features of paraganglioma syndromes associated with SDHB and SDHD gene mutations. JAMA 2004; 292: 943 – 51. 136. Astuti D, et al. Genetic analysis of mitochondrial complex II subunits SDHD, SDHB and SDHC in paraganglioma and phaeochromocytoma susceptibility. Clin Endocrinol 2003; 59: 728 – 33. 137. Chandel NS, et al. Mitochonondrial reactive oxygen species trigger hypoxia-induced transcription. Proc Natl Acad Sci USA 1998; 95: 11715 – 20. 138. Melicow MM. One hundred cases of pheochromocytoma (107 tumors) at the Columbia Presbyterian medical centre, 1926 – 1976. A clinopathological analysis. Cancer 1977; 40: 1987 – 2004. 139. Manger EM, Gifford RW Jr. Hypertension secondary to pheochromocytoma. Bull NY Acad Med 1982; 58: 138 – 58. 140. St John Sutton MG, Sheps SG, Lie JT. Prevalence of clinically unsuspected pheochromocytoma. Mayo Clin Proc 1981; 56: 354 – 60. 141. Cryer PE. Pheochromocytoma. Clin Endocrinol Metab 1985; 14: 203 – 20. 142. Mazzocchi G, et al. Human pheochromocytomas express orexin receptor type 2 gene and display an in vitro secretory response to orexins A and B. J Clin Endocrinol Metab 2001; 86: 4818 – 21.
163
143. Korner M, Waser B, Reubi JC. High expression of neuropeptide Y receptors in tumors of the human adrenal gland and extra-adrenal paraganglia. Clin Cancer Res 2004; 10: 8426 – 33. 144. Takahashi K, et al. Expression of prolactin-releasing peptide and its receptor in the human adrenal glands and tumor tissues of adrenocortical tumors, pheochromocytomas and neuroblastomas. Peptides 2002; 23: 1135 – 40. 145. Portela-Gomes GM, et al. Expression of chromogranins A, B, and C (secretogranin II) in human adrenal medulla and in benign and malignant pheochromocytomas. An immunohistochemical study with region specific antibodies. APMIS 2004; 10: 663 – 73. 146. Grossman AB, Kaltas GA. Adrenal medulla and pathology. In Besser GM, Thorner MD (eds) Comprehensive Clinical Endocrinology. Philadelphia, Pennsylvania: Elsevier Science, 2002: 223 – 237. 147. Wilson BS, Lloyd RV. Detection of chromagranin in neuroendocrine cells with a monoclonal antibody. Am J Pathol 1984; 115: 458 – 68. 148. Adrian TE, et al. Neuropeptide Y in phaeochromocytomas and ganglioneuroblastomas. Lancet 1983; 2: 540 – 2. 149. Yoshimasa T, et al. Plasma methionine-enkephalin and leucineenkephalin in normal subjects and patients with pheochromocytoma. J Clin Endocrinol Metab 1983; 57: 706 – 12. 150. Long RG, et al. Clinicopathological study of pancreatic and ganglioneuroblastoma tumors secreting vasoactive intestinal polypeptide (vipomas). Br Med J 1981; 282: 1767 – 71. 151. Bostwick DG, Bensch KG. Gastrin releasing peptide in human neuroendocrine tumors. J Pathol 1985; 147: 237 – 44. 152. Totsune K, et al. Immunoreactive C-type natriuretic peptide in human adrenal glands and adrenal tumors. Peptides 1994; 15: 287 – 90. 153. Berelowitz M, et al. Somatostatin-like immunoactivity is present in a human pheochromocytoma. J Clin Endocrinol Metab 1983; 56: 134 – 8. 154. Sano T, et al. Production and secretion of immunoreactive growth hormone-releasing by pheochromocytoma. Cancer 1986; 57: 1788 – 93. 155. Weinstein RS, Ide LF. Immunoreactive calcitonin in pheochromocytomas. Proc Soc Exp Biol Med 1980; 165: 215 – 7. 156. Ang VTY, Jenkins JS. Neurohypophyseal hormones in the adrenal medualla. J Clin Endocrinol Metab 1984; 58: 688 – 91. 157. Spark RF, et al. ACTH secretion from a functioning pheochromocytoma. N Engl J Med 1979; 301: 416 – 8. 158. Neville AM. The adrenal medulla. In Symington T (ed) The Functional Pathology of the Human Adrenal Gland. Edinburgh, Scotland: E&S Livingstone, 1969: 243 – 271. 159. Vogel TWA, et al. Differential expression of erythropoietin and its receptor in von Hippel-Lindau-associated and MEN type 2-associated pheochromocytomas. J Clin Endocrinol Metab 2005; 90: 1 – 21. 160. Weiss LM. Adrenal gland. In Weidner N, et al. (eds) Modern Surgical Pathology. Philadelphia, Pennsylvania: Saunders, 2003: 1755 – 1780. 161. Thompson LDR. Pheochromocytoma of the adrenal gland: scaled score PASS to separate benign from malignant neoplasms. Am J Surg Pathol 2002; 26: 551 – 66. 162. August C, et al. CGH and CD 44/MIB-1 immunohistochemistry are helpful to distinguish metastasized from nonmetastasized sporadic pheochromocytomas. Mod Pathol 2004; 17: 1119 – 28. 163. Isobe K, et al. Expression of the human telomerase reverse transcriptase in pheochromocytoma and neuroblastoma tissues. Endoc J 2004; 51: 47 – 52. 164. Sjoerdsma A, et al. Pheochromocytoma: current concepts of diagnosis and treatment. Ann Intern Med 1966; 65: 1302 – 26. 165. Anton AH, et al. Dihydroxyphenylalanine secretion in a malignant pheochromocytoma. Am J Med 1976; 42: 469 – 75. 166. Gunawan B, et al. Cytogenetic characterization of 5 pheochromocytomas. Cancer Genet Cytogenet 2004; 154: 163 – 6. 167. Ohta S, et al. Downregulation of metastasis suppressor genes in malignant pheochromocytoma. Int J Cancer 2005; 114: 139 – 43. 168. Dannenberg H, et al. Molecular genetic alterations in adrenal and extra-adrenal pheochromocytomas and paragangliomas. Endocr Pathol 2003; 14: 329 – 50. 169. Werbel SS, Ober KP. Pheochromocytoma. Update on diagnosis, localization, and management. Med Clin North Am 1995; 79: 131 – 53.
164
ENDOCRINE TUMORS
170. Pacak K, et al. Recent advances in genetics, diagnosis, localization, and treatment of pheochromocytoma. Ann Intern Med 2001; 134: 315 – 29. 171. Lenders JW, et al. Plasma metanephrines in the diagnosis of pheochromocytoma. Ann Intern Med 1995; 123: 101 – 9. 172. Eisenhofer G. Editorial: biochemical diagnosis of pheochromocytoma – is it time to switch to plasma-free metanephrines? J Clin Endocrinol Metab 2003; 88: 550 – 2. 173. Eisenhofer G, et al. Biochemical diagnosis of pheochromocytoma: how to distinguish true- from false-positive test results. J Clin Endocrinol Metab 2003; 88: 2656 – 66. 174. Eisenhofer G, Lenders JW, Pacak K. Biochemical diagnosis of pheochromocytoma. Front Horm Res 2004; 31: 76 – 106. 175. Hsiao RJ, et al. Chromogranin A storage and secretion: sensitivity and specificity for the diagnosis of pheochromocytoma. Medicine 1991; 70: 33 – 45. 176. Stridsberg M, Husebye ES. Chromogranin A and chromogranin B are sensitive circulating markers for phaeochromocytoma. Eur J Endocrinol 1997; 136: 67 – 73. 177. Pacak K, Eisenhofer G, Ilias I. Diagnostic imaging of pheochromocytoma. Front Horm Res 2004; 31: 107 – 20. 178. Scott BA, Gatenby RA. Imaging advances in the diagnosis of endocrine neoplasia. Curr Opin Oncol 1998; 10: 37 – 42. 179. LI S, Beheshti M. The radionuclide molecular imaging and therapy of neuroendocrine tumors. Curr Cancer Drug Targets 2005; 5: 139 – 48. 180. Lauriero F, et al. I-131 MIBG scintigraphy of neuroectodermal tumors. Comparison between I-131 MIBG and In-111 DTPAoctreotide. Clin Nucl Med 1995; 20: 243 – 9. 181. Maurea S, et al. Iodine-131-metaiodobenzylguanidine scintigraphy in preoperative and postoperative evaluation of paragangliomas: comparison with CT and MRI. J Nucl Med 1993; 34: 173 – 9. 182. Shapiro B, Gross MD, Shulkin B. Radioisotope diagnosis and therapy of malignant pheochromocytoma. Trends Endocrinol Metab 2001; 12: 469 – 75. 183. van der Harst E, et al. [(123)I]metaiodobenzylguanidine and [(111)In]octreotide uptake in benign and malignant pheochromocytomas. J Clin Endocrinol Metab 2001; 86: 685 – 93. 184. Ilias I, et al. Superiority of 6-[18F]-fluorodopamine positron emission tomography versus [131I]-metaiodobenzylguanidine scintigraphy in the localization of metastatic pheochromocytoma. J Clin Endocrinol Metab 2003; 88: 4083 – 7. 185. Trampal C, et al. Pheochromocytomas: detection with 11C hydroxyephedrine PET. Radiology 2004; 230: 423 – 8.
186. Pacak K, et al. 6-[18F]fluorodopamine positron emission tomographic (PET) scanning for diagnostic localization of pheochromocytoma. Hypertension 2001; 38: 6 – 8. 187. Kwekkeboom D, dJ M, Krenning E. Somatostatin receptor scintigraphy. In Sandler M, et al. (eds) Diagnosis in Nuclear Medicine, 4th ed. Philadelphia, Pennsylvania: Lippincott & Wilkins, 2003: 747 – 754. 188. Kwekkeboom DJ, et al. Octreotide scintigraphy for the detection of paragangliomas. J Nucl Med 1993; 34: 873 – 8. 189. Eisenhofer G, et al. Malignant pheochromocytoma: current status and initiatives for future progress. Endocr Relat Cancer 2004; 11: 423 – 36. 190. Pacak K, Eisenhofer G, Goldstein DS. Functional imaging of endocrine tumors: role of positron emission tomography. Endocr Rev 2004; 25: 568 – 80. 191. Ulchaker JC, et al. Successful outcomes in pheochromocytoma surgery in the modern era. J Urol 1999; 161: 764 – 7. 192. Russell WJ, et al. Labetalol in the preoperative management of phaeochromocytoma. Anaesth Intensive Care 1982; 10: 160 – 3. 193. Winfield HN, et al. Laparoscopic adrenalectomy: the preferred choice? A comparison to open adrenalectomy. J Urol 1998; 160: 325 – 9. 194. Khorram-Manesh A, et al. Long-term outcome of a large series of patients surgically treated for pheocromacytoma. J Intern Med 2005; 258: 55 – 66. 195. Harper MA, et al. Phaeochromocytoma in pregnancy. Five cases and a review of the literature. Br J Obstet Gynaecol 1989; 96: 594 – 606. 196. Averbuch SD, et al. Malignant pheochromocytoma: effective treatment with a combination of cyclophosphamide, vincristine, and dacarbazine. Ann Intern Med 1988; 109: 267 – 73. 197. Loh KC, et al. The treatment of malignant pheochromocytoma with iodine-131 metaiodobenzylguanidine (131I-MIBG): a comprehensive review of 116 reported patients. J Endocrinol Invest 1997; 20: 648 – 58. 198. Matsuda T, et al. Case of combined adrenal cortical adenoma and myelolipoma. Pathol Int 2004; 54: 725 – 9. 199. Wagnerova H, et al. Adrenal myelolipoma. 6 cases and a review of the literature. Neoplasma 2004; 51: 300 – 5. 200. Varkarakis IM, et al. Adenomatoid of the adrenal gland. J Urol 2005; 65: 175.e26 – 175.e8. 201. Favia G, et al. Management of incidentally discovered adrenal masses and risk of malignancy. Surgery 2000; 128: 918 – 24. 202. Zarco-Gonzlez JA, Herrera MF. Adrenal Incidentaloma. Scand J Surg 2004; 93: 298 – 30.
Section 3 : Endocrine Tumors
11
Uncommon Cancers of the Thyroid Mark Bloomston and Manisha H. Shah
INTRODUCTION Thyroid cancers are relatively uncommon with 25 690 new cases and 1490 deaths estimated in the United States for 2005.1 This accounts for 1.8% of all new cancer diagnoses and 0.3% of cancer-related deaths. Cancer of the thyroid is the fastest growing cancer among women in the United States and is, by far, the most common endocrine malignancy, representing 92% of all endocrine cancers. However, thyroid cancer accounts for only 62% of endocrine cancer-related deaths documenting the rather indolent nature of the disease relative to other cancers. Thyroid cancers are traditionally classified as well differentiated, medullary, or anaplastic, accounting for 90, 5, and 1–2% of thyroid cancers, respectively. Well-differentiated tumors are further divided into papillary (80–85%), follicular (10–15%), and the less common H¨urthle cell carcinoma (3–5%). These tumors tend to arise from follicular cells, which line the colloid follicles within the thyroid and are involved in the concentration of iodine and the production of thyroid hormone. These same cells likely also give rise to anaplastic thyroid carcinomas (ATC). Medullary thyroid carcinoma (MTC) develops from the parafollicular cells, sometimes called calcitonin-producing C cells, which are found in the upper and middle thirds of the thyroid gland and are embryologically derived from the neural crest. Immune cells are commonly found within the substance of the thyroid and give rise to lymphomas, which represent 1–3% of thyroid malignancies. Most rare tumors account for less than 1% of thyroid malignancies and include sarcoma, insular carcinoma, primary squamous carcinoma, malignant teratoma, mucoepidermoid carcinoma, and hemangioendothelioma. This chapter will focus on the diagnosis and management of medullary thyroid cancer (both sporadic and familial), H¨urthle cell carcinoma, anaplastic cancer, lymphoma, and these other rare tumors.
MEDULLARY THYROID CARCINOMA First described in 1959 as a solid thyroid neoplasm without follicular histology and a marked propensity for lymph node metastasis, the origin of MTC would not be known until the
late 1960s when the parafollicular C cell was described.2,3 The correlation between C cells and MTC was further strengthened in the 1970s when calcitonin was described as, what would become, one of the most sensitive and specific tumor markers in oncology.4 Investigators quickly began developing an understanding of the familial associations of MTC, which has led to the definition of genetic changes in the RET (rearranged during transfection) proto-oncogene important for both the heritable and sporadic forms of the disease.5 MTC is now one of the best-characterized solid tumor malignancies, given our understanding of its genetic, molecular, pathological, biochemical, and clinical features.
Hereditary Features of MTC MTC accounts for approximately 3–5% of all thyroid cancers and is associated with familial syndromes in 25–30% of cases. It is a principal component of multiple endocrine neoplasia (MEN) type 2A (MTC, pheochromocytoma, hyperparathyroidism) and 2B (MTC, pheochromocytoma, mucosal ganglioneuromatosis, marfanoid body habitus) as well as being part of its own syndrome of familial medullary thyroid cancer (FMTC). These syndromes show an autosomaldominant pattern of inheritance with MEN 2 variants being caused by mutations in the RET proto-oncogene.5 MEN 2A is caused by mutations in extracellular cysteine residues such as mutation at codon 634 while MEN 2B results from a methionine to threonine mutation at codon 918 in the tyrosine kinase catalytic domain. Familial MTC is caused by the same mutation as MEN 2A as well as, less commonly, mutations in the intracellular portion of the protein.6 These distinct, welldefined mutations coupled with effective treatment options make genetic screening for the associated familial syndromes mandatory in patients diagnosed with MTC. Referral of such patients to a cancer geneticist for evaluation and genetic counseling is essential. Also, the prominence of pheochromocytoma in MEN 2 emphasizes the importance of ruling out MEN 2 by genetic screening or pheochromocytoma by plasma metanephrines or urine catecholamines in patients presenting with MTC. MTC, in the setting of MEN 2, is characterized by 100% penetrance, with invasive carcinoma present by the first year
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
166
ENDOCRINE TUMORS
of life in MEN 2B and lymph node metastases reported by age 2.7 MTC in patients with MEN 2A, albeit less aggressive than in MEN 2B, is often present by age 2 with lymph node metastases by age 5.8 FMTC, the least aggressive of the three familial MTC syndromes, is characterized by MTC presenting in the second or third decade of life. These differences in presentation affect the recommendations regarding timing of prophylactic thyroidectomy in children with the associated mutations. Early thyroidectomy in patients with biochemical evidence of MTC by provocative (with calcium and pentagastrin stimulation) calcitonin screening of at-risk family members or without biochemical evidence of MTC in patients with germline mutations can be curative or preventive in all patients with MEN 2.9,10 The current recommendations for the management of the thyroid, based on the 2001 consensus guidelines for MEN 2, emphasize the use of RET mutation testing rather than calcitonin testing in at-risk individuals.11 This is based on several points: (i) early thyroidectomy can significantly alter the course of MTC in MEN 2; (ii) thyroidectomy is relatively well tolerated and safe; (iii) waiting for biochemical evidence of MTC may leave patients with residual disease following thyroidectomy; and (iv) genetic testing is a more sensitive screening method than stimulated calcitonin testing. On this basis, patients with the MEN 2B genotype (codon 883, 918, and 922 mutations) are at highest risk and it is recommended that they should undergo thyroidectomy and central node dissection by 6 months of age, and preferably within the first month of life. Mutations at codons 634, 620, 618, and 611 are also at relatively high risk and it is recommended that patients with these mutations should undergo thyroidectomy with or without central node dissection by 5 years of age. The recommendations for patients at lowest risk (codon 609, 768, 790, 791, 804, and 891 mutations) are less rigid, with some advocating thyroidectomy by age five, by age 10, or on the basis of provocative calcitonin testing.
Clinical Presentation and Diagnosis Most MTCs present with sporadic disease as an asymptomatic thyroid mass. Patients with heavy tumor burden and very high calcitonin levels may present with secretory diarrhea as their primary symptom. At the time of presentation, up to 75% of patients will have lymph node metastases with distant metastases being common. Thus, MTC is typically a systemic disease with distant metastasis to the lungs, liver, and/or bones. Serum calcitonin is a sensitive and specific tumor marker in MTC that has diagnostic and prognostic value. In addition, serum carcinoembryonic antigen (CEA) has been used to assess the degree of dedifferentiation of the tumors in these patients. Since MTC derives from C cells and not follicular cells, they do not concentrate iodine, thus rendering radioactive iodine scan ineffective for workup.12 Conventional computed tomography (CT) or magnetic resonance imaging (MRI) scans, and nuclear imaging techniques such as 131 I-metaiodobenzylguanidine, somatostatin receptor scintigraphy, and radiolabeled CEA antibody or anticalcitonin antibody have been utilized to survey for metastatic disease, each with suboptimal sensitivity.13 – 16 All patients with apparent sporadic MTC should undergo genetic
testing, as up to 6% of such patients have been found to carry the germline RET mutation.11 The diagnosis of MTC can be confirmed by fine needle aspiration (FNA). Histologically, MTC is a type of neuroendocrine carcinoma demonstrating several patterns: glandular, solid, spindle cell, oncocytic, clear cell, papillary, small cell, and giant cell. The nuclei of MTC cells are round with stippled “salt-and-pepper” chromatin, similar to other neuroendocrine tumors found elsewhere in the body (see Figure 1). Features associated with poor outcome include necrosis, squamous pattern, presence of oxyphil cells in the tumor and absence of cells with intermediate cytoplasm, and less than 50% calcitonin reactivity.17
Treatment As systemic chemotherapy and radiation are ineffective against MTC, surgical resection is the mainstay of treatment.18 Total thyroidectomy is recommended for familial and sporadic cases, given the multifocal nature of the MTC in 90 and 20% of patients, respectively. Total thyroidectomy also enhances the ability to utilize calcitonin levels to assess for adequacy of resection and monitor for recurrence. Since MTC has a propensity for early lymph node metastasis, thyroidectomy should be accompanied by central node dissection, clearing all tissue from the hyoid down to the innominate vessels inferiorly and the internal jugular veins laterally. Thorough removal of all tissue in these central compartments by an experienced surgeon reduces local recurrence and improves overall survival.19 Clinically suspicious nodes in the lateral neck should be sampled and, if positive, lateral neck dissection undertaken. Routine removal of nodes in levels II through V bilaterally or ipsilateral modified radical neck dissection should also be considered in larger palpable thyroid tumors (i.e., >2 cm) since occult metastases can occur in 75% of ipsilateral and 47% of contralateral jugular nodes.20,21 Controversy exists regarding the ideal management of the parathyroid glands at the time of
Figure 1 Medullary thyroid carcinoma. Note finely stippled “salt-andpepper” chromatin (arrows).
UNCOMMON CANCERS OF THE THYROID
167
prophylactic or curative thyroidectomy. Some experts advocate total parathyroidectomy with reimplantation to ensure complete removal of all thyroid tissue and all central nodal tissue. Also, this may prevent future hyperparathyroidism in patients with MEN 2A.6 Others believe that this is unnecessary to achieve adequate thyroid resection and that the risk of postoperative hyperparathyroidism, even in the face of MEN 2A, is small.22
etoposide, the taxanes, and gemcitabine. This has led investigators to seek novel therapeutic strategies targeting genetic and molecular defects in MTC. Tyrosine kinase inhibitors targeting RET in MTC are now in the early phases of development, and such clinical trials are one of the best treatment options available for patients with advanced MTC.
Management of Persistent or Recurrent Calcitonin Elevation
H¨urthle cell carcinomas are oxyphilic tumors that arise from follicular cells. They do not arise from the parafollicular C cells attributed to the 19th-century pathologist after whom these tumors were mistakenly named by Ewing in 1919.30 This uncommon variant of follicular carcinoma is referred to as oxyphil tumor, oncocytoma, mitochondrioma, or Askanazy cell tumor. The natural history and optimal treatment of H¨urthle cell tumors are not completely agreed upon. Their histologic features and thyroglobulin production suggest that they arise from follicular cells; however, they do not appear to be variants of follicular carcinoma.31
¨ HURTHLE CELL CARCINOMA
While thyroidectomy is often curative in children with MEN 2A and FMTC diagnosed by genetic testing, more than 50% of patients presenting with palpable disease will fail to achieve a “biochemical cure” following resection of all known gross disease.23 The course of these patients is rather indolent, with only 35% progressing to demonstrable disease as shown by van Heerden et al. in 1990.23 In this study of 31 patients, 11 patients recurred during the 12-year followup period. These patients all went on to resection and none had normalization of their calcitonin levels. However, 5and 10-year survival rates for all patients were 90 and 86%, respectively. On the basis of these findings, patients with persistently elevated calcitonin levels without obvious disease should be followed with interval CT scanning of the neck, chest, and abdomen, and surgery reserved for those patients with appearance of measurable disease. However, if a patient has not had adequate clearance of all nodal tissue at risk as described above at the time of thyroidectomy, central neck dissection with or without lateral neck dissection should be considered. If adequate neck dissection has been completed, then blind neck reexploration is discouraged.6 Distant metastases are present at the time of diagnosis in 12% of patients with sporadic MTC, compared to 20% in MEN 2B. Only 3% of patients with MEN 2A and 2% with FMTC will present with synchronous metastases.24 The liver is a common site of distant disease, but metastatic deposits are often small and therefore difficult to detect with radiographic imaging techniques. This has led some clinicians to propose staging laparoscopy in patients being considered for clearance of persistent or recurrent neck disease with occult liver metastases identified in 20%.25 Resection of recurrent MTC offers the best chance of longterm survival and symptom palliation with median survival of over 8 years reported.26
H¨urthle cell carcinoma presents in the fifth to seventh decade of life with a 3 : 1 female preponderance. The typical presentation is that of a slow-growing thyroid nodule with lymphadenopathy, vocal cord paralysis, or distant metastasis being uncommon. Approximately only 10% of metastases from H¨urthle cell carcinomas concentrate iodine, compared with 75% of metastases from follicular carcinoma. Diagnosis of H¨urthle cell neoplasm is confirmed by FNA though the determination of invasive carcinoma cannot be made by cytological examination alone.32 H¨urthle cells show characteristic abundant granular acidophilic cytoplasm filled with mitochondria and nuclei that are usually vesicular and uniform (see Figure 2). The determination of invasive malignancy is based on the demonstration of capsular or angioinvasion that can only be seen in the resected specimen.33 Clearly, local invasion, lymph node spread, and distant metastasis also classify H¨urthle cell neoplasms as malignant. Grossly, these tumors have a distinct tan-brown color and are typically well encapsulated. Approximately 20% of H¨urthle cell tumors will turn out to be cancers. Thyroid scanning typically demonstrates a “cold” nodule but is not reliable in distinguishing between benign and malignant tumors.
Nonoperative Treatment
Treatment
Radioactive iodine treatment is ineffective in MTC since thyroid C cells do not concentrate iodine. Also, the results of external beam radiation have been disappointing. Studies administering doses greater than 5000 rads to the neck result in great difficulty in reoperating in the radiation field, with high local recurrence rates.27 Similarly, various chemotherapeutic regimens consisting of doxorubicin, dacarbazine, streptozocin, and 5-fluorouracil administered as single agents or in combination consistently report short-lived response rates of 15–30%.28,29 There is no clear information regarding the utility of the more modern agents, such as cisplatin,
The debate about the optimal resection for H¨urthle cell carcinoma (total thyroidectomy vs lobectomy) follows the discussion of the optimal management of differentiated thyroid cancers. Although lobectomy may be adequate for smaller (i.e., <1 cm) tumors, total thyroidectomy offers several advantages: (i) although local recurrence rates are low following lobectomy, they are fatal in over one-third of patients; (ii) multifocal disease is found in up to 35%; (iii) removal of all thyroid tissue facilitates radioactive iodine uptake in persistent or recurrent disease for surveillance and ablative purposes (though these tumors tend not to
Clinical Presentation and Diagnosis
168
ENDOCRINE TUMORS
RAF kinase (BRAF) gene in papillary thyroid cancer (PTC)43 has advanced our understanding of progression of differentiated thyroid tumors to ATC. The frequency of BRAF mutation in PTC is 45% (28–69%), while in ATC it is 24% (0–100%).44 However, the BRAF mutation has been consistently present in both differentiated and undifferentiated components of ATC, suggesting a possible role of BRAF in dedifferentiation. Furthermore, frequent mutations in p53 (83%),45 β-catenin (61%),46 and RAS (52%)47 have been described in ATC. While activating mutations in RAS and BRAF oncogenes are implicated in differentiated thyroid carcinoma, inactivating mutations in p53 tumor suppressor gene are exclusively present in poorly differentiated thyroid tumors. Finally, upregulation of minichromosome maintenance (MCM) proteins has recently been reported in 65% of ATC.48
Clinical Presentation and Diagnosis Figure 2 H¨urthle cell carcinoma.
concentrate iodine); and (iv) thyroglobulin assays are more sensitive for recurrent disease in the absence of normal thyroid tissue.34,35 Any remaining normal thyroid tissue should be ablated with radioactive iodine postoperatively. Surveillance consists of thyroglobulin measurement in the hypothyroid state. When thyroglobulin levels are still detectable after total thyroidectomy and ablation or if they begin to rise, recurrence should be sought. Radioiodine scanning is a reasonable early test to detect the few H¨urthle cell carcinomas that take up iodine.36 Recurrences occur in 23–29% of patients after a median of 4 years from resection.35 Most recurrences are in the neck, with the lung being the most common site of distant metastasis. Surgical resection is the mainstay of treatment for recurrent disease as chemotherapy and external beam radiation are of little benefit. Median survival after recurrence is 34 months.37
Unlike well-differentiated thyroid cancers, ATC tends to be a disease of the elderly, presenting in the seventh decade of life (two decades later than differentiated cancers). Like other thyroid cancers, there is a female preponderance of disease, though less drastic (1.5–2 : 1). Patients almost uniformly present with a rapidly enlarging mass with evidence of local invasion (i.e., dysphagia, hoarseness, stridor) being common. The primary is typically large (8–9 cm) with nearly half of patients presenting with distant metastases.38 Common sites of metastasis are lung, liver, and bone. ATC are divided into three histologic subtypes: large cell, spindle cell, and squamoid. Each subtype is named for cell patterns to which they are similar with squamoid resembling nonkeratinizing squamous cell carcinoma, spindle cell being “sarcoma-like,” and large cell demonstrating more pleomorphism than the other patterns. All subtypes are characterized by high mitotic activity, foci of necrosis, and both intra- and extrathyroidal extensive invasion.
ANAPLASTIC THYROID CARCINOMA
Treatment
The marked aggressive nature of anaplastic thyroid carcinoma clearly distinguishes it from its well-differentiated counterparts where long-term survival is expected. With a median survival of 4 to 5 months, anaplastic carcinoma is one of the most rapidly growing and deadly cancers.38 It would appear that the incidence of anaplastic carcinoma is falling. This decline may be related to previous overestimates of the incidence of disease due to misclassification of lymphomas, medullary carcinomas, and “insular” variants of follicular carcinoma.39 Some reports do suggest that the true incidence of ATC is indeed declining, likely related to the decrease in endemic goiter associated with dietary iodine supplementation.40 ATC commonly coexists with differentiated thyroid cancer leading many investigators to surmise that the former may represent dedifferentiation of the more indolent welldifferentiated neoplasms.41,42 Several genetic alterations have been implicated in the pathogenesis of ATC. The recent breakthrough discovery of activating mutations in the B-type
Given the rarity of ATC, clinical trials reported to date are limited by their retrospective nature, small sample size, limitations of single arm trial design and interpretation, lack of uniformity of therapy within specific trials, and decadeslong duration of study. In general, a multimodality regimen with surgery, radiation, and chemotherapy is recommended for this aggressive cancer. Most patients present with extensive local disease, which ultimately leads to their demise due to airway compromise. For this reason, local disease control, whether palliative or curative, is of utmost importance. When feasible, total thyroidectomy with or without neck dissection should be sought as the mainstay of treatment. Survival in resected patients is significantly longer in patients amenable to resection, though still very poor.38 Debulking operations to control local disease, combined with adjuvant hyperfractionated external beam radiation, have shown significant improvements in survival (7–8 months with clearing of neck disease vs 1 month with residual neck disease) even when distant disease was present.49 Adjuvant radiation in
UNCOMMON CANCERS OF THE THYROID
hyperfractionated doses in combination with low-dose doxorubicin is recommended for all patients undergoing either curative or palliative resection for anaplastic carcinoma.50,51 Since 30–50% of patients have unresectable local disease at presentation, neoadjuvant approaches have been applied with a resection rate of 70% and local disease control of 48%.52 Nonetheless, survival beyond 1 year is rare. Chemotherapy, although widely employed, has a limited role, with relatively low response rates from doxorubicin, the fluoropyrimidines, and cisplatin. The role of novel compounds, such as the taxanes and gemcitabine, has not yet been defined, but most cytotoxic responses are of short duration. External beam radiation as primary therapy for symptom palliation has had some success but not without toxicity, mostly related to esophageal dysfunction.53 Radioactive iodine plays no role in the treatment of recurrent or metastatic disease in anaplastic carcinoma. Because of the radio- and chemoresistant nature of this cancer, effective systemic therapies are desperately needed. Targeted therapies are currently being tested in clinical trials and such therapy may be the best treatment option for patients with advanced ATC.
INSULAR CARCINOMA First described in 1907 by Langhans and named in 1984 by Carcangiu,54 insular carcinoma represents a follicular cell-derived poorly differentiated variant of thyroid cancer. Though generally considered as having a better prognosis than anaplastic carcinoma, prognosis is still poor. With only a few series reported, insular carcinomas appear to represent less than 5% of thyroid cancers.55
Clinical Presentation and Diagnosis Insular carcinoma typically presents as a rapidly enlarging mass in patients over the age of 50 with a female predilection; however, it has also been reported in the very young.56 Local invasion as evidenced by hoarseness, dyspnea, and dysphagia is common. Cervical lymph node involvement at the time of presentation is common with distant metastasis typically occurring in the lungs and bone. Definitive diagnosis is made histologically with demonstration of solid clusters of tumor cells with small follicles within the clusters. Foci of necrosis, vascular invasion, and frequent mitotic figures are commonplace along with extrathyroidal extension of disease.57 Definitive cytologic characteristics are lacking making FNA ineffective for diagnosis.
169
PRIMARY THYROID LYMPHOMA Primary thyroid lymphoma accounts for 1% of all lymphomas and are almost all non-Hodgkin’s lymphomas.59 Since the thyroid does not inherently contain lymphoid tissue, primary thyroid lymphomas may arise in the context of acquired lymphoid aggregation, such as in autoimmune thyroid disease (e.g., Hashimoto’s thyroiditis).60 The majority (70–90%) are intermediate grade diffuse large B cell lymphomas or, less commonly, mucosa-associated lymphoid tissue (MALT) lymphomas or MALTomas.
Clinical Presentation and Diagnosis Primary thyroid lymphomas tend to present in the seventh decade of life and have at least a 3 : 1 female preponderance. These tumors typically present as a slowly growing thyroid mass with invasive symptoms of hoarseness, stridor, or dysphagia being less common. “B type” symptoms of fever, night sweats, and weight loss are only seen in about 20% of patients.59 The key to diagnosing and planning therapy for primary thyroid lymphoma, like other lymphomas, rests on accurate histologic subtyping. This generally requires core or open surgical biopsy to obtain enough tissue for immunophenotypic and cytogenetic analysis, though FNA is sometimes sufficient. Next, accurate staging is accomplished by either CT or MRI. Though MRI has been shown to be more accurate than CT in evaluating the extent of local disease in the neck, advancements in CT technology makes both imaging modalities equally effective in preoperative planning.61 Radioactive iodine and technetium uptake scans are generally not helpful due to variable uptake patterns.
Treatment The role of surgery in the management of primary thyroid lymphoma outside the context of obtaining adequate tissue for diagnosis is debatable. Many believe that diffuse large B cell lymphomas should be considered a systemic disease with no role for thyroidectomy. These patients have been managed with chemoradiation with 5-year survivals of 70% reported using doxorubicin-based regimens (also see Chapter on uncommon lymphoproliferative disorders).62 The more indolent MALT lymphomas when localized to the thyroid (stage IE) respond well to total thyroidectomy with 5-year survivals of up to 90% when complete resection is possible.63
SARCOMA Treatment Total thyroidectomy has been the mainstay of therapy in the only three large reports.54,55 Neck dissection is added to thyroidectomy given the high incidence of lymph node metastasis. Radioactive iodine has been utilized with variable response rates.55 External beam radiation is largely ineffective, as is chemotherapy. Even with early aggressive surgical therapy with or without radioiodine treatment, the prognosis in these patients is poor with 10-year survival in fewer than 50%.55,58
Often confused with spindle cell variant of anaplastic carcinoma, sarcoma of the thyroid includes liposarcoma, angiosarcoma, leiomyosarcoma, and dendritic cell types.64,65 The management of these entities is similar to other sarcomas and consists of aggressive surgical resection including total thyroidectomy with removal of any involved lymph nodes or adjacent structures. Patients harboring tumors with pathologic features suggestive of advanced disease, including extracapsular spread, high mitotic rate, abundant nuclear pleomorphism, or unresectable disease, warrant postoperative
170
ENDOCRINE TUMORS
radiotherapy with adjuvant chemotherapy recommended for more aggressive subtypes (e.g., dendritic cell sarcoma).64 The choice and sequencing of chemotherapy regimens is beyond the scope of this chapter, but essentially reflects standard patterns of care for locally advanced soft tissue sarcomas at other sites.
SQUAMOUS CELL CARCINOMA Primary squamous cell carcinoma (SCCA) of the thyroid is extremely uncommon, accounting for less than 1% of all thyroid malignancies. Several theories exist as to the origin of this rare cancer. Some have suggested that SCCA arises from remnants of the thyroglossal duct with cystic formation within the field of cancer cells being seen.66 Chronic inflammation of the thyroid such as that seen in Hashimoto’s disease may lead to squamous metaplasia and, ultimately, SCCA.67 Finally, SCCA may arise as a direct transition from papillary or follicular carcinoma.68
Clinical Presentation and Diagnosis Patients with primary SCCA of the thyroid present in the fifth or sixth decade of life with a rapidly enlarging thyroid mass. There is a female preponderance (2–3 : 1) with many patients having a long-standing history of goiter with sudden enlargement. Evidence of local invasion as well as lymph node metastasis are common at the time of diagnosis. Distant metastasis is uncommon. The diagnosis of primary SCCA of the thyroid requires exclusion of an alternate source of SCCA that has involved the thyroid by hematogenous or contiguous spread. These may arise from the aerodigestive tract most commonly, although spread from breast, gastrointestinal, and renal sources should be considered. Workup, therefore, consists of FNA of the thyroid mass to confirm the diagnosis of SCCA along with upper endoscopy and CT of the neck, chest, and abdomen. Histologic findings of primary thyroid SCCA are like those of other primaries and not unique to the thyroid, emphasizing the importance of thorough workup.
Treatment Radical resection, when possible, including total thyroidectomy with resection of adjacent involved structures and neck dissection yields the best chance for local control and long-term survival.69 This is often not possible because of extensive esophageal and/or tracheal involvement. External beam radiation may be beneficial as adjuvant therapy when complete resection has been achieved but is otherwise not helpful.68 These characteristics render SCCA of the thyroid highly lethal similar to anaplastic carcinoma with survival rates generally less than 1 year.
MALIGNANT TERATOMA Extragonadal germ cell tumors are very uncommon in adults and are always malignant. They even less commonly arise in the thyroid with the mediastinum and retroperitoneum being more common sites.70 Early descriptions reported poor
survival, although more recent reports suggest improved outcomes with the addition of chemotherapy.71 The experience of the pathologist is of critical importance, and biopsies should be referred for expert opinion from a specialty center. It is possible to confuse a benign teratoma (which only requires complete surgical resection) with a more malignant variant (which usually mandates cytotoxic chemotherapy), and thus this distinction is essential to optimal management. The median age at presentation is 28 years with a thyroid mass being the most common finding. Current treatment recommendations are total thyroidectomy followed by chemotherapy using a BEP (bleomycin, etoposide, and cisplatin) regimen and possibly external beam radiation.72 These recommendations are based on anecdotal reports with no large experiences reported.
MALIGNANT HEMANGIOENDOTHELIOMA These heterogeneous vascular tumors are exceedingly rare, with few cases reported in the literature. Histologically intermediate between a benign hemangioma and a conventional angiosarcoma, these tumors have been classified into four categories including epithelioid, spindle cell, kaposiform, and endovascular papillary. Malignant hemangioendotheliomas were first described by Weiss and Enzinger in 1982 and have similar morphology to other tumors reported in the skin, bone, lung, pleura, liver, peritoneum, and anterior mediastinum.73 They can be distinguished from anaplastic carcinomas by immunohistochemistry and their endothelial origin confirmed by staining for factor VIII –related antigen, ulex europaeus, and CD31.74
Clinical Presentation and Diagnosis Patients tend to present in the seventh decade of life, with men more commonly diagnosed than women (2 : 1). The typical presentation is that of an enlarging neck mass, often with local compressive effects. Metastases to lymph nodes and lung are common at the time of presentation. Histologically, hemangioendotheliomas are often unencapsulated and ill defined. The cells are eosinophilic and pleomorphic with uniform nuclei. Cytoplasmic vacuolization is remarkable and distinctive, suggesting a failed attempt at lumen formation.74 Epithelioid cells predominate and are arranged in nests and short cords embedded in an abundant matrix.
Treatment Hemangioendotheliomas are aggressive tumors. As such, radical resection is the mainstay of therapy.75 External beam radiation as adjuvant therapy for patients with extensive disease or with positive margins has been applied. Radiation may also play a role for palliation in patients with advanced disease. Chemotherapy has been applied in the setting of metastatic or recurrent disease utilizing more typical sarcoma regimens. Extensive experience with this management scheme is lacking. Median survivals range from 2 to 14 months with few long-term survivors reported.75,76
UNCOMMON CANCERS OF THE THYROID
MUCOEPIDERMOID CARCINOMA These tumors are most commonly found in salivary glands, particularly the parotid, but have been described rarely in the thyroid.77 The origin of these tumors is debated, with early reports hypothesizing that they arose from thyroglossal duct remnants and ectopic salivary gland tissue.78,79 This has given way to the two more modern theories that continue to be debated of an ultimobranchial origin based purely on histological evidence80 and a follicular epithelial origin using both histological and molecular evidence.81 Though they typically are reported as low-grade malignancies, aggressive tumors with rapid patient demise have been reported.
Clinical Presentation and Treatment Patients present with a neck mass in the fifth to eighth decades of life. More commonly present in women (2–3 : 1), lymph node metastasis at the time of diagnosis is common. Distant metastases to lungs and bone are rare.81 While lobectomy or subtotal thyroidectomy is often adequate, Steele et al. advocate more aggressive surgery until we understand more about the natural history of the disease.82 Chemotherapy, external beam radiation, and radioactive iodine have all been used in different combination, but limited experience prevents definitive recommendations regarding metastatic or recurrent disease.
SECONDARY THYROID MALIGNANCIES The thyroid gland is an unusual site for clinically evident metastasis, representing 1 to 2% of thyroid malignancies, though autopsy studies report incidence figures as high as 24% in patients dying from metastatic disease.83 These lesions are usually multifocal and rarely the only site of metastasis but may present prior to the primary lesion being found. Renal cell carcinoma is the most common primary, with esophagus, breast, stomach, colon, skin, lung, pancreas, and melanoma also reported.84 The latency period between initial therapy for the primary malignancy and the development of thyroid metastasis varies by tumor type, with very long delays of 7 to 9 years commonly seen in renal cell carcinoma metastases.84 Patients typically present with a thyroid nodule. Diagnosis is made by FNA or thyroidectomy. Cytologic evaluation is most helpful in patients with known malignancies or poor surgical risk. Distinguishing between primary anaplastic carcinoma of the thyroid and poorly differentiated adenocarcinoma metastasis can prove difficult, though positive immunostaining for thyroglobulin suggests a primary thyroid malignancy. Currently, FNA cytology is recommended prior to thyroidectomy, particularly in the face of previous cancer. The management of metastases to the thyroid is individualized. Since it is a harbinger of systemic disease, it is often not curable. Thus, treatment and, ultimately, survival is predicated on the presence of other sites of metastasis and the natural history of the primary. When appropriate, thyroidectomy can result in long-term survival when the thyroid represents the only site of metastasis. Patients with renal cell
171
carcinoma have been reported to fare better following metastasectomy with 56% 4-year survival reported, although this may reflect case selection bias.84 Palliative thyroidectomy is an option for local symptoms but, as expected, few patients survive beyond 2 years.85 Systemic therapy is tailored to the site of origin and is beyond the scope of this chapter.
SUMMARY The thyroid gland is a complex, vascular organ composed of several cell types, making it fertile ground for many unusual malignancies. While the thyroid gland is, in general, the endocrine organ most commonly affected by malignancy, the cancers described above are fairly uncommon. As such, few endocrinologists or surgeons have amassed a great experience with each type, and the relevant literature is generally sparse. Still, a protocol of thorough workup, early diagnosis, and aggressive surgical therapy is a common thread among all tumor types, recognizing the potential of advanced and often incurable disease.
REFERENCES 1. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55(1): 10 – 30. 2. Hazard JB, Hawk WA, Crile G Jr. Medullary (solid) carcinoma of the thyroid; a clinicopathologic entity. J Clin Endocrinol Metab 1959; 19(1): 152 – 61. 3. Williams ED. Histogenesis of medullary carcinoma of the thyroid. J Clin Pathol 1966; 19(2): 114 – 8. 4. Wells SA Jr, et al. Provocative agents and the diagnosis of medullary carcinoma of the thyroid gland. Ann Surg 1978; 188(2): 139 – 41. 5. Komminoth P. The RET proto-oncogene in medullary and papillary thyroid carcinoma. Molecular features, pathophysiology and clinical implications. Virchows Arch 1997; 431(1): 1 – 9. 6. Quayle FJ, Moley JF. Medullary thyroid carcinoma: including MEN 2A and MEN 2B syndromes. J Surg Oncol 2005; 89(3): 122 – 9. 7. Skinner MA, et al. Medullary thyroid carcinoma in children with multiple endocrine neoplasia types 2A and 2B. J Pediatr Surg 1996; 31(1): 177 – 81; discussion 181 – 2. 8. Modigliani E et al., The GETC Study Group. Prognostic factors for survival and for biochemical cure in medullary thyroid carcinoma: results in 899 patients. Groupe d’etude des tumeurs a calcitonine. Clin Endocrinol (Oxf) 1998; 48(3): 265 – 73. 9. Wells SA Jr, et al. Early diagnosis and treatment of medullary thyroid carcinoma. Arch Intern Med 1985; 145(7): 1248 – 52. 10. Rodriguez GJ, et al. Prophylactic thyroidectomy in MEN 2A syndrome: experience in a single center. J Am Coll Surg 2002; 195(2): 159 – 66. 11. Brandi ML, et al. Guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001; 86(12): 5658 – 71. 12. Sone T, et al. Metastatic medullary thyroid cancer: localization with iodine-131 metaiodobenzylguanidine. J Nucl Med 1985; 26(6): 604 – 8. 13. Thomas CC, et al. Detection of medullary carcinoma of the thyroid with I-131 MIBG. Clin Nucl Med 1994; 19(12): 1066 – 8. 14. Edington HD, et al. Radioimmunoimaging of metastatic medullary carcinoma of the thyroid gland using an indium-111-labeled monoclonal antibody to CEA. Surgery 1988; 104(6): 1004 – 10. 15. Frank-Raue K, et al. Somatostatin receptor imaging in persistent medullary thyroid carcinoma. Clin Endocrinol (Oxf) 1995; 42(1): 31 – 7. 16. Krausz Y, et al. Somatostatin receptor scintigraphy for early detection of regional and distant metastases of medullary carcinoma of the thyroid. Clin Nucl Med 1999; 24(4): 256 – 60. 17. Franc B, et al. Medullary thyroid carcinoma: search for histological predictors of survival (109 proband cases analysis). Hum Pathol 1998; 29(10): 1078 – 84.
172
ENDOCRINE TUMORS
18. Evans DB, et al. The surgical treatment of medullary thyroid carcinoma. Semin Surg Oncol 1999; 16(1): 50 – 63. 19. Dralle H, et al. Compartment-oriented microdissection of regional lymph nodes in medullary thyroid carcinoma. Surg Today 1994; 24(2): 112 – 21. 20. Duh QY, et al. Medullary thyroid carcinoma. The need for early diagnosis and total thyroidectomy. Arch Surg 1989; 124(10): 1206 – 10. 21. Moley JF, DeBenedetti MK. Patterns of nodal metastases in palpable medullary thyroid carcinoma: recommendations for extent of node dissection. Ann Surg 1999; 229(6): 880 – 7; discussion 887 – 8. 22. Decker RA, et al. Prophylactic surgery for multiple endocrine neoplasia type IIa after genetic diagnosis: is parathyroid transplantation indicated? World J Surg 1996; 20(7): 814 – 20; discussion 820 – 1. 23. van Heerden JA, et al. Long-term course of patients with persistent hypercalcitoninemia after apparent curative primary surgery for medullary thyroid carcinoma. Ann Surg 1990; 212(4): 395 – 400; discussion 400 – 1. 24. Kouvaraki MA, et al. Surgical management of thyroid carcinoma. J Natl Compr Canc Netw 2005; 3(3): 458 – 66. 25. Moley JF, et al. Surgical management of patients with persistent or recurrent medullary thyroid cancer. J Intern Med 1998; 243(6): 521 – 6. 26. Chen H, et al. Effective long-term palliation of symptomatic, incurable metastatic medullary thyroid cancer by operative resection. Ann Surg 1998; 227(6): 887 – 95. 27. Nguyen TD, et al. Results of postoperative radiation therapy in medullary carcinoma of the thyroid: a retrospective study by the French Federation of Cancer Institutes – the Radiotherapy Cooperative Group. Radiother Oncol 1992; 23(1): 1 – 5. 28. Orlandi F, et al. Chemotherapy with dacarbazine and 5-fluorouracil in advanced medullary thyroid cancer. Ann Oncol 1994; 5(8): 763 – 5. 29. Schlumberger M et al., The Groupe d’Etude des Tumeurs a Calcitonine (GETC). Treatment of advanced medullary thyroid cancer with an alternating combination of 5 FU-streptozocin and 5 FU-dacarbazine. Br J Cancer 1995; 71(2): 363 – 5. 30. Ewing J. Neoplastic Disease, 3rd ed. Philadelphia, Pennsylvania: W. B. Saunders, 1928. 31. Masood S, et al. Differential oncogenic expression in thyroid follicular and Hurthle cell carcinomas. Am J Surg 1993; 166(4): 366 – 8. 32. Tyler DS, et al. Indeterminate fine-needle aspiration biopsy of the thyroid: identification of subgroups at high risk for invasive carcinoma. Surgery 1994; 116(6): 1054 – 60. 33. Massidda B, et al. Hurthle-cell tumors of the thyroid. Minerva Chir 1992; 47(10): 913 – 7. 34. Ditkoff BA, et al. Hurthle cell cancer of the thyroid: the incidence of multifocal and bilateral disease. Thyroidol Clin Exp 1995; 7: 49 – 53. 35. McDonald MP, et al. Hurthle cell carcinoma of the thyroid gland: prognostic factors and results of surgical treatment. Surgery 1996; 120(6): 1000 – 4; discussion 1004 – 5. 36. Caplan RH, Abellera RM, Kisken WA. Hurthle cell neoplasms of the thyroid gland: reassessment of functional capacity. Thyroid 1994; 4(3): 243 – 8. 37. Grossman RF, Clark OH. Hurthle cell Carcinoma. Cancer Control 1997; 4(1): 13 – 7. 38. McIver B, et al. Anaplastic thyroid carcinoma: a 50-year experience at a single institution. Surgery 2001; 130(6): 1028 – 34. 39. Samaan NA, Ordonez NG. Uncommon types of thyroid cancer. Endocrinol Metab Clin North Am 1990; 19(3): 637 – 48. 40. Agrawal S, et al. Histologic trends in thyroid cancer 1969 – 1993: a clinico-pathologic analysis of the relative proportion of anaplastic carcinoma of the thyroid. J Surg Oncol 1996; 63(4): 251 – 5. 41. Hadar T, et al. Anaplastic carcinoma of the thyroid. Eur J Surg Oncol 1993; 19(6): 511 – 6. 42. Venkatesh YS, et al. Anaplastic carcinoma of the thyroid. A clinicopathologic study of 121 cases. Cancer 1990; 66(2): 321 – 30. 43. Cohen Y, et al. BRAF mutation in papillary thyroid carcinoma. J Natl Cancer Inst 2003; 95(8): 625 – 7. 44. Xing M. BRAF mutation in thyroid cancer. Endocr Relat Cancer 2005; 12(2): 245 – 62. 45. Fagin JA, et al. High prevalence of mutations of the p53 gene in poorly differentiated human thyroid carcinomas. J Clin Invest 1993; 91(1): 179 – 84.
46. Garcia-Rostan G, et al. Frequent mutation and nuclear localization of beta-catenin in anaplastic thyroid carcinoma. Cancer Res 1999; 59(8): 1811 – 5. 47. Garcia-Rostan G, et al. Ras mutations are associated with aggressive tumor phenotypes and poor prognosis in thyroid cancer. J Clin Oncol 2003; 21(17): 3226 – 35. 48. Guida T, et al. Mitogenic effects of the up-regulation of minichromosome maintenance proteins in anaplastic thyroid carcinoma. J Clin Endocrinol Metab 2005; 90(8): 4703 – 9. 49. Levendag PC, De Porre PM, van Putten WL. Anaplastic carcinoma of the thyroid gland treated by radiation therapy. Int J Radiat Oncol Biol Phys 1993; 26(1): 125 – 8. 50. Simpson WJ. Anaplastic thyroid carcinoma: a new approach. Can J Surg 1980; 23(1): 25 – 7. 51. Kim JH, Leeper RD. Treatment of anaplastic giant and spindle cell carcinoma of the thyroid gland with combination adriamycin and radiation therapy. A new approach. Cancer 1983; 52(6): 954 – 7. 52. Tennvall J et al., The Swedish Anaplastic Thyroid Cancer Group. Combined doxorubicin, hyperfractionated radiotherapy, and surgery in anaplastic thyroid carcinoma. Report on two protocols. Cancer 1994; 74(4): 1348 – 54. 53. Mitchell G, Huddart R, Harmer C. Phase II evaluation of high dose accelerated radiotherapy for anaplastic thyroid carcinoma. Radiother Oncol 1999; 50(1): 33 – 8. 54. Carcangiu ML, Zampi G, Rosai J. Poorly differentiated (“insular”) thyroid carcinoma. A reinterpretation of Langhans’ “wuchernde Struma”. Am J Surg Pathol 1984; 8(9): 655 – 68. 55. Ashfaq R, et al. Papillary and follicular thyroid carcinomas with an insular component. Cancer 1994; 73(2): 416 – 23. 56. Rijhwani A, Satish GN. Insular carcinoma of the thyroid in a 10-yearold child. J Pediatr Surg 2003; 38(7): 1083 – 5. 57. Sobrinho-Simoes MS, Fonesca E. Recently described tumors of the thyroid. In Anthony PP, Macsween RNM (eds) Recent Advances in Histopathology. Edinburgh: Churchill Livingstone, 1994, Vol. 16: 213 – 229. 58. Flynn SD, et al. Poorly differentiated (“insular”) carcinoma of the thyroid gland: an aggressive subset of differentiated thyroid neoplasms. Surgery 1988; 104(6): 963 – 70. 59. Glass AG, Karnell LH, Menck HR. The national cancer data base report on non-Hodgkin’s lymphoma. Cancer 1997; 80(12): 2311 – 20. 60. Thieblemont C, et al. Primary thyroid lymphoma is a heterogeneous disease. J Clin Endocrinol Metab 2002; 87(1): 105 – 11. 61. Takashima S, et al. Primary thyroid lymphoma: evaluation with US, CT, and MRI. J Comput Assist Tomogr 1995; 19(2): 282 – 8. 62. Skarsgard ED, Connors JM, Robins RE. A current analysis of primary lymphoma of the thyroid. Arch Surg 1991; 126(10): 1199 – 203; discussion 1203 – 4. 63. Widder S, Pasieka JL. Primary thyroid lymphomas. Curr Treat Options Oncol 2004; 5(4): 307 – 13. 64. Galati LT, Barnes EL, Myers EN. Dendritic cell sarcoma of the thyroid. Head Neck 1999; 21(3): 273 – 5. 65. Merimsky O, et al. Sarcoma and thyroid disorders: a common etiology? Oncol Rep 2002; 9(4): 863 – 9. 66. Akbari Y, Richter RM, Papadakis LE. Thyroid carcinoma arising in thyroglossal duct remnants. Report of a case and review of the literature. Arch Surg 1967; 94(2): 235 – 9. 67. LiVolsi VA, Merino MJ. Squamous cells in the human thyroid gland. Am J Surg Pathol 1978; 2(2): 133 – 40. 68. Harada T, et al. Squamous cell carcinoma of the thyroid gland – transition from adenocarcinoma. J Surg Oncol 1982; 19(1): 36 – 43. 69. Sarda AK, et al. Squamous cell carcinoma of the thyroid. J Surg Oncol 1988; 39(3): 175 – 8. 70. Buckley NJ, Burch WM, Leight GS. Malignant teratoma in the thyroid gland of an adult: a case report and a review of the literature. Surgery 1986; 100(5): 932 – 7. 71. Ueno NT, et al. Primary malignant teratoma of the thyroid gland: report and discussion of two cases. Head Neck 1998; 20(7): 649 – 53. 72. Tsang RW, et al. Malignant teratoma of the thyroid: aggressive chemoradiation therapy is required after surgery. Thyroid 2003; 13(4): 401 – 4. 73. Suster S, Moran CA, Koss MN. Epithelioid hemangioendothelioma of the anterior mediastinum. Clinicopathologic, immunohistochemical,
UNCOMMON CANCERS OF THE THYROID
74.
75.
76.
77. 78. 79.
and ultrastructural analysis of 12 cases. Am J Surg Pathol 1994; 18(9): 871 – 81. Siddiqui MT, et al. Epithelioid haemangioendothelioma of the thyroid gland: a case report and review of literature. Histopathology 1998; 32(5): 473 – 6. Rhomberg W, et al. Treatment options for malignant hemangioendotheliomas of the thyroid. Int J Radiat Oncol Biol Phys 2004; 60(2): 401 – 5. Ladurner D, et al. Malignant hemangioendothelioma of the thyroid gland. Pathology, clinical aspects, and prognosis. Wien Klin Wochenschr 1990; 102(9): 256 – 9. Bhandarkar ND, Chan J, Strome M. A rare case of mucoepidermoid carcinoma of the thyroid. Am J Otolaryngol 2005; 26(2): 138 – 41. Diaz-Perez R, Quiroz H, Nishiyama RH. Primary mucinous adenocarcinoma of thyroid gland. Cancer 1976; 38(3): 1323 – 5. Rhatigan RM, Roque JL, Bucher RL. Mucoepidermoid carcinoma of the thyroid gland. Cancer 1977; 39(1): 210 – 4.
173
80. Harach HR. A study on the relationship between solid cell nests and mucoepidermoid carcinoma of the thyroid. Histopathology 1985; 9(2): 195 – 207. 81. Wenig BM, Adair CF, Heffess CS. Primary mucoepidermoid carcinoma of the thyroid gland: a report of six cases and a review of the literature of a follicular epithelial-derived tumor. Hum Pathol 1995; 26(10): 1099 – 108. 82. Steele SR, et al. Mucoepidermoid carcinoma of the thyroid gland: a case report and suggested surgical approach. Am Surg 2001; 67(10): 979 – 83. 83. Berge T, Lundberg S. Cancer in Malmo 1958 – 1969. An autopsy study. Acta Pathol Microbiol Scand Suppl 1977; 260: 1 – 235. 84. Mirallie E, et al. Management and prognosis of metastases to the thyroid gland. J Am Coll Surg 2005; 200(2): 203 – 7. 85. Rosen IB, et al. Secondary malignancy of the thyroid gland and its management. Ann Surg Oncol 1995; 2(3): 252 – 6.
Section 3 : Endocrine Tumors
12
Parathyroid Carcinoma Alliric I. Willis and John A. Ridge
INTRODUCTION Parathyroid carcinoma is a rare malignancy, representing 0.005% of all cancer cases in the U.S. National Cancer Database Report of 1985–1995.1 It is an uncommon cause of hyperparathyroidism, accounting for <5% of primary hyperparathyroid cases.2 – 4 This chapter discusses the pathophysiology, clinical presentation, diagnosis, treatment, and prognosis of this disease.
ANATOMY AND PHYSIOLOGY Normally, four parathyroid glands are present, two superior and two inferior, with an average size and weight of approximately 5 × 3 × 2 mm and 40–60 mg each. The superior parathyroid glands arise from the fourth branchial pouch and descend into the neck to associate with the upper pole of the thyroid. The superior parathyroid glands are usually found along the posterior border of the superior pole of the thyroid and may be found within the gland’s parenchyma. Infrequently, the superior glands continue their migration caudally, and they may be found anywhere along the course of the tracheoesophageal groove into the posterior mediastinum.5,6 The inferior parathyroid glands arise from the third branchial pouch in association with the thymus. They are found within the thymus or thyrothymic ligament, but can vary significantly in location from the skull base to the anterior mediastinum.5,6 Each gland, surrounded by a thin capsule, contains parathyroid cells that regulate calcium homeostasis by sensing serum calcium levels through calcium receptors and secreting parathyroid hormone regulated by a negative feedback loop. Adenomatous, hyperplastic, or malignant change in parathyroid glands results in the loss of feedback inhibition, increased secretion of parathyroid hormone, and subsequent hypercalcemia.
EPIDEMIOLOGY Parathyroid carcinoma is uncommon with a reported incidence from <1% to as high as 5% among cases of primary
hyperparathyroidism. In earlier series published in the United States, the incidence ranged from 1% (by Cope) in 1966 to as high as 4% (by Schantz and Castleman) in 1973.2,7,8 More recently, reported estimates of incidence have been as low as <1% in the United States.9,10 In series from Italy and Japan, it has been reported as nearly 5%.11 – 13 Obara and Fujimoto, in a collective review, found an incidence of 2.1% among more than 4000 cases of primary hyperparathyroidism.3 It is not clear whether the lower incidence in more recently reported series of parathyroid cancer cases in the United States (in comparison with those from Japan, Italy, and earlier United States series) is due to differences in populations or to increased calcium screening and incidental diagnosis of primary hyperparathyroidism in the United States. In contrast to hyperparathyroidism, which predominantly affects women, there is no gender predilection with parathyroid carcinoma. The largest series of parathyroid carcinoma patients, collected from the U.S. National Cancer Database between 1985 and 1995, found a nearly even ratio of 51% men to 49% women.1 Parathyroid carcinoma tends to present at a younger age than benign hyperparathyroidism, often in the fifth decade of life compared with primary hyperparathyroidism, which frequently presents in the sixth decade.14
PATHOLOGY The classic histologic criteria for parathyroid cancers were articulated by Schantz and Castleman. They distinguished parathyroid carcinoma as a separate entity and found that there was no evidence that cancer developed from adenomas or hyperplasia of the parathyroid glands. In diagnosing parathyroid cancer, they found fibrous trabeculae in 90%, mitotic figures in 81%, capsular invasion in 67%, and blood vessel invasion in 12%.2 As evidenced by their frequency, such findings are not consistent. A recent review of 27 cases of parathyroid cancer reported fibrous bands in 44%, mitotic figures in 40%, vascular invasion in 37%, capsular invasion in 26%, trabeculae in 11%, and lymphatic invasion in 11%.15 Another study compared histopathologic features found in 16 parathyroid carcinomas with those found in 45 typical adenomas and 8 atypical adenomas. The qualities that were found to be most
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
PARATHYROID CARCINOMA
specific for parathyroid carcinoma, and not present in typical or atypical adenomas, were capsular invasion (94%), adjacent soft tissue invasion (69%), and vascular invasion (81%).16 A series of 95 cases of parathyroid carcinoma from the Swedish Cancer Registry reported on reevaluations of histopathology by pathologists unaware of the initial diagnosis. At the first surgical procedure, only 74% of cases were correctly diagnosed as parathyroid carcinoma, and in 19% of cases a benign initial diagnosis was rendered. On reevaluation, 43% were diagnosed unequivocally as malignant, while 57% were read as histologically equivocal, that is, displaying varying combinations of suspicious findings.13 These results highlight the difficulty of assigning a conclusive diagnosis of this rare cancer. There is no standard staging system for parathyroid cancer. A TNM (tumor, nodes, metastasis) staging system is difficult to develop for parathyroid carcinoma because tumor size and lymph node status at presentation have not been found to influence survival for this disease. However, Shaha and Shah proposed a TNM-based staging system focused upon tumor size greater or less than 3 cm, the presence of regional lymph node metastases, and the presence of distant metastases.1,17 Given the invasive nature of parathyroid cancer, one potentially useful means of characterizing the extent of disease is limited to the parathyroid gland, locally invasive, or having metastases. A histological classification system may give an indication of prognosis. In one proposed system, capsular and vascular invasion identifies parathyroid carcinoma. Lowgrade lesions with limited local invasion correlated with cure, while high-grade tumors with widespread infiltration correlated with metastases.18
GENETICS No specific gene has been incriminated as a cause of parathyroid carcinoma; however, advances in research have implicated genes that may be involved in the pathogenesis of this malignancy. The HRPT2 gene is a tumor suppressor gene that may contribute to the pathogenesis of parathyroid carcinoma when mutated.19 The normal HRPT2 gene encodes the protein parafibromin.20,21 Parafibromin has been shown to be a tumor suppressor by transfection studies in which wild-type parafibromin inhibited tumor cell growth, but transfection with missense parafibromin mutants had no impact on cell growth.22
175
HRPT2 gene mutations have been identified in patients with the hyperparathyroidism-jaw tumor (HPT-JT) syndrome, a rare, autosomal-dominant syndrome that includes hyperparathyroidism (90%), tumors of the mandible and maxilla (30%), parathyroid cancer (15%), renal cysts (10%), and rare solid renal tumors.21,22 Parathyroid carcinoma is overrepresented among HPT-JT patients with a frequency of 15% (in contrast to < 1% –5% among all patients with hyperparathyroidism).23 In 15 patients with sporadic parathyroid carcinomas, 10 had HRPT2 mutations and 30% of the mutations were germ-line. Each of these mutations resulted in failure of production of parafibromin, thereby implicating HRPT2 and its product, parafibromin, as tumor suppressors of parathyroid cancer.19 Cyclin D1, an oncogene located at chromosome band 11q13, is overexpressed in parathyroid adenomas and carcinomas.23 The level of cyclin D1 is significantly higher in parathyroid carcinomas than adenomas.22 Transfection of NIH3T3 mouse cells and HEK-293 human cells with wild-type parafibromin inhibited expression of cyclin D1. Transfection with mutant parafibromin had no impact on cyclin D1 expression. These findings suggest that HRPT2 mutations with loss of parafibromin expression may result in increases in cyclin D1 expression and contribute to neoplasia. However, parafibromin is present in other tissues without neoplastic transformation associated with mutations.22
CLINICAL PRESENTATION The clinical presentation of patients with parathyroid carcinoma almost invariably results from hypercalcemia due to the parathyroid hormone produced by the tumor. Symptoms include the classic “bones, stones, groans, and psychic overtones” associated with hypercalcemia of hyperparathyroidism; however, in patients with parathyroid carcinoma, they occur more frequently and in more severe forms. The clinical manifestations at presentation of hypercalcemia from parathyroid carcinoma in several reported series are shown in Table 1. The organ systems most commonly affected are skeletal and renal. Skeletal complications including osteopenia, osteoporosis, bone pain, or pathologic fractures are reported in 30–91% of cases. Renal complications including calculi and intrinsic renal failure are reported in 21–60% of cases.
Table 1 Frequency of clinical manifestations of hypercalcemia with parathyroid carcinoma.
Busaidy et al.15 No. of patients Skeletal Renal Neuromuscular 33%c , Pancreatitis Gastrointestinal Asymptomatic a
Calculi. Intrinsic. c Weakness, lethargy. d Myalgia, arthralgia. e Headaches. b
27 30% 44% 22%d , 26%e 4% 15% 30%
Iacobone et al.11 19 63% 53% 37% 16% 21%
Wynne et al.24 40 91% (20/22) 56% (14/25)a , 38%b 27%d
7%
Obara and Fujimoto.3 Shane and Bilezikian.9 Schantz and Castleman.2 163
62
39% 48%
70% 60%
5%
10%
2%
2%
61 62% 30%a , 21%b 8%c 10% 13%
176
ENDOCRINE TUMORS
Neuromuscular and neuropsychiatric manifestations are frequently vague in presentation but can often be elicited on history and physical examination including a thorough review of systems. In a series reported by Busaidy et al., symptoms of fatigue were reported in 33%, arthralgia in 22%, and headaches in 26%.15 The physical finding most associated with parathyroid carcinoma is a palpable cervical lymph node. Cervical lymphadenopathy is present on examination of patients with parathyroid carcinoma with reported frequencies of 15–75%.2,3,13,15,16,24 However, it is exceedingly rare in patients with benign causes of primary hyperparathyroidism. Stojadinovic et al. studied a series of 73 patients with parathyroid adenomas, atypical adenomas, and carcinomas. They reported the incidence of a palpable neck mass in carcinoma, 75% (16/20), versus atypical adenoma, 25% (2/8), and adenoma 0% (0/45).16 A serum calcium level greater than 14 mg dL−1 should raise suspicion of parathyroid carcinoma. Average calcium levels measured in patients with parathyroid cancer in multiple published series range from 13.4 to 15.9 mg dL−1 .2,3,9,13,15,16,24,25 In a published series comparing parathyroid carcinoma to atypical adenomas and typical adenomas, the average serum calcium levels were 14, 12, and 11 mg dL−1 respectively.16 Patients with parathyroid carcinoma show markedly elevated parathyroid hormone levels. Primary hyperparathyroidism is usually associated with parathyroid hormone levels less than twice the upper limit of normal values. Patients with parathyroid carcinoma often have parathyroid hormone levels five times greater than the upper limit of normal values.3,13,24 Table 2 shows average calcium levels and the magnitudes of parathyroid hormone elevations from published series in which both values were reported. Imaging studies that are useful in operative planning for parathyroid carcinoma are those commonly employed in primary hyperparathyroidism. The sestamibi nuclear medicine scan is the most sensitive test for identifying parathyroid tumors. Sestamibi with single photon emission computed tomography (SPECT) adds three-dimensional imaging. Ultrasound is increasingly employed for imaging the neck and can identify tumor invasion of local structures. Computed tomography (CT) scan and magnetic resonance imaging (MRI) are useful for imaging the mediastinum and sites of distant metastases. Needle biopsy of suspected parathyroid carcinoma should not be performed because of the risk of tumor seeding.3,10
Table 2 Serum calcium and parathyroid hormone levels in patients with parathyroid carcinoma.
Author Wynne et al.24 Sandelin et al.13 Obara and Fujimoto,3 a
Serum calcium (No. of patients)
Magnitude of parathyroid hormone level elevationa (No. of patients)
14.6 mg dL−1 (43) 14.4 mg dL−1 (95) 15.0 mg dL−1 (163)
10.2x (21) 1.3 – 75x (24) >5x (53)
Elevation is the multiple of the upper limit of normal values.
OPERATIVE FINDINGS Clinical diagnosis of parathyroid carcinoma at operation is important so that the appropriate cancer resection can be performed. Tumor appearance, size, and local invasion are important. The appearance often differs markedly from a typical parathyroid adenoma. Instead, the surgeon encounters a gray-white, lobulated, firm tumor greater than 2.5 cm in diameter, weighing more than 4 g, and invading or adherent to adjacent structures. The average diameter of parathyroid carcinoma tumors in published series ranges 2.5–3.3 cm.2,13,15 The largest reported series of parathyroid carcinomas, from the U.S. National Cancer Database, reported an average tumor diameter of 3.3 cm in 174 measurable cases.1 Schantz and Castleman reported a weight range of 0.8–42.4 g in 24 cases.2 The weight range reported in a published series from the Swedish Cancer Registry was 1.05–40 g in 35 cases.13 Parathyroid carcinoma tends to arise in inferior parathyroid glands. In a study of 16 cases, 6 had cancers in the left inferior gland, 9 were in the right inferior gland, and 1 was in a mediastinal fifth gland, while no cases involved superior glands.4 In another series, out of 19 cases with tumor locations reported, 15 involved inferior glands, 3 involved superior glands, and 1 involved a mediastinal gland.15 An important distinguishing feature of parathyroid carcinoma at operation is tumor invasion of local tissue and adjacent structures. Invasion may occur into local adipose tissue or into adjacent structures such as thyroid, esophagus, and recurrent laryngeal nerve. The incidence of local invasion is 44–70% in reported series, with invasion of fat most common.13,15 Invasion of adjacent structures has been reported as 23–37%.3,15 Structures most frequently invaded are the thyroid, recurrent laryngeal nerve, esophagus, trachea, and strap muscles. In a series of 163 patients with parathyroid carcinoma, 38 had invasion of local structures with the thyroid (24) and recurrent laryngeal nerve (6) most frequently involved.3 Cervical lymph node management is controversial. The incidence of cervical lymph node metastasis has been reported as high as 32%, which resulted in recommendations for radical lymph node dissection at the initial operation.25 However, several reported series cite an incidence of cervical lymph node involvement at less than 5%.3,4,13 Hence, there has been no standard surgical approach to cervical lymph nodes. The lack of a standard regarding lymph node dissection is likely a confounding factor in the reporting of the incidence of lymph node metastases at presentation. The Swedish Cancer Registry found only three lymph node metastases in 95 cases; however, 84 cases (88%) had resections without lymphadenectomy.13 In published series in which lymph node dissections were performed, the incidence of lymph node metastasis at initial operation is 11–15%.1,2,15 From the U.S. National Cancer Database review of 286 cases, only 105 had cervical lymph node status reported, of which 15% showed lymph node involvement at initial operation.1
PARATHYROID CARCINOMA
TREATMENT The only curative treatment for parathyroid carcinoma is surgical resection. The excision should be performed without disruption of the tumor capsule, as this can result in tumor seeding. Similarly, disruption of adhesions between tumor and local structures can result in tumor seeding. Local cervical lymph nodes may harbor metastases in as many as 15% of patients. Therefore, the preferred operation for parathyroid carcinoma is en bloc resection including tumor, adherent tissues, ipsilateral thyroid, with ipsilateral lymphadenectomy of levels IV, VI, and the tracheoesophageal groove.3,10,14,17 The recurrent laryngeal nerve should be sacrificed if involvement is recognized preoperatively by a paralyzed vocal cord or if invasion is identified at operation. If cervical lymph nodes are clinically positive at operation, then a comprehensive cervical lymphadenectomy should be performed.17 The majority of patients with parathyroid carcinoma undergo an inadequate initial operation. This is probably due to lack of recognition of parathyroid carcinoma at operation in the overwhelming majority of cases.1 Many patients are not identified until recurrence. As many as 19% of cases are read initially as benign by pathologists.13 Analysis of the U.S. National Cancer Database results found that only 13% of operations performed for patients with parathyroid carcinoma were the recommended en bloc resection of tumor with adjacent structures and lymphatic tissue, while 60% underwent resection of parathyroid glands only, and 6% had excisional biopsies performed. The inadequacy of resection is further exemplified by the finding that 39% of cases did not report tumor size, suggesting that the tumor was disrupted or incompletely resected, which could result in tumor seeding.1 Analysis of results of the Swedish Cancer Registry found that only 44% of cases received the recommended operation for parathyroid carcinoma (tumor resection with partial or total thyroidectomy, with or without neck dissection). Resection of tumor alone or multiple parathyroid glands occurred in 51% of cases. Multivariate analysis showed that patients treated with more extensive, en bloc operations had a longer relapse-free period and longer survival.13 A published series of seven cases suggests the importance of a complete initial operation. Two patients underwent en bloc resection of tumor and ipsilateral thyroid with adherent structures as well as ipsilateral central compartment lymph node dissection and ipsilateral thymectomy, with no adjuvant therapy. They were cured. The other five patients underwent lesser operations; three developed lung metastases, and two died from recurrent or metastatic parathyroid carcinoma. Of note is that one of those five patients had a tumor resection alone that was read as complete, but local recurrence and lung metastases ensued and led to death from disease.26 Patients with local recurrence or distant metastasis should be submitted to reoperation and metastasectomy, with the goal of reducing parathyroid hormone levels and hypercalcemia.11,27 There is no established role for radiation therapy in the treatment of parathyroid carcinoma. However, several small series studying radiation therapy in
177
the adjuvant setting have shown promising results in prevention of local relapse in patients with locally advanced disease.15,28 The results in patients with bulk recurrence have not been as favorable.26 Chemotherapy has not been an effective treatment option for parathyroid carcinoma. Regimens including nitrogen mustard; adriamycin, cyclophosphamide, and 5-fluorouracil; and others have been unsuccessful.14,23 Calcium-reducing agents, including bisphosphonates, plicamycin, and calcitonin, have been utilized in the treatment of hypercalcemia secondary to parathyroid carcinoma with varying success and without durable effect.3,14 Calcimimetics work by modulating the calcium receptor. They are molecules that allosterically modulate the calcium receptor selectively by increasing receptor sensitivity to ionized calcium levels, affecting the negative feedback loop and decreasing parathyroid hormone secretion from parathyroid cells.29 This treatment has been successful in primary and secondary hyperparathyroidism and may have potential in treating hypercalcemia secondary to parathyroid carcinoma.14,29
PROGNOSIS Parathyroid carcinoma often recurs. Reported rates range from 22 to 100%, perhaps owing in part to the fact that many cases are not appreciated until review at the time of recurrence.27 This frequently occurs years after the initial presentation. The diagnosis is often assigned on the basis of recurrent hypercalcemia and hyperparathyroidism. Metastases may occur through either lymphatic or hematogenous routes. The most common sites of recurrence are cervical lymph nodes and the tumor bed (50–75%), followed by lung (22–40%), liver (10–28%), and bone (5–28%).2,13,17 Other sites, such as brain, are reported, but rare.30 Tumor can be detected by means of sestamibi, CT, and MRI.10,27 Schantz and Castleman reported a recurrence rate of 30% in 59 cases with a clear relationship between early recurrence and death from the disease. Ten patients had recurrence within 2 years of initial diagnosis and nine (90%) died from recurrent tumor. Eight patients had recurrence more than 2 years after presentation and only three of the eight (38%) died of recurrent disease.2 In a large series reporting on disease recurrence in 95 parathyroid carcinoma patients, the Swedish Cancer Registry found that the incidence of recurrence was 42% with an average time to recurrence of 33 months (range 1–228 months). A significant advantage for en bloc resection was demonstrated with a recurrence incidence of 21% after en bloc resection, versus 58% after tumor resection alone (p = 0.0002 for recurrence, p = 0.003 for survival).13 This study is particularly important as it demonstrated that the initial type of operation, en bloc resection, has a statistically significant impact on recurrence and survival; a benefit that was not able to be established by smaller studies.13,15,27 Multivariate analysis of the Swedish Cancer Registry demonstrated that the type of operation (en bloc versus tumor resection) and histopathology (invasive versus atypical) were statistically significant factors correlating with both increased disease-free and overall survival.13 Evaluation of the U.S.
178
ENDOCRINE TUMORS
Table 3 Recurrence and survival rates for parathyroid carcinoma.
Author Kleinpeter et al.27 Busaidy et al.15 Hundahl et al.1 Sandelin et al.13
Table 4 Summary of diagnosis and treatment of parathyroid carcinoma.
Number of Recurrence 5-year 10-Year patients incidence (%) survival (%) survival (%) 23 26 286 95
22 42 N/A 42
86 85 86 85
69 77 49 70
National Cancer Database found that tumor size and lymph node status had no significant impact on survival.1 Favorable long-term survival rates are achievable in patients with parathyroid carcinoma. Five-year survival rates were reported as 85% from the U.S. National Cancer Database and Swedish Cancer Registry, and 50–90% in other series.1,13,16,27 Ten-year survival of 49–77% has been reported.1,13,27 The cause of death is hypercalcemia in the overwhelming majority of cases.2,3,15,16 Table 3 shows recurrence and survival rates for parathyroid carcinoma.
Presentation • Skeletal and renal complications are common Physical Examination • Palpable cervical mass Laboratory Values • Calcium >14 mg dL−1 • Parathyroid hormone level increased >5x Radiology • Sestamibi, ultrasound Operative Findings • Gross: Gray-white tumor (>3 cm) adherent to or invading local tissue and structures • Micro: Invasion through gland capsule into surrounding tissue, lymphatics, and vessels Treatment • En bloc resection of tumor, adherent tissue and structures, ipsilateral thyroid, and ipsilateral lymphadenectomy of the central compartment and lower cervical chain • For recurrence: reoperation and metastasectomy. Radiation has no established role. Chemotherapy has been ineffective.
SUMMARY Parathyroid carcinoma is an uncommon cause of primary hyperparathyroidism. The morbidity and mortality are related to hypercalcemia resulting from tumor production of parathyroid hormone. The initial goals of therapy for this cancer are complete removal of the tumor and prevention of recurrence. The first-line treatment is adequate resection at the initial operation. This consists of en bloc resection of tumor, ipsilateral thyroid, adherent tissues, and regional lymphadenectomy (including the ipsilateral central compartment and lower cervical chain). It is important to be aware of preoperative findings that might suggest the diagnosis of parathyroid carcinoma. These include: skeletal and renal complications from hypercalcemia, the presence of a palpable neck mass, and markedly elevated serum calcium and parathyroid hormone levels. Sestamibi scanning is the most sensitive noninvasive study for localization, while the addition of ultrasound can be helpful for identifying tumor invasion of surrounding structures. At operation, it is essential to recognize findings indicative of parathyroid carcinoma, such as tumor invasion or adherence to surrounding structures. Otherwise, the initial operation, the best opportunity for cure, will be lost. The frequency of recurrence in parathyroid carcinoma is high; however, good long-term survival rates can be achieved. Operation is the most effective treatment for recurrence or metastasis. Second-line treatment options with calcium-reducing medications may serve a future role in reducing morbidity as advances are made. A summary of diagnosis and treatment of parathyroid carcinoma is given in Table 4.
REFERENCES 1. Hundahl SA, et al. Two hundred eighty-six cases of parathyroid carcinoma treated in the U.S. between 1985 – 1995: a National Cancer Data Base Report. The American College of Surgeons Commission on Cancer and the American Cancer Society. Cancer 1999; 86(3): 538 – 44.
2. Schantz A, Castleman B. Parathyroid carcinoma. A study of 70 cases. Cancer 1973; 31(3): 600 – 5. 3. Obara T, Fujimoto Y. Diagnosis and treatment of patients with parathyroid carcinoma: an update and review. World J Surg 1991; 15(6): 738 – 44. 4. Favia G, et al. Parathyroid carcinoma: sixteen new cases and suggestions for correct management. World J Surg 1998; 22(12): 1225 – 30. 5. van Heerden JA, Smith SL. Parathyroidectomy for Primary Hyperparathyroidism (Adenoma and Carcinoma), 3rd ed. Boston, Massachusetts: Little, Brown and Company, 1997. 6. Doherty GM, Wells SA. Parathyroid glands. In Greenfield LJ, Mulholland MW (eds) Surgery: Scientific Principles and Practice. Philadelphia, Pennsylvania: Lippincott-Raven Publishers, 1997. 7. Cope O. The study of hyperparathyroidism at the Massachusetts general hospital. N Engl J Med 1966; 274(21): 1174 – 82. 8. Cope O, Nardi GL, Castleman B. Carcinoma of the parathyroid glands: 4 cases among 148 patients with hyperparathyroidism. Ann Surg 1953; 138(4): 661 – 71. 9. Shane E, Bilezikian JP. Parathyroid carcinoma: a review of 62 patients. Endocr Rev 1982; 3(2): 218 – 26. 10. Fraker DL. Update on the management of parathyroid tumors. Curr Opin Oncol 2000; 12(1): 41 – 8. 11. Iacobone M, Lumachi F, Favia G. Up-to-date on parathyroid carcinoma: analysis of an experience of 19 cases. J Surg Oncol 2004; 88(4): 223 – 8. 12. Fujimoto Y, et al. Surgical treatment of ten cases of parathyroid carcinoma: importance of an initial en bloc tumor resection. World J Surg 1984; 8(3): 392 – 400. 13. Sandelin K, et al. Prognostic factors in parathyroid cancer: a review of 95 cases. World J Surg 1992; 16(4): 724 – 31. 14. Shane E. Clinical review 122: parathyroid carcinoma. J Clin Endocrinol Metab 2001; 86(2): 485 – 93. 15. Busaidy NL, et al. Parathyroid carcinoma: a 22-year experience. Head Neck 2004; 26(8): 716 – 26. 16. Stojadinovic A, et al. Parathyroid neoplasms: clinical, histopathological, and tissue microarray-based molecular analysis. Hum Pathol 2003; 34(1): 54 – 64. 17. Shaha AR, Shah JP. Parathyroid carcinoma. Cancer 1999; 86(3): 378 – 80. 18. Kameyama K, Takami H. Proposal for the histological classification of parathyroid carcinoma. Endocr Pathol 2005; 16(1): 49 – 52. 19. Shattuck TM, et al. Somatic and germ-line mutations of the HRPT2 gene in sporadic parathyroid carcinoma. N Engl J Med 2003; 349(18): 1722 – 9.
PARATHYROID CARCINOMA 20. Carpten JD, et al. HRPT2, encoding parafibromin, is mutated in hyperparathyroidism-jaw tumor syndrome. Nat Genet 2002; 32(4): 676 – 80. 21. Szabo J, et al. Hereditary hyperparathyroidism-jaw tumor syndrome: the endocrine tumor gene HRPT2 maps to chromosome 1q21-q31. Am J Hum Genet 1995; 56(4): 944 – 50. 22. Woodard GE, et al. Parafibromin, product of the hyperparathyroidismjaw tumor syndrome gene HRPT2, regulates cyclin D1/PRAD1 expression. Oncogene 2005; 24(7): 1272 – 6. 23. Mittendorf EA, McHenry CR. Parathyroid carcinoma. J Surg Oncol 2005; 89(3): 136 – 42. 24. Wynne AG, et al. Parathyroid carcinoma: clinical and pathologic features in 43 patients. Medicine (Baltimore) 1992; 71(4): 197 – 205.
179
25. Holmes EC, Morton DL, Ketcham AS. Parathyroid carcinoma: a collective review. Ann Surg 1969; 169(4): 631 – 40. 26. Kirkby-Bott J, et al. One stage treatment of parathyroid cancer. Eur J Surg Oncol 2005; 31(1): 78 – 83. 27. Kleinpeter KP, et al. Is parathyroid carcinoma indeed a lethal disease? Ann Surg Oncol 2005; 12(3): 260 – 6. 28. Munson ND, et al. Parathyroid carcinoma: is there a role for adjuvant radiation therapy? Cancer 2003; 98(11): 2378 – 84. 29. Weigel RJ. Nonoperative management of hyperparathyroidism: present and future. Curr Opin Oncol 2001; 13(1): 33 – 8. 30. Kern M, et al. Intracranial metastatic parathyroid carcinoma. Case report and review of the literature. J Neurosurg 2004; 101(6): 1065 – 9.
Section 4 : Breast Cancer
13
Metaplastic Breast Carcinoma Helenice Gobbi, Ingrid A. Mayer and A. Bapsi Chakravarthy
INTRODUCTION Metaplastic carcinomas represent less than 5% of all breast carcinomas. They are characterized by tumors containing both epithelial and mesenchymal cell types. The World Health Organization (WHO) classifies metaplastic carcinomas into pure epithelial metaplastic carcinomas and mixed epithelial–mesenchymal carcinomas. From a diagnostic viewpoint classifying metaplastic carcinomas as monophasic or biphasic sarcomatoid carcinomas is more appropriate. Macroscopically, the majority of the tumors are circumscribed, hard, and rubbery. Microscopically, a spindle cell pattern is the most common. It can be pure or in combination with squamous, glandular, or sarcomatous elements, such as, bone, cartilage, and osteoclastic-like giant cells. The pathologic phenotype is the best predictor of the clinical behavior of metaplastic carcinoma. The incidence of lymph node metastases is generally low in metaplastic breast tumors. Because of its rarity, treatment decisions are made based on limited retrospective data. Following histological confirmation, surgical options should be discussed with the patient. Breast conservation can be offered to patients in whom a margin-negative excision can be obtained. Adjuvant therapy with anthracycline-containing regimens should be discussed with patients who have a component of adenocarcinoma or when lymph nodes are involved. Tamoxifen or aromatase inhibitors should be considered in patients with estrogen receptor (ER) and/or progesterone receptor (PR) positive tumors. Although the optimal agents in the metastatic setting are unknown, we would advise the use of agents that are used in the more usual infiltrating ductal adenocarcinomas.
HISTORICAL BACKGROUND The term metaplastic carcinoma is used for a heterogeneous group of tumors containing cells of both epithelial and mesenchymal phenotype. The metaplastic changes include epithelial (squamous) and sarcomatous elements (osseous, chondroid, loose spindle and fibromyxoid stroma, dense spindle and fibrosarcomatoid stroma).1 Various synonyms have
been used for similar lesions but the large number of terms has not helped to better understand the biology and prognosis of these tumors.2 In the new WHO classification of breast tumors, metaplastic carcinomas were grouped as: pure epithelial metaplastic carcinomas (including squamous cell carcinoma, adenocarcinoma with spindle cell metaplasia, adenosquamous carcinoma, and mucoepidermoid carcinoma) and mixed epithelial–mesenchymal metaplastic carcinomas.3 The WHO classification is difficult to follow because spindle cell metaplastic carcinomas without admixed adenocarcinoma, heterologous elements or squamous differentiation are neither included in the “purely epithelial” nor the “mixed epithelial and mesenchymal” category.4 The classification of metaplastic carcinomas into two main categories of monophasic (“sarcomatoid” or spindle cell carcinomas) and biphasic “sarcomatoid” carcinomas (“carcinosarcomas” or “malignant mixed tumors”) would be more appropriate from a diagnostic point of view.2
BIOLOGY AND EPIDEMIOLOGY The incidence of metaplastic carcinoma represents less than 5% of breast malignancies. However, an accurate assessment of the incidence of metaplastic breast cancers is difficult because these tumors have been designated a variety of names, and authors tend to report incidences based on specific subtypes. There are many examples of initial “misdiagnosis” in the literature. High-grade monophasic tumors have been classified as pure sarcomas and low-grade spindle cell metaplastic tumors are often misinterpreted as fibromatosis.5 – 7 Metaplastic carcinomas may arise within fibrosclerotic breast lesions such as papillomas, complex sclerosing lesions, and nipple adenomas.8,9 Two main variants of low-grade metaplastic tumors are described in association with fibrosclerotic lesions: fibromatosis-like tumors6 and adenosquamous carcinoma.8 The neoplastic nature of such low-grade tumors is sometimes difficult to distinguish from reactive myofibroblastic proliferation that can occur within complex sclerosing lesions and papillomas.10
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
182
BREAST CANCER
PATHOLOGY Gross Pathology Metaplastic tumors show no specific macroscopic features, but the majority of cases have been reported to be circumscribed, hard, and rubbery. Tumors range in size from 0.5 to 21 cm, with an average size of 3 to 5 cm. Heterologous elements, especially bone and cartilage may be evident on gross examination. Tumors that are predominately squamous may have cystic degeneration.7,11 – 13 Although spindle cell carcinomas are often grossly spiculated or well circumscribed, microscopically they show a more infiltrative border.6,7
Microscopic Appearances Because the microscopic pathology of metaplastic breast tumors is so varied, multiple histological sections are necessary for an accurate diagnosis of these lesions and to determine whether the tumor is of monophasic (pure spindle cell pattern) or biphasic (epithelial and mesenchymal) morphology.2 Although in most cases two patterns, a dominant and a minor pattern are recognized, more often they are intermixed. The most common pattern of metaplastic tumors is the spindle cell pattern, which can be pure or in combination with squamous, glandular, or sarcomatous elements, such as, bone, cartilage, and osteoclastic-like giant cells.
Figure 1 Fibromatosis-like metaplastic tumor showing bland spindle cell proliferation with few glandular elements (center and left). H&E, ×200.
Spindle Cell Metaplastic Carcinoma Tumors comprised of dominant spindle cells frequently associated with an invasive squamous or glandular component are referred to as monophasic sarcomatous carcinoma, or spindle cell metaplastic carcinomas. The spindle cells in these tumors can vary from a relatively bland appearance to aggressive patterns resembling high-grade sarcomas.7 The lowest-grade spindle cell metaplastic tumors show low cellularity, intermixed with more fibromyxoid stroma, with only sparse recognizable squamous or glandular elements closely mimicking fibromatosis (see Figure 1). We chose the term “fibromatosis-like metaplastic tumors” for such lesions, to avoid the word carcinomas because neither the phenotype nor the behavior is that of a carcinoma.6,14 In tumors that are almost entirely composed of spindle cells, the epithelial elements may be difficult to recognize.7,15,16 The spindle cells coexpress myoepithelial (actins, p63, CD10, maspin, P-cadherin) and mesenchymal markers such as vimentin. The consistent expression of myoepithelial markers in spindle cell metaplastic carcinomas is helpful in the diagnosis and indicates myoepithelial differentiation of such tumors.4,6,17,18 The spindle cells of metaplastic carcinomas are usually negative for ER and PR and Her-2/neu. Positivity, when present in these tumors, is confined to the adenocarcinomatous component.19
Adenosquamous Carcinoma Adenosquamous carcinoma is a variant of metaplastic carcinoma composed of small glandular structures intimately admixed with variable amounts of solid nests of squamous differentiation in a collagenous stroma (see Figure 2).11,20
Figure 2 Low-grade adenosquamous metaplastic carcinoma showing islands of squamous cells merging with the spindle cells. H&E, ×200.
Squamous metaplasia varies from syringoma-like differentiation to well differentiated keratinizing areas and poorly differentiated nonkeratinizing foci. Although benign squamous metaplasia has been described in different types of benign lesions and invasive mammary carcinomas (representing <5 to 10% of the tumor) the term “metaplastic carcinoma” is reserved for tumors that show metaplasia as a dominant pattern.10,11 The squamous component is negative for ER, PR, and Her-2/neu while the positivity of the glandular component is dependent on its degree of differentiation.19 Primary squamous carcinoma of the breast denotes a carcinoma entirely composed of metaplastic squamous cells with no connection to the skin or continuity with the nipple. Tumors with less squamous differentiation are referred to as adenocarcinomas with extensive squamous differentiation
METAPLASTIC BREAST CARCINOMA
or adenosquamous carcinomas. Squamous cell carcinomas assume several phenotypes including keratinizing, nonkeratinizing, or spindled.3,21
Metaplastic Carcinoma with Sarcomatous Metaplasia A broad range of patterns is seen in this wide variety of tumors including those designated as “matrix-producing carcinomas” (see Figure 3) and carcinosarcomas.3 These tumors are composed of infiltrating carcinoma mixed with spindle cells and heterologous elements ranging from bland chondroid and osseous differentiation to high-grade sarcomas including fibrosarcoma, malignant fibrous histiocytoma, osteosarcoma, chondrosarcoma, liposarcoma, rhabdomyosarcoma, and leiomyosarcoma. The epithelial component of the sarcomatous metaplastic carcinomas may have squamous features, but it is more commonly of ductal nonspecial type and usually of grade 2 or 3 morphology.3 When the mesenchymal component is malignant, the designation of carcinosarcoma has been used.3,13 Some authors use the term “pseudosarcomatous” to describe these lesions.22 Although heterologous elements, such as bone and cartilage, may be associated with metaplastic carcinoma, foci of benign heterologous elements may also be present in other mammary carcinomas without known adverse clinical implications.23 The majority of biphasic sarcomatous carcinomas are negative for ER, PR, and Her-2/neu both in the adenocarcinoma and the mesenchymal elements. The spindle cells may show focal positivity for cytokeratin and p63.18,19,23
Grade The proportion and grading of epithelial and sarcomatous elements present in metaplastic tumors of the breast would be expected to have potential clinical relevance.6,14 However, in most published series a correlation of histologic grade and prognosis has not been considered, despite the accepted
183
clinical utility of histologic grading in both – soft tissue sarcoma and mammary carcinoma.7 Similarly, grading of sarcomatous elements in metaplastic carcinomas would be expected to have potential clinical relevance. It is also important to grade the carcinomatous component when it is significant in extent. In general, most cases with predominant squamous and glandular lesions are intermediate to highgrade adenocarcinomas with areas of squamous metaplasia of similar grade. It is recommended that squamous carcinomas should be graded mainly on the basis of nuclear features and, to a lesser degree, on cytoplasmic differentiation.3
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS As these tumors can contain both carcinomatous and mesenchymal elements, it is likely that the clinical behavior will also be a combination of these two elements. The average age at diagnosis and clinical presentation of metaplastic carcinomas are similar to the more common invasive mammary carcinomas. In spite of the variety of histological patterns of metaplastic breast carcinomas, most patients present with a solitary, nontender palpable mass or a mammographic density. Metaplastic carcinomas may present as a density on mammography and a microlobulated mass on ultrasonography. Although microcalcifications are usually absent, large coarse calcifications have been associated with metaplastic tumors with osteosarcomatous elements. Complex echogenecity with solid and cystic components may be seen sonographically and high signal intensity on T2-weighted magnetic resonance imaging (MRI). These image features are related to necrosis and cystic degeneration found histopathologically.24 – 26 Because of its heterogeneous histopathological pattern, the use of fine needle aspiration to diagnose metaplastic carcinoma has been considered limited with a relatively high false-negative rate.27 The accuracy of preoperative diagnosis by this method is reported to be approximately 50%. The final diagnoses of cases initially misdiagnosed as metaplastic carcinomas using cytology have included both benign and malignant conditions. The carcinomatous components are usually easily identified, but the squamous and/or sarcomatous components are generally less obvious.28 The use of core needle biopsy and mammotome are also limited in the diagnosis of metaplastic carcinomas, and the final diagnosis should be made after complete excision and adequate histopathologic examination of the entire tumor.29
PROGNOSIS
Figure 3 Matrix-producing metaplastic carcinoma presenting groups of epithelial cells within chondromyxoid stroma. H&E, ×200.
Metaplastic carcinomas represent a heterogeneous and rare group of tumors whose prognosis is unclear since much of the data is derived from small series of cases.2,5,8,25 The best predictor for the clinical behavior of metaplastic carcinoma is the phenotype. This approach is helpful in tumors with pure mesenchymal appearance. However, metaplastic carcinoma may contain significant components of both – invasive mammary carcinoma and mesenchymal neoplasm. In this setting, one might predict the tumor to behave with the combined
184
BREAST CANCER
potential of each of the components taken separately.1 The clinical behavior of the low-grade fibromatosis-like metaplastic tumors appears to be that of fibromatosis with a tendency for local recurrence and low or no metastatic potential.6,9 The low-grade adenosquamous carcinomas also have an excellent prognosis, and a small number of cases can behave in a locally aggressive manner. Recurrence appears to be related to the adequacy of local excision.11 Lymph node metastases in these tumors are extremely rare. The higher-grade biphasic sarcomatous carcinomas tend to behave more like sarcomas, with a potential for both local recurrence and distant metastases especially by hematogenous spread to the lung.30 In general, the incidence of lymph node metastases in metaplastic carcinoma is much lower when compared with invasive ductal carcinoma of a similar size, ranging between 5 and 26%. The likelihood of lymph node involvement with metaplastic carcinoma may be a reflection of the amount and grade of epithelial elements.3 Low-grade lesions such as matrix-producing metaplastic carcinoma and low-grade adenosquamous carcinoma appear to have an extremely low incidence of metastases to regional lymph nodes.3,20 Tumors that consist predominantly of epithelial elements tend to behave like carcinomas and may have lymph node involvement by the epithelial elements. In contrast, the mesenchymal elements of sarcomatous metaplasia of the breast typically do not metastasize to the regional lymph node. Tumors that consist predominantly of high-grade mesenchymal elements tend to behave like pure sarcomas and spread hematogenously to distant sites.14 Although some series have found no correlation between lymph node involvement and prognosis, this is likely because of the small number of patients involved in these studies. The pattern of progression is usually that of local recurrence followed by metastases to the lungs followed by spread to other anatomic sites. However, patients with low-grade tumors, such as fibromatosis-like and adenosquamous, usually present with local recurrences, but no distant metastases.9,20,31 Recurrence appears to be related to the extent of initial surgery. Patients undergoing excisional biopsy have higher rates of local recurrence than patients undergoing mastectomy. The increased recurrence rate after excisional biopsy alone for the low-grade tumors did not translate into increased mortality.6,31 Some authors reported a correlation between the size of the primary metaplastic tumor in the breast and the recurrence rates, as well as survival rates. Tumors that are larger in size are more likely to recur and patients are more likely to die of disease.3 In general, 5-year survival rates of patients with metaplastic carcinoma made up predominately of epithelial elements are reported to be around 65%, comparable to ductal carcinoma of similar size.2,3 The presence of squamous metaplasia appears to have no effect on the clinical behavior of the breast lesion with which it is associated. Patients with biphasic sarcomatoid carcinomas appear to have a very poor prognosis and lower survival rate, with 5-year survival of around 40%.2 However, Wargotz et al. reported a 5-year survival of 64% for patients with the “matrix-producing” subtype of metaplastic carcinoma.12
TREATMENT Metaplastic carcinoma is a very rare and heterogeneous disease. Although there are several series that have described the pathologic findings of this disease, these consist of generally fewer than 25 patients accrued over many years and treated with many different modalities. These retrospective series generally describe mastectomy as the surgery most commonly performed for these patients. Although some of these small retrospective studies have remarked on the use of radiation, they have not distinguished between radiation used as a component of breast conservation and postmastectomy radiation.32 In a retrospective analysis from MD Anderson Cancer Center (MDACC), 100 sarcomatoid breast cancer patients have been identified. All but one had surgery and 54% have had radiation as a component of treatment. Although 50% of first recurrences were local, the abstract does not report whether this was more common in patients who did not receive postoperative radiation.33 Given the lack of data on the role of radiation in the treatment of metaplastic carcinoma, we have generally applied the same criteria as that used in the treatment of soft tissue sarcomas in other sites. If a lumpectomy with negative margins (at least 2 mm) can be achieved, postlumpectomy radiation to a total dose of 6000 cGy is utilized. Radiation therapy is used to maximize local control without sacrificing cosmetic results. If the tumor size or breast : tumor ratio precludes breast conservation, mastectomy would be recommended. If the tumor can be removed with wide surgical margins (≥2 cm), postmastectomy radiation is not to be recommended. If the margins are close, postmastectomy radiation to a total dose of 6000 cGy can be recommended. We treat the regional nodes only if four or more lymph nodes are involved at the time of axillary dissection. Adjuvant systemic therapy has not been routinely administered for metaplastic carcinoma. The few reported series in which chemotherapy, radiation therapy, or the combination was administered in the adjuvant setting cover an admixture of tumors, including those with both predominant epithelial and mesenchymal components.7,21,22,34,35 Of the 28 patients with primary metaplastic carcinoma, reviewed pathologically at the Mayo Clinic, 13 received adjuvant systemic therapy.36 These mostly consisted of “standard” anthracycline-containing regimens and tamoxifen for women with ER and/or PR positive tumors. Since no difference was observed in the rate or recurrence between patients who did and those who did not receive adjuvant therapy, they suggest that the use of “standard” regimens for adenocarcinoma of the breast may be relatively ineffective for metaplastic breast cancer. Conversely, in an MDACC review series,32 of the 50 patients identified with metaplastic cancer of the breast, the addition of systemic chemotherapy with anthracycline-containing regimens to mastectomy (but not to wide local excision) yielded statistically significant better recurrence rates, particularly in stage II disease. In a more recent retrospective review of a 100 patients from MDACC with sarcomatoid/metaplastic carcinoma of the breast33 no differences were observed in the 5-year relapse-free survival between patients who did and those who did not
METAPLASTIC BREAST CARCINOMA
receive adjuvant chemotherapy, although a total of three patients treated with anthracycline/ifosfamide-based regimen did remain recurrence-free. Metaplastic carcinomas are rarely estrogen and/or progesterone positive.7 Nevertheless, in those patients who are hormone positive, we would recommend the use of hormonal therapy using the same guidelines as that used for the more common infiltrating ductal carcinomas, despite the seeming lack of efficacy in this setting. Although metaplastic carcinomas are rarely Her-2 positive, they often express Her-1. This raises the intriguing possibility of exploring treatments with EGFR tyrosine kinase inhibitors in select patients in the context of clinical trials.37 Once distant disease develops, patients uniformly do poorly. The vast majority of patients die from their disease. Due to the lack of data on the management of metastatic disease, it is impossible to make comments on the most appropriate agents to choose, but once again it seems reasonable to use those agents found to be most active in ductal carcinoma if the largest component is adenocarcinoma or agents active in sarcomas if the sarcomatoid features are most prevalent in the metastatic lesion.
AUTHORS’ RECOMMENDATIONS Metaplastic carcinoma is a rare disease with heterogeneous pathologic features. Because of its rarity, treatment decisions are made primarily based on retrospective data. The prognosis is variable depending on the degree of mesenchymal component in the tumor. Five-year survivals range from 40 to 65%. Following histological confirmation of the diagnosis, surgical options should be discussed with the patient. Breast conservation should be offered to patients who can obtain a margin-negative excision. Adjuvant therapy with anthracycline-containing regimens should be discussed with patients who have a component of adenocarcinoma or when lymph nodes are involved. Tamoxifen or aromatase inhibitors should be considered in patients with ER and/or PR positive tumors. Although the optimal agents in the metastatic setting are unknown, we would advise the use of agents that are used in the more usual infiltrating ductal adenocarcinomas.
REFERENCES 1. Page DL, Anderson TJ. Uncommon types of invasive carcinoma. In Page DL, Anderson TJ (eds) Diagnostic Histopathology of the Breast. Edinburgh, Scotland: Churchill Livingstone, 1987: 236 – 252. 2. Pinder SE, Elston CW, Ellis IO. Invasive carcinoma-unusual histological types. In Elston CW, Ellis IO (eds) The Breast. Edinburgh, Scotland: Churchill Livingstone, 1998: 283 – 337. 3. Ellis IO, et al. Invasive breast carcinoma. In Tavassoli F, Devilee P (eds) WHO Classification of Tumours: Pathology and Genetics of Tumours of the Breast and Female Genital Organs. Lyon, France: IARC Press, 2003: 13 – 59. 4. Leibl S, et al. Metaplastic breast carcinomas: are they of myoepithelial differentiation?: immunohistochemical profile of the sarcomatoid subtype using novel myoepithelial markers. Am J Surg Pathol 2005; 29: 347 – 53. 5. Gersell DJ, Katzenstein AL. Spindle cell carcinoma of the breast. A clinicopathologic and ultrastructural study. Hum Pathol 1981; 12: 550 – 61.
185
6. Gobbi H, et al. Metaplastic breast tumors with a dominant fibromatosislike phenotype have a high risk of local recurrence. Cancer 1999; 85: 2170 – 82. 7. Wargotz ES, Deos PH, Norris HJ. Metaplastic carcinomas of the breast. II. Spindle cell carcinoma. Hum Pathol 1989; 20: 732 – 40. 8. Denley H, et al. Metaplastic carcinoma of the breast arising within complex sclerosing lesion: a report of five cases. Histopathology 2000; 36: 203 – 9. 9. Gobbi H, et al. Metaplastic spindle cell breast tumors arising within papillomas, complex sclerosing lesions, and nipple adenomas. Mod Pathol 2003; 16: 893 – 901. 10. Gobbi H, et al. Reactive spindle cell nodules of the breast after core biopsy or fine-needle aspiration. Am J Clin Pathol 2000; 113: 288 – 94. 11. Rosen PP, Oberman HA. Invasive carcinoma. In Tumors of the Mammary Gland. Washington, District of Columbia: Armed Forces Institute of Pathology, 1993: 157 – 257. 12. Wargotz ES, Norris HJ. Metaplastic carcinomas of the breast. I. Matrixproducing carcinoma. Hum Pathol 1989; 20: 628 – 35. 13. Wargotz ES, Norris HJ. Metaplastic carcinomas of the breast. III. Carcinosarcoma. Cancer 1989; 64: 1490 – 9. 14. Borowsky A, Gobbi H. Metaplastic carcinoma of the breast: grading and behavior of predominantly spindle cell tumors. Pathol Case Rev 1999; 4: 208 – 13. 15. Adem C, et al. Wide spectrum screening keratin as a marker of metaplastic spindle cell carcinoma of the breast: an immunohistochemical study of 24 patients. Histopathology 2002; 40: 556 – 62. 16. Kurian KM, Al-Nafussi A. Sarcomatoid/metaplastic carcinoma of the breast: a clinicopathological study of 12 cases. Histopathology 2002; 40: 58 – 64. 17. Koker MM, Kleer CG. p63 expression in breast cancer: a highly sensitive and specific marker of metaplastic carcinoma. Am J Surg Pathol 2004; 28: 1506 – 12. 18. Reis-Filho JS, et al. Novel and classic myoepithelial/stem cell markers in metaplastic carcinomas of the breast. Appl Immunohistochem Mol Morphol 2003; 11: 1 – 8. 19. Barnes PJ, et al. Metaplastic breast carcinoma: clinical-pathologic characteristics and HER2/neu expression. Breast Cancer Res Treat 2005; 91: 173 – 8. 20. Rosen PP, Ernsberger D. Low-grade adenosquamous carcinoma. A variant of metaplastic mammary carcinoma. Am J Surg Pathol 1987; 11: 351 – 8. 21. Wargotz ES, Norris HJ. Metaplastic carcinomas of the breast. IV. Squamous cell carcinoma of ductal origin. Cancer 1990; 65: 272 – 6. 22. Kaufman MW, et al. Carcinoma of the breast with pseudosarcomatous metaplasia. Cancer 1984; 53: 1908 – 17. 23. Popnikolov NK, et al. Benign myoepithelial tumors of the breast have immunophenotypic characteristics similar to metaplastic matrixproducing and spindle cell carcinomas. Am J Clin Pathol 2003; 120: 161 – 7. 24. Chang YW, et al. Magnetic resonance imaging of metaplastic carcinoma of the breast: sonographic and pathologic correlation. Acta Radiol 2004; 45: 18 – 22. 25. Gunhan-Bilgen I, et al. Metaplastic carcinoma of the breast: clinical, mammographic, and sonographic findings with histopathologic correlation. AJR Am J Roentgenol 2002; 178: 1421 – 5. 26. Velasco M, et al. MRI of metaplastic carcinoma of the breast. AJR Am J Roentgenol 2005; 184: 1274 – 8. 27. Ribeiro-Silva A, et al. Limitations of fine-needle aspiration cytology to diagnose metaplastic carcinoma of the breast. Pathol Oncol Res 2001; 7: 298 – 300. 28. Johnson TL, Kini SR. Metaplastic breast carcinoma: a cytohistologic and clinical study of 10 cases. Diagn Cytopathol 1996; 14: 226 – 32. 29. Hoda SA, Rosen PP. Observations on the pathologic diagnosis of selected unusual lesions in needle core biopsies of breast. Breast J 2004; 10: 522 – 7. 30. Pitts WC, et al. Carcinomas with metaplasia and sarcomas of the breast. Am J Clin Pathol 1991; 95: 623 – 32. 31. Sneige N, et al. Low-grade (fibromatosis-like) spindle cell carcinoma of the breast. Am J Surg Pathol 2001; 25: 1009 – 16. 32. Gutman H, et al. Biologic distinctions and therapeutic implications of sarcomatoid metaplasia of epithelial carcinoma of the breast. J Am Coll Surg 1995; 180: 193 – 9.
186
BREAST CANCER
33. Hennessy B, et al. Sarcomatoid (metaplastic) carcinoma of the breast: the U.T. M. D. Anderson Cancer Center (MDACC) and SEER database experience. In Proceedings American Society of Clinical Oncology, Orlando, Florida, 2005. 34. Lazarevic B, Katatikarn V, Marks RA. Primary squamous-cell carcinoma of the breast. Diagnosis by fine needle aspiration cytology. Acta Cytol 1984; 28: 321 – 4. 35. Toikkanen S. Primary squamous cell carcinoma of the breast. Cancer 1981; 48: 1629 – 32.
36. Rayson D, et al. Metaplastic breast cancer: prognosis and response to systemic therapy. Ann Oncol 1999; 10: 413 – 9. 37. Leibl S, Moinfar F. Metaplastic breast carcinomas are negative for Her2 but frequently express EGFR (Her-1): potential relevance to adjuvant treatment with EGFR tyrosine kinase inhibitors? J Clin Pathol 2005; 58: 700 – 4.
Section 4 : Breast Cancer
14
Adenoid Cystic Carcinoma of the Breast
Melinda E. Sanders, Masako Kasami, Julie Means-Powell and David L. Page
INTRODUCTION Adenoid cystic carcinoma (ACC) of the breast, when properly and rigidly defined, accounts for about 0.1% of all invasive breast malignancies. Strict criteria were identified by the 1970s, aided by the use of electron microscopic features and histochemistry that recognized the orderly arrangement of dichotomous features.1 Carcinomas of this type and of similar histopathology have been described in the literature by a variety of terms including carcinoma adenoides cysticum, adenocystic carcinoma, pseudoadenomatous basal cell carcinoma, adenocystic basal cell cancer, basaloma, adenomyoepithelioma, cylindromatous carcinoma, and adenocystic basaloid carcinoma.1,2 Unfortunately, there is no guarantee that these related terms will identify cases with special attributes. Adenocystic has been used generically for tumors of similar patterns, probably including what we now call “invasive cribriform carcinoma”.1 However, the term ACC of the breast, which is credited to Spies,3 is the appropriate specific term for this special tumor type.
BIOLOGY AND EPIDEMIOLOGY Carcinomas termed “ACC” occur in several sites, including the major and minor salivary glands, lacrimal glands, maxillary sinuses, nasopharynx, external ear, esophagus, skin, tongue, trachea, bronchi, uterus, cervix, and Bartholin’s glands.2,4 – 6 ACC of the breast is so named because of its similarity microscopically to tumors of the major and minor salivary glands of the same name. Despite its rarity in the breast, it is important to be aware of the unique, defining elements of mammary ACC because its clinical behavior is distinctly different compared to ACC elsewhere, as well as similar appearing cancers of the breast. A recent important observation associates the origin of some ACC from examples of microglandular adenosis (MGA).7 Superficially, the histologic pattern described by ACC is similar to that of invasive cribriform carcinomas.8 – 10 Their differences have been described by electron microscopy,
histochemistry, and immunohistochemistry as well as by standard histology. Some of the special features of invasive cribriform carcinoma may be referable to ACC, particularly their increased representation in prevalence screening mammography, presumably because of their slow growth and lack of significant metastatic activity.11 – 14 ACC of the breast, when properly and rigidly defined, accounts for about 0.1% of all invasive breast malignancies.1 Age at diagnosis is varied, with a range from 31 to 90 years. Similar to those with other slower-growing mammary carcinomas, patients present with a movable breast mass, which may be present for days to years before diagnosis.5,6 The size of these tumors at diagnosis may be large, but define a range similar to invasive breast cancers in general. Current cases are diagnosed by mammographic imaging and other methods of detection at an average size of approximately 2 cm. However, in circumstances in which the breast mass was neglected, tumors as large as 10 cm have been reported.4,15 Unlike typical mammary carcinoma, the breast mass may be painful or tender.5,6,16,17 A significant feature, particularly with regard to complete excision, is the relatively frequent presence of a diffuse pattern of infiltration peripherally that is not apparent on gross examination, associated with a dominant, more compact centrally located tumor mass. There is no predilection for either the left or the right breast,5,18 and no bilateral lesions have been reported. ACC has a tendency to develop in the region of the nipple or areola.2,4,19 Fixation to skin, nipple, or pectoral muscles,5,17,20 however, is uncommon and bloody discharge is rare.16 As with typical mammary carcinoma, ACC in males is rare,20 – 24 but when present, it generally presents and behaves similarly to ACC in females.
PATHOLOGY Breast Cancer Classification Among the several carefully defined special types of breast cancer, ACC is a prime example of the necessity of a
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
188
BREAST CANCER
cluster of pathologic features that guarantee a unique, usually excellent clinical behavior and prognosis.25 – 27 Unlike other special types of invasive breast cancer,11,12 ACC may not be enriched in screening settings, although this phenomenon may not be discernable because of the rarity of ACC. In general, the special types of breast cancer are associated with an improved overall survival, as is readily identified by their greatly increased representation in 20-year survivors of invasive breast cancer.28
Histopathology of Adenoid Cystic Carcinoma of the Breast Breast ACC was first proposed as an entity because of its histopathologic similarity to the tumors of the same name in the salivary glands. A dual cell population composed of cytokeratin-positive epithelial cells and vimentin-positive myoepithelial cells present in well-formed islands of fairly cohesive cells forming small, sharply defined cystic spaces within these islands are the hallmarks of this tumor (see Figures 1 and 2). Nodular and diffusely infiltrating smaller, usually rounded collections of infiltrating tumor may coexist, but maintenance of the same cytology and intercellular arrangements is a defining feature. The typical cribriform morphology consists of glandular and pseudoglandular structures (see Figure 3), with true glands filled with brightly eosinophilic mucin and pseudolumens filled with more basophilic basement membrane-like material. In some cases, the basement membrane-like material predominates, forming irregular islands (see Figure 4). On a periodic acid Schiff stain, the contents of the “true” lumens stain pink as is typical of epithelial mucins, which are of near neutral pH whereas the basement membrane material or “stromal mucin” in the “pseudolumens” is more acidic and stains pale pink or blue. (see Figure 5).1,29 The main proliferating element is a population of modified myoepithelial cells as demonstrated by positive staining for smooth muscle actins, vimentin,30 and p6331 (see Figure 6). These cells are grouped in nests outlining the “pseudolumens” or
Figure 1 Low power view of adenoid cystic carcinoma. Notice the sharp borders of the tumor islands within the surrounding stroma and the crisp definition of rounded spaces in each infiltrating island.
Figure 2 Low power view of infiltrating adenoid cystic carcinoma with larger island pattern than in Figure 1 and some stromal reaction.
Figure 3 Higher power view of “true” lumens filled with eosinophilic mucin and “pseudolumens” filled with basement membrane-like material.
more irregular islands of basement membrane like material. The true epithelial lumens, also referred to as the “glandular” or “adenoid” component, are lined by cytokeratin and epithelial membrane antigen positive cells and demonstrate conserved basolateral markers of normal epithelial polarity including fodrin, E-cadherin, and β-catenin.30 Electron microscopy can also be used to demonstrate these two defining patterns.2,15,29 All strictly defined ACC reported in the literature are estrogen receptor (ER) and progesterone receptor (PR) negative.32 – 36 However, these findings should not be used as an indicator of poor outcome. In fact, the detailed study of markers of epithelial polarity in ACC by Kasami et al.30 demonstrating a status of advanced normal differentiation despite the capacity for local invasion is consistent with the usual lack of distant metastasis. Although not often mentioned in the literature, in situ patterns may be occasionally present in conjunction with areas
ADENOID CYSTIC CARCINOMA OF THE BREAST
Figure 4 Example of adenoid cystic carcinoma with basement membrane-like material predominating and surrounded by largely myoepithelial cells.
Figure 5 Periodic acid Schiff stain of adenoid cystic carcinoma stains the mucin in the few “true” lumena bright pink while the predominating basement membrane-like material stains pale pink.
of invasive tumor,37 although there is no recognized adenoid cystic form of in situ carcinoma without attendant invasion. We have identified several cases of ACC with in situ-like areas involving papillomas but also with classic ACC present in the surrounding soft tissue. Single case reports of ACC arising in a fibroadenoma38 and adenomyoepithelioma39 have been described; however, the ACC-like areas were apparently focal and confined to these otherwise benign lesions without evidence of classic ACC in the surrounding breast tissue. Shin and Rosen37 have recently described a solid variant of ACC that exhibits a >90% solid growth pattern composed of basaloid cells with moderate to occasionally marked nuclear atypia and rare to brisk mitotic activity. Ductules reminiscent of the intercalated ducts in salivary gland tumors identifiable by the presence of larger cells with eosinophilic cytoplasm
189
Figure 6 The myoepithelial stain p63 highlights the predominantly myoepithelial population in this adenoid cystic carcinoma while the epithelial cells forming the true lumena remain unstained.
and normochromatic nuclei arranged around the lumen were also present in tumor islands. These tumors show the same immunohistochemical profile of a typical ACC. Despite the presence of a single lymph node metastasis in two of the nine described patients, no patient developed recurrent or distant disease after initial treatment.37 Rarely, ACC may be intermixed with or appear to arise from, a benign, but infiltrative and equally rare condition, MGA.7 Altered myoepithelial cells appear to be the major neoplastic element in both ACC and “atypical MGA”, which shows recognizable features of typical MGA but with glandular proliferations of greater architectural complexity and cytologic atypia. In such cases, the single layer of lining epithelium typical of MGA is replaced by stratified cells with intraluminal bridging.40 – 42 In cases of MGA with coexistent ACC (see Figure 7), many of the haphazardly spaced glands of MGA contain the typical dual cell population and architectural features of ACC.7 There is an important mimicker of mammary ACC, usually easily recognized as bounded by the basement membrane of lobular units, consisting of aggregates of wellcircumscribed eosinophilic spherules. This benign condition has been termed “collagenous spherulosis” (see Figure 8). Although relatively uncommon, it is an extremely important pattern to recognize and is usually confined to a single lobular unit or a clustering of lobular units in an otherwise benign breast. The spaces are lined by basement membrane material as originally defined by Clement and colleagues43 using an antibody to type IV collagen (a basement membrane-related collagen).
Similarities between Adenoid Cystic and Invasive Cribriform Carcinoma Both of these special types of breast carcinoma are characterized by small sharply defined, rounded spaces within islands of epithelial cells. The first difference is that invasive cribriform carcinoma is named because it mimics quite
190
BREAST CANCER
Figure 7 Example of microglandular adenosis with coexistent adenoid cystic carcinoma showing an admixture of the simplified glands of microglandular adenosis lined by a single layer of epithelial cells with small nuclei and eosinophilic cytoplasm, more complex epithelial-lined glands with cribriform spaces or solid growth consistent with atypical MGA as well as glands expanded by myoepithelial cells and pseudolumena with basophilic basement membrane-like material.
Figure 8 Example of collagenous spherulosis showing regular, round spaces within an island of benign proliferating epithelium containing rigid, spherular deposits of eosinophilic hyaline material composed of laminin and type IV collagen.
precisely the patterns of noninvasive ductal carcinoma in situ.8 – 10 This pattern is only superficially similar to ACC and does not include the dual cell population with attendant special histochemical findings described above.
CLINICAL PRESENTATION AND PROGNOSIS In sharp contrast to similar tumors in other organs, ACC presenting in the breast is much less aggressive. In most instances, the tumor grows slowly,20,44,45 and has a low malignant potential.45 It is unusual for patients to have
metastases at the time of diagnosis; however, these tumors can recur both locally, likely as a result of inadequate excision, and rarely with limited patterns of metastatic disease many years after initial presentation. In fact, ACC of the breast has been reported to recur more than 20 years after diagnosis.44 Nevertheless, patients may live for many years following documented distant metastases,46,47 and most of the few reported deaths from mammary ACC are aberrant high-grade examples or lack clear case definition and histologic verification.23,24,32,48 – 53 As with other special types of breast cancer, it is mandatory to identify features of breast ACC that are predictive of excellent prognosis, and one is to demonstrate dual differentiation in the majority of the neoplasm. Ro and colleagues have presented a few cases attempting to further refine the prognostic ability of the histologic features.32 These investigators devised a grading system for breast ACC, dividing cases into three categories: grade I tumors were completely glandular and cystic, lacking a solid component, grade II tumors contained solid areas constituting less than 30% of the mass, and grade III tumors were those in which the solid component made up more than 30% of the mass. Unfortunately, this effort was hampered by the paucity of cases and lack of strict adherence to the dimorphic criteria. Indeed, the single grade III case in this study was not the dimorphic “special type” of breast cancer and should be regarded as a generic, high-grade breast cancer. This is very important because this woman who presented with positive axillary nodes at the time of diagnosis and died 2 years later with liver and brain metastases is often cited as evidence that ACC of the breast can act badly. In fact, this tumor was twice as large as any other tumor in the study (6.0 cm) and was >90% solid. Although demonstrated to be of myoepithelial origin by electron microscopy and positive for S-100, the photomicrographs show a tumor composed of very large, solid neoplastic islands containing cells of significantly higher grade than the usual ACC. This approach has also been specifically denied by the later important works of Shin and Kleer, which described solid variants of ACC that have few of the lumina but the same dual cell population and excellent prognosis.37,54 Kleer additionally found that nuclear grade and proliferative activity did not predict capacity for local recurrence or distant metastasis.37,54 In contrast to the usual breast cancer histologies, in breast ACC the routine sampling of axillary lymph nodes at the time of initial surgery has not been considered useful in providing prognostic information nor has this procedure contributed to local control. In fact, regional lymph node involvement is quite rare. After a review of over 200 cases in the literature, we were only able to find six histologically confirmed cases of breast ACC involving axillary lymph nodes.15,24,32,34,37 Controversy surrounds the diagnosis of one of these cases24 because of the presence of psammoma bodies and necrosis. These features are not typical of breast ACC, calling into question the accuracy of the diagnosis.18,50,55 At least one group suggests that this case in fact represents a variant of papillary and cribriform carcinoma.18 Of the remaining five accepted cases, the tumors were large, ranging from 5 to 15 cm.15,32,34,37 Two of these patients had clinically
ADENOID CYSTIC CARCINOMA OF THE BREAST
positive lymph nodes at presentation15,32,34 and two were from patients with solid variant tumors described by Shin but did not show evidence of local or distant recurrence on follow-up.37
Local Recurrence and Metastatic Potential Local recurrence after primary surgical therapy is an infrequent occurrence. We were able to confirm 10 cases of local recurrence from ACC,2,5,32,47,56 – 58 nine of which had undergone local excision. The remaining patient had a simple mastectomy. The margins from the initial resection of these tumors were not well documented. Therefore, considering that local recurrence is such an infrequent occurrence, it may be that inadequate attention was given to the margins at the time of initial resection in cases of local recurrence. Furthermore, two cases of local recurrence were reported as grade II tumors by Ro et al. and are not consistent with pure ACC.32 These two tumors are described as behaving more aggressively, which may account for the development of local recurrence in these cases. Similar to local recurrence, metastasis from ACC is rare. In fact, we were able to confirm only seven cases in the literature since 1970.15,32,35,46,47,59 One additional case was presented by Peters and Wolff18 of a patient who developed lung and liver metastases by radiological studies, but no biopsy was performed to confirm the presence of metastases from ACC. When metastasis does occur, the lung15,46,47,59 is the most frequently observed site. In fact, there are no accepted cases of distant metastases in which pulmonary metastasis did not occur. Other sites of metastases include bone,32,35 kidney,35,46 and brain.47 Although data on distant metastases from breast ACC are appropriately sparse, they present a very different pattern than that seen in most metastasizing breast cancers. Patients typically developed a distant solitary metastasis, several years from initial presentation. However, one patient with a 10 cm mass present for 10–20 years had evidence of disease metastatic to the lung at the time of presentation.15 Despite developing distant metastases, only one patient had died of disease at the time of follow-up.32 Importantly, this particular patient did not have classic histology for breast ACC, which probably accounts for the tumor’s uncharacteristically aggressive behavior.
TREATMENT Management of Primary Tumors Historically, surgical treatment of the primary tumor has varied from excisional biopsy to radical mastectomy. In the literature, several authors have advocated simple mastectomy as the treatment of choice unless the tumors are large or there are clinically enlarged axillary lymph nodes.2,5,18,33,50 However, we believe the preferred initial treatment of breast ACC is local excision to negative margins since this is a relatively indolent disease. Furthermore, although the incidence of local recurrence increases with more conservative surgery (similar to other breast cancer histologies) distant metastasis following local failure is exceedingly rare. In addition,
191
patients who experience local failures can be salvaged with more extensive surgical procedures. Only one patient initially treated with local excision who developed local recurrence subsequently died from distant metastases.49 An additional patient for whom the type of initial treatment was unknown, but axillary nodes were negative, developed a metastasis to the clavicle at 3 years, metastases to the lung, kidney, scalp, and eye at 5 years and subsequently died at 6 years following her initial diagnosis.35 When excisional biopsy is performed as the primary initial surgical procedure, special attention must be given to the margins because these tumors grow in an infiltrative pattern. In the series presented by Ro et al., all 12 tumors showed focal infiltration at the border, with tumor extension into adipose tissue.32 Also, Peters and Wolff noted insidious extensions around a grossly obvious lesion, which would be impossible to detect clinically.18 We believe that the primary reason these tumors recur locally is inadequate resection of the tumor with histologically positive margins. Because axillary lymph node metastasis is unusual with breast ACC and of questionable prognostic significance, we agree with others who advocate performing axillary node dissection only in patients with large tumors or clinically involved axillary lymph nodes.4,5,50 There is little reliable data regarding adjuvant hormonal or radiation therapy and its effects on local recurrence and survival in this disease. In addition, there does not appear to be a proven means of predicting patients who may benefit from adjuvant systemic therapy,17 and chemotherapy is not advocated. Four recent studies purporting to examine outcome in ACC23,51,53 are all flawed by limited or no central pathologic review and no mention of pathologic criteria for accepting a case as ACC. Cases included were diagnosed from 1960–2000,23,53 1970–1988,51 or 1974–200252 during which time criteria for ACC have been refined considerably. In addition, two of these studies report 46%51 and 29%53 of cases as ER positive, a finding that argues that these tumors were not actually ACC but rather invasive cribriform carcinoma or other low to intermediate grade cancers with focal cribriform features. In the first,51 all women given hormonal therapy had ER positive tumors. Of 11 women with ER and PR negative tumors, six received no adjuvant therapy, three had an unknown adjuvant therapy status but all were alive and well with no evidence of disease at 1, 5, and 11 years after initial surgery, and two women received radiotherapy, one of whom was alive and well without evidence of recurrence at 5 years while the other experienced distant metastasis to an unspecified site 6 years following diagnosis but was alive and well 2 years later. In fact, the only reported death in this study51 was a woman with an ER positive tumor associated with distant metastatic disease at the time of diagnosis. In the second study,53 it is not mentioned whether adjuvant therapy was given following surgery and its effect on outcome. Until these reports, there was only one report of an ER positive1 ACC of the breast, and this even case has been questioned because there has not been any reported confirmation of this histologic diagnosis.50 Therefore, there are no definitively diagnosed cases in the literature using hormone therapy in the adjuvant setting. In
192
BREAST CANCER
one report, tamoxifen was given in the metastatic setting, but the tumor did not respond.52 Therefore, antiestrogen hormone therapy does not appear to have a role in this malignancy.
Management of Local Recurrence Local recurrence is commonly treated by repeat surgical excision. A variety of surgical approaches have been used, including wide local excision, simple mastectomy, modified radical mastectomy, radical mastectomy, and en bloc resection of the pectoralis muscle and axilla. The median time from initial diagnosis to local recurrence was 5 years, but occurred as early as 6 months24 and as late as 22 years.44 Only one patient who developed local recurrence subsequently died with distant metastases.49 Of interest, this patient did not have a pure ACC tumor by pathologic description. Therefore, the occurrence of local recurrence with typical ACC histology does not seem to be a harbinger of systemic disease.
Management of Distant Recurrence Several modalities have been used for the management of distant metastases, including surgical resection, chemotherapy, radiation therapy, and hormone therapy. However, the effectiveness of these modalities is difficult to comment on because of the paucity of patients involved. The presence of solitary, large distant metastases may respond best to surgical excision.46,47 This is exemplified by a woman who had a pulmonary metastasis successfully surgically resected 6 years after mastectomy and a subsequent renal metastasis successfully resected 12 years after mastectomy.
AUTHORS’ RECOMMENDATIONS We believe the preferred initial treatment of breast ACC is local excision to negative margins since this is a relatively indolent disease. Although the incidence of local recurrence increases with more conservative surgery (similar to other breast cancer histologies) distant metastasis following local failure is exceedingly rare. Patients who experience local recurrences can be salvaged with more extensive surgical procedures. We advocate performing axillary node dissection only in patients with large tumors or clinically involved axillary lymph nodes. Adjuvant chemotherapy or radiotherapy is not advocated. Antiestrogen hormone therapy does not appear to have a role in this malignancy. The effectiveness of surgical resection, chemotherapy, radiation therapy, and hormone therapy in the management of distant metastases is difficult to comment on because of the paucity of patients involved in the literature. The presence of solitary, large distant metastases may respond best to surgical excision because patients may live for many years following documented distant metastases.
REFERENCES 1. Azzopardi JG, Ahmed A, Millis RR. Problems in breast pathology. Major Probl Pathol 1979; 11: i – xvi, 1 – 466. 2. Qizilbash AH, Patterson MC, Oliveira KF. Adenoid cystic carcinoma of the breast. Light and electron microscopy and a brief review of the literature. Arch Pathol Lab Med 1977; 101(6): 302 – 6.
3. Spies J. Adenoid cystic carcinoma; generalized metastases in 3 cases of basal cell type. Arch Surg 1930; 21: 365 – 404. 4. Anthony PP, James PD. Adenoid cystic carcinoma of the breast: prevalence, diagnostic criteria, and histogenesis. J Clin Pathol 1975; 28(8): 647 – 55. 5. Cavanzo FJ, Taylor HB. Adenoid cystic carcinoma of the breast. An analysis of 21 cases. Cancer 1969; 24(4): 740 – 5. 6. Galloway JR, Woolner LB, Clagett OT. Adenoid cystic carcinoma of the breast. Surg Gynecol Obstet 1966; 122(6): 1289 – 94. 7. Acs G, et al. Microglandular adenosis with transition into adenoid cystic carcinoma of the breast. Am J Surg Pathol 2003; 27(8): 1052 – 60. 8. Page DL, et al. Invasive cribriform carcinoma of the breast. Histopathology 1983; 7(4): 525 – 36. 9. Simpson JF, Page DL. Status of breast cancer prognostication based on histopathologic data. Am J Clin Pathol 1994; 102(4, Suppl. 1): S3 – 8. 10. Venable JG, Schwartz AM, Silverberg SG. Infiltrating cribriform carcinoma of the breast: a distinctive clinicopathologic entity. Hum Pathol 1990; 21(3): 333 – 8. 11. Anderson TJ, et al. Comparative pathology of breast cancer in a randomised trial of screening. Br J Cancer 1991; 64(1): 108 – 13. 12. Cowan WK, et al. The pathological and biological nature of screendetected breast carcinomas: a morphological and immunohistochemical study. J Pathol 1997; 182(1): 29 – 35. 13. Page DL. Prognosis and breast cancer. Recognition of lethal and favorable prognostic types. Am J Surg Pathol 1991; 15(4): 334 – 49. 14. Stutz JA, et al. The radiological appearances of invasive cribriform carcinoma of the breast. Nottingham Breast Team. Clin Radiol 1994; 49(10): 693 – 5. 15. Wells CA, Nicoll S, Ferguson DJ. Adenoid cystic carcinoma of the breast: a case with axillary lymph node metastasis. Histopathology 1986; 10(4): 415 – 24. 16. Jaworski RC, Kneale KL, Smith RC. Adenoid cystic carcinoma of the breast. Postgrad Med J 1983; 59(687): 48 – 51. 17. Leeming R, Jenkins M, Mendelsohn G. Adenoid cystic carcinoma of the breast. Arch Surg 1992; 127(2): 233 – 5. 18. Peters GN, Wolff M. Adenoid cystic carcinoma of the breast. Report of 11 new cases: review of the literature and discussion of biological behavior. Cancer 1983; 52(4): 680 – 6. 19. Friedman BA, Oberman HA. Adenoid cystic carcinoma of the breast. Am J Clin Pathol 1970; 54(1): 1 – 14. 20. Hjorth S, Magnusson PH, Blomquist P. Adenoid cystic carcinoma of the breast. Report of a case in a male and review of the literature. Acta Chir Scand 1977; 143(3): 155 – 8. 21. Ferlito A, Di Bonito L. Adenoid cystic carcinoma of the male breast: report of a case. Am Surg 1974; 40(1): 72 – 6. 22. Miliauskas JR, Leong AS. Adenoid cystic carcinoma in a juvenile male breast. Pathology 1991; 23(4): 298 – 301. 23. Millar BA, et al. The potential role of breast conservation surgery and adjuvant breast radiation for adenoid cystic carcinoma of the breast. Breast Cancer Res Treat 2004; 87(3): 225 – 32. 24. Verani RR, Van der Bel-Kahn J. Mammary adenoid cystic carcinoma with unusual features. Am J Clin Pathol 1973; 59(5): 653 – 8. 25. Page DL, Anderson TJ. How should we categorize breast cancer. Breast 1993; 2: 217 – 9. 26. Pereira H, et al. Pathological prognostic factors in breast cancer. IV: should you be a typer or a grader? A comparative study of two histological prognostic features in operable breast carcinoma. Histopathology 1995; 27(3): 219 – 26. 27. Simpson JF, Page DL. Prognostic value of histopathology in the breast. Semin Oncol 1992; 19(3): 254 – 62. 28. Dixon JM, et al. Long-term survivors after breast cancer. Br J Surg 1985; 72(6): 445 – 8. 29. Koss LG, Brannan CD, Ashikari R. Histologic and ultrastructural features of adenoid cystic carcinoma of the breast. Cancer 1970; 26(6): 1271 – 9. 30. Kasami M, et al. Maintenance of polarity and a dual cell population in adenoid cystic carcinoma of the breast: an immunohistochemical study. Histopathology 1998; 32(3): 232 – 8. 31. Mastropasqua MG, et al. Immunoreactivity for c-kit and p63 as an adjunct in the diagnosis of adenoid cystic carcinoma of the breast. Mod Pathol 2005; 18: 1277 – 82.
ADENOID CYSTIC CARCINOMA OF THE BREAST 32. Ro JY, Silva EG, Gallager HS. Adenoid cystic carcinoma of the breast. Hum Pathol 1987; 18(12): 1276 – 81. 33. Zaloudek C, Oertel YC, Orenstein JM. Adenoid cystic carcinoma of the breast. Am J Clin Pathol 1984; 81(3): 297 – 307. 34. Pastolero G, et al. Proliferative activity and p53 expression in adenoid cystic carcinoma of the breast. Mod Pathol 1996; 9(3): 215 – 9. 35. Trendell-Smith NJ, Peston D, Shousha S. Adenoid cystic carcinoma of the breast: a tumour commonly devoid of oestrogen receptors and related proteins. Histopathology 1999; 35(3): 241 – 8. 36. Sheen-Chen SM, et al. Adenoid cystic carcinoma of the breast: truly uncommon or easily overlooked? Anticancer Res 2005; 25(1B): 455 – 8. 37. Shin SJ, Rosen PP. Solid variant of mammary adenoid cystic carcinoma with basaloid features: a study of nine cases. Am J Surg Pathol 2002; 26(4): 413 – 20. 38. Blanco M, et al. Adenoid cystic carcinoma arising in a fibroadenoma. Ann Diagn Pathol 2005; 9(3): 157 – 9. 39. Van Dorpe J, De Pauw A, Moerman P. Adenoid cystic carcinoma arising in an adenomyoepithelioma of the breast. Virchows Arch 1998; 432(2): 119 – 22. 40. James BA, Cranor ML, Rosen PP. Carcinoma of the breast arising in microglandular adenosis. Am J Clin Pathol 1993; 100(5): 507 – 13. 41. Kay S. Microglandular adenosis of the female mammary gland: study of a case with ultrastructural observations. Hum Pathol 1985; 16(6): 637 – 41. 42. Rosenblum MK, Purrazzella R, Rosen PP. Is microglandular adenosis a precancerous disease? A study of carcinoma arising therein. Am J Surg Pathol 1986; 10(4): 237 – 45. 43. Clement PB, Young RH, Azzopardi JG. Collagenous spherulosis of the breast. Am J Surg Pathol 1987; 11(6): 411 – 7. 44. Lusted D. Structural and growth patterns of adenoid cystic carcinoma of breast. Am J Clin Pathol 1970; 54(3): 419 – 25. 45. Prioleau PG, et al. Sweat gland differentiation in mammary adenoid cystic carcinoma. Cancer 1979; 43(5): 1752 – 60.
193
46. Herzberg AJ, Bossen EH, Walther PJ. Adenoid cystic carcinoma of the breast metastatic to the kidney. A clinically symptomatic lesion requiring surgical management. Cancer 1991; 68(5): 1015 – 20. 47. Koller M, et al. Brain metastasis: a rare manifestation of adenoid cystic carcinoma of the breast. Surg Neurol 1986; 26(5): 470 – 2. 48. Nayer HR. Case report section; cylindroma of the breast with pulmonary metastases. Dis Chest 1957; 31(3): 324 – 7. 49. O’Kell RT. Adenoid cystic carcinoma of the breast. Mo Med 1964; 61: 855 – 8. 50. Sumpio BE, et al. Adenoid cystic carcinoma of the breast. Data from the Connecticut tumor registry and a review of the literature. Ann Surg 1987; 205(3): 295 – 301. 51. Arpino G, et al. Adenoid cystic carcinoma of the breast: molecular markers, treatment, and clinical outcome. Cancer 2002; 94(8): 2119 – 27. 52. Kontos M, Fentiman IS. Adenoid cystic carcinoma of the breast. Int J Clin Pract 2003; 57(8): 669 – 72. 53. McClenathan JH, de la Roza G. Adenoid cystic breast cancer. Am J Surg 2002; 183(6): 646 – 9. 54. Kleer CG, Oberman HA. Adenoid cystic carcinoma of the breast: value of histologic grading and proliferative activity. Am J Surg Pathol 1998; 22(5): 569 – 75. 55. Steinman A, Pepus M, McSwain G. Adenoid cystic carcinoma of the breast. South Med J 1978; 71(7): 851 – 4. 56. Peters GN, Wolff M, Haagensen CD. Tubular carcinoma of the breast. Clinical pathologic correlations based on 100 cases. Ann Surg 1981; 193(2): 138 – 49. 57. Wilson WB, Spell JP. Adenoid cystic carcinoma of breast: a case with recurrence and regional metastasis. Ann Surg 1967; 166(5): 861 – 4. 58. Woyke S, Domagala W, Olszewski W. Fine structure of mammary adenoid cystic carcinoma. Pol Med J 1970; 9(5): 1140 – 8. 59. Lim SK, Kovi J, Warner OG. Adenoid cystic carcinoma of breast with metastasis: a case report and review of the literature. J Natl Med Assoc 1979; 71(4): 329 – 30.
Section 4 : Breast Cancer
15
Non-Hodgkin Lymphoma of the Breast David S. Morgan and Jean F. Simpson
HISTORICAL BACKGROUND Non-Hodgkin’s lymphoma (NHL) localized to the breast is an uncommon disease. Most series and cases reported in the literature use a case definition devised by Wiseman and Liao,1 and modified by Hugh et al.2 which describes cases of “primary” breast lymphoma (PBL) as having the following characteristics: (i) both mammary tissue and lymphomatous infiltrate present in close association in an adequate histologic specimen and (ii) no evidence of widespread lymphoma by standard staging techniques or preceding extramammary lymphoma, although ipsilateral axillary node involvement is allowed if both lesions are present simultaneously. Considerably more common is “secondary” NHL of the breast, that is, NHL that involves the breast as one of the several sites of nodal and/or extranodal involvement at the first diagnosis or as a site of relapse. “Secondary” breast lymphoma is perhaps twice as common as PBL.3 Hodgkin’s disease involving the breast4 – 9 is very rare and will not be discussed here. The traditional case definition for PBL is somewhat at odds with the modern understanding that most NHLs are disseminated processes because of the potential mobility of lymphoma cells.10,11 It is not now customary to define lymphoma as “primary” or “secondary”. However, because of the specific diagnostic and therapeutic challenges it presents to clinicians, radiologists, and pathologists, this review will focus on localized, “primary” NHL of the breast. The diagnosis and management of so-called “secondary” cases is similar to that of other cases of NHL with one or more extranodal foci of involvement, and are not discussed here. The nomenclature, classification, and treatment of the diseases we now call NHLs have changed radically since they were first recognized in the 19th century, and therefore it is difficult to assess data from reports in the older literature. Lymphoid neoplasms of the breast have been recognized at least since 1880 when Gross removed an upper outer quadrant breast mass from a 22-year-old woman which was
said to show “lymphadenoid sarcoma”.12 A handful of cases were reported in the late 19th and early 20th centuries.13 – 15 In 1944, Adair and Hermann reported five cases of breast lymphosarcoma among 3033 cases of malignant breast tumors (0.16%) seen over a period of 20 years at Memorial Hospital.16 Since then, numerous reports, mostly case reports and small case series, as well as a few larger series, have described the clinical and pathologic features and discussed the management of this disease.2,3,5,17 – 37 Until recently, most reports, particularly in radiology literature, have discussed NHL of the breast with respect to its differences from, or similarities to carcinoma of the breast, or even as a subtype of breast cancer. This approach reflects the view and the concerns of the clinician who is expecting to diagnose, stage, and treat breast carcinoma using the usual paradigms and instead discovers lymphoma, raising the questions of how to differentiate the two. However this has unfortunately clouded the discussion of this entity in its most straightforward form, that is, as an extranodal NHL.
BIOLOGY AND EPIDEMIOLOGY Extranodal involvement in advanced stage lymphoma is quite common, and its incidence varies by histologic subtype, the most common sites being the gastrointestinal tract, the paranasal sinuses, and the respiratory tract. Breast involvement is not uncommon, and is said to be particularly common in Burkitt or other high-grade lymphomas.20,21 The discovery of NHL in an extranodal site as the sole (Ann Arbor stage IE38 ) focus of involvement is also common in clinical practice, found in perhaps one-third of NHL cases. It has been suggested that some extranodal lymphomas arise in mucosa-associated lymphoid tissue (MALT),39 which is found in a variety of sites, most notably the gastrointestinal tract, but also occasionally in the breast.40 MALT lymphomas are reported to have a distinctive morphology, immunohistochemical phenotype, and natural history (see below). The incidence of NHL with breast localization only (with or without clinically apparent involvement of the associated
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
NON-HODGKIN LYMPHOMA OF THE BREAST
axillary lymph nodes) is quite low. Reflecting the clinical problem of distinguishing this entity from carcinoma of the breast, many authors have reported its rate of occurrence as a percentage of all operable breast malignancies seen in single institution series. These historical reviews have placed the relative incidence at 0.05 to 0.18% of breast neoplasms2,26,29,33,41 – 44 although a few series put this figure as high as 1.1%.1,45,46 PBL is said to make up about 2.2% of all extranodal lymphomas47 and 0.7% of all NHLs in population-based studies.2 Most cases have been reported in women; fewer than 10 cases have been reported in men.27,29,37,48 – 51 Although most reported cases have been in persons of European or Asian27,50,52 descent, it is unclear whether this represents a higher incidence in those groups or reflects reporting bias. The reported age at diagnosis varies widely, from 9 to 95 years,27,33,49,53 with the greatest frequency in the sixth decade.2,5,24,26,29,33,54 Many investigators have reported a bimodal age distribution,16,27,33,55 with distinct clinical characteristics and outcomes in the two groups (see below sections under “Pathology” and “Treatment”).
PATHOLOGY Recent advances may improve our understanding of the clinicopathologic features of PBL. Acceptance of clinically relevant classification systems and the standard practice of immunophenotyping lymphoid neoplasms may result in more accurate comparison of published series of lymphomas. In the past, some studies of breast lymphomas may not have been carefully defined, and have included pseudolymphomas,3,41 granulocytic sarcomas,5 or even poorly differentiated carcinomas.5 Adopting more precise diagnostic approaches, as well as the use of defining criteria for PBL1,2 should result in greater understanding of these unusual neoplasms. The following review of histopathology of PBL is based on studies that have adhered to these tenets. PBL is a heterogeneous disease, even when limited to the defining criteria stated above.1,2 It does seem that there are two distinct clinicopathologic groups of patients, however. Women in the first group are younger and often have a bilateral presentation, often associated with pregnancy or recent childbirth, and the disease behaves aggressively. Classically, the histologic correlate of this type of PBL is high grade, small noncleaved cell, Burkitt-type lymphoma,56 although diffused large cell or immunoblastic lymphoma is also described5,37 in this setting. The second group of patients has an age range that overlaps that of primary breast carcinoma, and the presenting symptom is usually a unilateral breast mass. This group demonstrates the broad morphologic spectrum of malignant lymphoma, with the preponderance of cases being intermediate to high grade (see below).5 The same histologic subtypes found in node-based lymphomas are also present in breast lymphoma, and there does not appear to be any unique subtype occurring exclusively in the breast, although diffused large cell lymphoma is commonly reported.5,37,57,58
195
Some authors have suggested that PBLs are often MALTtype.3 MALT lymphoma is a low-grade lymphoma characterized by a proliferation of follicle marginal zone-type cells (centrocyte-like cells or monocytoid B cells), often with surrounding reactive germinal centers, plasmacytic differentiation, Dutcher bodies, follicular colonization, and lymphoepithelial lesions. These lymphomas typically express pan B cell markers. Although mammary tissue is not considered a normal site of MALT, it is possible that low-grade lymphomas similar to MALT lymphoma could arise in the breast in “acquired MALT”, as in the setting of infection or autoimmune disease.39 In support of the concept of MALT-type lymphomas of the breast, Mattia et al.3 have presented clinical and pathologic findings from nine cases of PBL. Of these, eight were of low-grade (Working Formulation and Kiel classification), and four were classified as MALT-type lymphomas (Table 1). No lymphoepithelial lesions, typically found in MALT-type lymphomas of more usual sites, were identified. In these cases, MALT-type lymphomas constituted a distinct subset with a markedly increased potential for prolonged diseasefree survival after local therapy (see below). Hugh et al.2 presented the clinical and pathologic features of 20 cases of PBL and found seven (35%) to be monocytoid B cell lymphomas, histologically identical to lymphomas arising in MALT. One histopathologic feature of lymphomas of MALT origin is the presence of lymphoepithelial lesions, characterized by malignant lymphoid infiltrates involving glandular epithelium.39 However, these lymphoepithelial lesions may be seen in PBL not otherwise characteristic of MALT lymphoma. In contrast, lymphoepithelial lesions were not identified in MALT-like lymphomas of the breast in the series presented by Mattia et al.3 Burke stresses site-specific differences that amount to MALT lymphomas, including those involving the breast.62 Other studies (see Table 1) have not found MALT-type lymphomas to be characteristic of PBLs,59,60 including the largest reported series37 of PBL (41 cases). In this study, derived largely from consultative cases, as well as others,27,36,60,61,63 intermediate and high-grade lymphomas predominate.
Table 1 Reported incidence of primary breast lymphoma presenting as MALT-type lymphoma.
Study Bobrow et al.59 Prevot et al.60 Mattia et al.3 Ariad et al.36 Lin et al.5 Hugh et al.2 Tan et al.61 Jeon et al.27 Arber et al.37 Kuper-Hommel et al.57 Farinha et al.31
Number of cases 9 14 9 6 21 20 14 7 41 38 14
Number (%) of MALT lymphomas 0 0 4 0 0 7 0 0 2 8 9
(44%)
(35%)
(5%) (20%) (64%)
196
BREAST CANCER
In summary, the histopathology of PBL is heterogeneous. Although the Burkitt-type PBL that affects younger women does appear to be a distinctive clinicopathologic entity, the majority of PBL have histopathologic characteristics that overlap with lymphomas in other sites, with both aggressive and relatively indolent forms represented. The rarity of PBL has been attributed to the paucity of normal lymphocytes in this location.40 Some investigators have suggested that PBL occurs in the setting of a preexisting breast lymphocytic lobulitis.63,64 The immunophenotype of the lymphocytes in lobulitis is predominantly B cells, with increased expression of HLA-DR by the lobular epithelium.65 – 67 In contrast, however, the lymphocytic infiltrate in PBL has been shown to be of T cell origin.59,60,68 And although lobular atrophy was a frequent finding in the study by Arber et al.,37 lobulitis in surrounding benign breast epithelium was not identified, nor was it seen in the series by Mattia et al.3
Immunohistochemical Analysis of Primary Breast Lymphoma A detailed comparison of the immunophenotype of reported PBLs is difficult, because reported series vary widely in reagents and methodology used. In spite of these inconsistencies, the vast majority (>90%) of PBLs are of B cell lineage by immunophenotyping studies.2,3,5,27,36,37,59 – 61,63 A few studies have also explored immunoglobulin heavy chain expression (Ig), with documentation of monoclonal Ig.2,3,41 Immunohistochemical evidence supporting MALT-type differentiation in PBL has not been studied in great detail. In the series presented by Mattia et al.,3 which described four PBL as being MALT-like (see above) while one case was further characterized as being CD5, CD10, and CD23 negative, an immunophenotype reported for MALT lymphoma.69 Arber et al.37 studied frozen sections of seven PBL with antibodies to mantle (UCL3D3) and marginal cells (UCL4D12). One of the seven cases, a monocytoid B cell lymphoma, was positive for both, while the remaining six cases representing other types of lymphomas were negative. Farinha et al. reported nine of 14 PBL as MALTomas, and stress the importance of immunohistochemistry in assessing the MALT characteristics of PBL.31 Thus, in the few examples reported, the immunophenotype of MALT-type lymphomas involving the breast is consistent with that histopathologic diagnosis.
Molecular and Cytogenetic Studies Reports of cytogenetic and molecular analysis of PBLs are infrequent. In a karyotypic study of MALT-type lymphomas, Ott et al.69 described the t(11;18)(q21;q21) chromosome translocation as a frequent and specific aberration in lowgrade MALT-type lymphomas. The single breast low-grade lymphoma included in this series did not contain clonal aberration, however. Bobrow et al.59 used polymerase chain reaction (PCR) to screen their cases of PBL for t(14 : 18), but none was positive for this translocation. This finding was not unexpected since less than 5% of extranodal lymphomas show this finding.70 Others have used the lack
of immunoglobulin heavy chain gene rearrangement as supportive evidence of T cell lineage in a case of PBL.71
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS Breast lymphoma is rarely suspected clinically. The demographic profile, presenting symptoms, and initial evaluation of patients generally point to a diagnosis of carcinoma of the breast or occasionally to a benign lesion, both of which are much more common than NHL of the breast. Therefore, the rate of clinical misdiagnosis is quite high, and the potential for inappropriate treatment is likewise high. The most common presentation is a painless, palpable mass which is often reported to have enlarged very rapidly, or a new lesion seen on screening mammogram.3,72 Reported sizes range from 1 to 19 cm, and multiple lesions may be seen. Less common is a presentation reminiscent of inflammatory breast cancer, with diffuse thickening or hardening of the breast, sometimes with an overlying violaceous skin color. Skin retraction, nipple retraction, and nipple discharge are not usually seen; Liberman et al. report that these findings were absent in all 32 patients in their series.73 The constitutional (“B”) symptoms of NHL (fever, night sweats, and weight loss) are unusual but occasionally seen with highgrade histologies. Many investigators have reported that involvement of the right breast is more common.17,24,29,74,75 although others have found no lateralization.3 Up to 25% are reported to be bilateral at presentation,2 with additional patients (9% in one series20 ) developing involvement of the contralateral breast later or as a relapse site.35,76 Involvement of the upper outer quadrant appears to be more common than involvement of other parts of the breast.42,49,77 The right predominance and the upper outer quadrant localization are unexplained, but neither seems to have prognostic or management implications. Recently, several groups have reported the occurrence of lymphoma in proximity to the capsule of silicone breast implants.78 – 80 The majority of cases are of T cell origin. Whether or not these lymphomas are causally related to the implants remains unknown. In summary, the clinical presentation is very similar to breast carcinoma, although bilaterality, rapid breast enlargement, absence of nipple discharge and skin retraction, multiple lesions, violaceous skin changes over the lesion, and soft and mobile axillary lymph nodes (as opposed to the harder lymph nodes typical of metastatic carcinoma) have been reported as features suggestive of breast lymphoma.5 Similarly, the radiographic findings are not characteristic; the mammogram and ultrasound findings are similar to those in other breast diseases, both malignant and benign.73 The most common mammogram finding is one or multiple focal masses, sometimes with a thin rim of radiolucency (a finding usually associated with nonmalignant lesions). Less common is diffusely increased breast density with skin thickening, which may be interpreted as consistent with inflammatory breast cancer.73,81 Spiculated masses and miliary densities have been described infrequently.82,83
NON-HODGKIN LYMPHOMA OF THE BREAST
Liberman reviewed 32 cases of breast NHL, 21 of them primary, and correlated mammogram and histologic findings. The majority of mammograms, 69%, showed a solitary uncalcified mass, 9% showed multiple uncalcified masses, 9% diffuse increased opacity, and 13% no abnormal findings. In the cases with mammographically detected masses, 28% were well circumscribed, while the rest were incompletely circumscribed. None of the masses was spiculated. The findings did not correlate with histologic subtypes.73 Thus no mammographic finding is specific for breast lymphoma. A small series by Schouten, which reported the prospective interpretation of mammograms, is perhaps illustrative: of seven mammograms, all were read as “abnormal,” two as “suggesting lymphoma,” and three as “malignancy”.20 The ultrasonographic findings are even less well described, and are reported as showing a wide spectrum of findings. In one retrospective series, eight patients had sonograms; seven of them showed a mass. All the masses were hypoechoic, with homogeneous echotexture in six.73 Magnetic resonance imaging (MRI) of breast NHL has been described.84 – 86 The findings on MRI are not pathognomonic and are similar to other breast neoplasms. As there is for all breast masses, there is a low threshold for biopsy. The most reliable diagnosis requires an incisional or excisional biopsy, and some authors have warned that diagnoses made by frozen section examination may be inaccurate in distinguishing lymphoma from breast carcinoma.46 While some authors have also cautioned against core needle biopsies,33 others have pointed out that immunoperoxidase studies performed on core samples or fine needle aspiration samples may facilitate making a diagnosis of breast lymphoma.82 Once a diagnosis of breast lymphoma is made, the investigation should proceed as for the staging of any other extranodal lymphoma, and should include the following: 1. Detailed history with attention to systemic symptoms of lymphoma (unexplained fever greater that 101◦ F, loss of more than 10% of body weight, and unexplained night sweats) 2. Computed tomography scans of the chest, abdomen, and pelvis 3. Bone marrow biopsy 4. Blood studies including a complete blood count with differential and platelets, chemistries including lactate dehydrogenase, and perhaps β-2 microglobulin. Nuclear medicine scans may be useful in staging NHL of the breast. Several reports describe positivity of breast lymphoma on Gallium-67. In many centers, [18 F] fluorodeoxyglucose (FDG) positron emission tomography (PET) is replacing Gallium scanning as a staging tool for aggressive lymphoma (low-grade lymphomas are variably FDG-avid). FDG-avidity of breast lymphomas has been reported, and PET is likely to be a useful tool for staging and follow-up of aggressive breast lymphoma.87
TREATMENT Determining definitive treatment recommendations for breast lymphoma from the reported case series is difficult. There
197
are no clinical trials, the case series are small and typically describe nonuniform treatments. The classification system for NHLs has evolved and the concept of MALT lymphoma has been explored. In general during the time period covered by these case series, common treatments for NHL have evolved from surgery to adjuvant radiotherapy to combination chemotherapy and chemoimmunotherapy. In general, early investigators favored surgical treatment,16 while later authors have advocated radiotherapy, chemotherapy, or both after excisional biopsy.20,24,26,33,36,42,44 Lamovec and Jancar, for example, advocated radiotherapy after excisional biopsy, based on their small series, in which those receiving radiotherapy after excisional biopsy had fewer recurrences than those who had excisional biopsy alone.26 Schouten et al. advocated aggressive chemotherapy based on their small series,20 and other authors have advocated chemotherapy when such factors as highgrade histology, high lactate dehydrogenase level, axillary node involvement, and bulky disease are present.24 More recently, some have advocated chemotherapy for all patients, with the type of chemotherapy tailored to the histologic subtype.17,18,22,23,25,28,35 In the absence of clear data to the contrary from the literature, it seems reasonable to base the treatment of PBL on the generally accepted recommendations for other extranodal lymphomas, where treatment decisions are based primarily on histologic subtype and extent of disease. PBLs, as defined for the purposes of this discussion (see above), are stage IE or IIE in the Ann Arbor staging system.38 Histologic subtypes may be grouped as low, intermediate or highgrade according to the Working Formulation classification of lymphoma.88 Low-grade lymphoma of MALT may represent another grouping important for management decisions. The standard therapy for apparently localized low-grade lymphoma is radiotherapy alone,89 with the caveat that dissemination of low-grade lymphoma is the rule and that many patients with apparently localized disease have occult disease elsewhere. For this reason, some clinicians would advocate single agent alkylator therapy such as chlorambucil or cyclophosphamide or a non–anthracycline-based combination such as cyclophosphamide, vincristine, and prednisone, rather than radiotherapy alone. A period of observation without treatment may be appropriate for selected patients with indolent histology and clinical behavior. Low-grade lymphomas of MALT in the gastrointestinal tract or salivary gland, where they are most common and best characterized, behave indolently.90 – 92 Although relatively few reports2,3,26 have addressed the question of whether breast low-grade MALT lymphomas are similarly indolent, there is evidence that this histologic subtype may represent a special case in which excision alone is adequate for some patients. For example, Mattia et al. observed four patients with low-grade lymphoma of MALT among nine cases of PBL who were treated with excision alone. Three were alive, with no evidence of disease, at 10, 12, and 48 months followup.3 In reviewing the literature, they identified 17 patients in whom this diagnosis is relatively secure, 10 of whom had a durable complete remission “most after local treatment only.” They concluded that there is a recognizable subset of patients
198
BREAST CANCER
with this histologic subtype, in the older literature, who had an indolent course with minimal treatment.3 Ample data now support the treatment of localized intermediate grade lymphomas with abbreviated chemotherapy (for example, three cycles of cyclophosphamide, doxorubicin, vincristine, and prednisone [CHOP], or the equivalent) followed by involved field radiotherapy.93 Most clinicians would add the anti-CD20 antibody rituximab to this regimen, based on the improvement in overall survival and disease-free survival seen in patients with advanced NHL.94 Although there is less consensus that this treatment plan is also acceptable in extranodal lymphomas, it seems reasonable to apply this approach to localized breast lymphomas as well. For those patients with bulky disease or other adverse prognostic features, some clinicians might consider longer courses of chemotherapy followed by radiotherapy. The optimal treatment regimen for extranodal high-grade lymphoma is not defined; however, it should be seen as a potentially curable disease and requires central nervous system (CNS) prophylaxis.
therapy with rituximab.3,17,26 Follicular lymphomas of the breast, like follicular lymphomas elsewhere, tend to relapse in disseminated sites, but survival times are fairly long, 5 years in one series. This compares to a median survival time usually reported as 8 to 12 years in node-based follicular lymphomas. The survival times for intermediate and high-grade histologic subtypes of PBL vary widely in the literature, probably reflecting the variety of treatment modalities used. With modern treatment with combination chemotherapy (for example, CHOP) or chemoimmunotherapy (CHOP and rituximab), intermediate grade breast lymphoma is curable,28 although some authors have postulated a worse survival rate than for other intermediate grade lymphomas. The International Prognostic Index has not been validated in this group of patients, but is likely to be of prognostic value.103 Of particular concern are reports17,22,30,35 that CNS relapse is particularly common with intermediate grade histology of breast lymphoma, raising the question of CNS prophylaxis.
AUTHORS’ RECOMMENDATIONS PROGNOSIS Like much of the data on PBL, the reported results of treatment vary widely. Five-year survival rates ranging from 995 to 85%33 are reported, probably reflecting the heterogeneous histologic subtypes, variations in staging, and nonstandardized treatment of patients in the various small series. In general, early reports described an aggressive, rapid course with early relapse. Lattes for example, reported 17 of 33 patients dead in the first year.95 It has been suggested that these poor outcome data reflect understaging of patients and thus inadequate treatment.20 As noted above, many authors have distinguished two patterns of clinical behavior of PBL. The first affects younger women, many of them pregnant or recently postpartum, and many of African or Italian descent, who have a bilateral presentation, Burkitt’s-like histologic subtype and a very rapid and aggressive clinical course.2,16,27,96 – 101 CNS relapses are reportedly common in this group.2 The other pattern is of older women with unilateral breast involvement. A variety of histologic subtypes is seen, and the clinical course is in general more indolent.2,102 Age, which probably discriminates these two groups, has been reported to be a significant prognostic factor. One series reported median survival times of 10.3 months in patients under 45 years old and 25.5 months in those older than 45,27 and, similarly, Hugh et al., extracting data on 235 patients described in the literature, reported median survival of 9.5 and 47 months for those younger than and older than 40, respectively.2 The differences in survival may be due to the higher incidence of Burkitt’s-like lymphoma in the younger group. In the more recent series, histologic subtype is reported as a major prognostic factor, as it is in other lymphomas.3,27,29,36,44 It is likely that the outcomes for patients with breast lymphoma are generally similar to other extranodal NHL patients. Low-grade lymphomas of MALT tend to have favorable outcomes with a variety of treatments, including radiotherapy, surgery, chemotherapy, or antibody
The rational treatment of localized breast lymphoma should be similar to that of other extranodal lymphomas. Although the data in the literature are difficult to interpret with confidence, one might expect outcomes with modern treatment similar to those of other extranodal lymphomas matched for histologic subtype and stage. Special note is made of the syndrome of bilateral aggressive tumors of childbearing age, for which the optimal treatment is undefined but may be more aggressive, of the reports of CNS relapse, and of the especially indolent nature of low-grade MALT lymphomas.
REFERENCES 1. Wiseman C, Liao KT. Primary lymphoma of the breast. Cancer 1972; 29(6): 1705 – 12. 2. Hugh JC, et al. Primary breast lymphoma. An immunohistologic study of 20 new cases. Cancer 1990; 66(12): 2602 – 11. 3. Mattia AR, Ferry JA, Harris NL. Breast lymphoma. A B-cell spectrum including the low grade B-cell lymphoma of mucosa associated lymphoid tissue. Am J Surg Pathol 1993; 17(6): 574 – 87. 4. Eufemio G. Case report. Primary malignant lymphoma of the breast. Acta Med Philipp 1966; 2(4): 201 – 5. 5. Lin Y, Govindan R, Hess JL. Malignant hematopoietic breast tumors. Am J Clin Pathol 1997; 107(2): 177 – 86. 6. Kueckens H. Ein lokales lymphogranulom der brust in form eines mammatumors. Beitr pathol Anat Allge Pathol 1928; 80: 135 – 7. 7. Raju GC, Jankey N, Delpech K. Localized primary extranodal Hodgkin’s disease (Hodgkin’s lymphoma) of the breast. J R Soc Med 1987; 80(4): 247 – 9. 8. Abuin JC, Gonzalez R. Mammary localization of Hodgkin’s disease. Rev Sanid Milit Argent 1968; 67(1): 45 – 8. 9. Lawler MR, Riddell DH. Jr. Hodgkin’s disease of the breast. Arch Surg 1966; 93(2): 331 – 4. 10. Horning SJ, et al. Detection of non-Hodgkin’s lymphoma in the peripheral blood by analysis of antigen receptor gene rearrangements: results of a prospective study. Blood 1990; 75(5): 1139 – 45. 11. Stetlet-Stevenson M, et al. Detection of occult follicular lymphoma by specific DNA amplification. Blood 1988; 72(5): 1822 – 5. 12. Gross S. Tumors of the Mammary Gland. New York: Appleton and Co., 1880. 13. Halsam W. Birmingh M. Rev 1889; 25: 286. 14. Geist S, Wilensky A. Sarcoma of the breast. Ann Surg 1915; 62: 11.
NON-HODGKIN LYMPHOMA OF THE BREAST 15. Bilroth I. Handbuch der Frauenkrankheiten. Stuttgart: F. Enke, 1880, Vol. 3. 16. Adair F, Hermann J. Primary lymphosarcoma of the breast. Surgery 1944; 16: 836. 17. Wong WW, et al. Primary non-Hodgkin lymphoma of the breast: the Mayo clinic experience. J Surg Oncol 2002; 80(1): 19 – 25; discussion 26. 18. Vigliotti ML, et al. Primary breast lymphoma: outcome of 7 patients and a review of the literature. Leuk Lymphoma 2005; 46(9): 1321 – 7. 19. Smith MR, Brustein S, Straus DJ. Localized non-Hodgkin’s lymphoma of the breast. Cancer 1987; 59(2): 351 – 4. 20. Schouten JT, Weese JL, Carbone PP. Lymphoma of the breast. Ann Surg 1981; 194(6): 749 – 53. 21. Sabate JM, et al. Lymphoma of the breast: clinical and radiologic features with pathologic correlation in 28 patients. Breast J 2002; 8(5): 294 – 304. 22. Ribrag V, et al. Primary breast lymphoma: a report of 20 cases. Br J Haematol 2001; 115(2): 253 – 6. 23. Park YH, et al. Primary malignant lymphoma of the breast: clinicopathological study of nine cases. Leuk Lymphoma 2004; 45(2): 327 – 30. 24. Misra A, Kapur BM, Rath GK. Primary breast lymphoma. J Surg Oncol 1991; 47(4): 265 – 70. 25. Lyons JA, et al. Treatment of prognosis of primary breast lymphoma: a review of 13 cases. Am J Clin Oncol 2000; 23(4): 334 – 6. 26. Lamovec J, Jancar J. Primary malignant lymphoma of the breast. Lymphoma of the mucosa-associated lymphoid tissue. Cancer 1987; 60(12): 3033 – 41. 27. Jeon HJ, et al. Primary non-Hodgkin malignant lymphoma of the breast. An immunohistochemical study of seven patients and literature review of 152 patients with breast lymphoma in Japan. Cancer 1992; 70(10): 2451 – 9. 28. Ha CS, et al. Localized primary non-Hodgkin lymphoma of the breast. Am J Clin Oncol 1998; 21(4): 376 – 80. 29. Giardini R, Piccolo C, Rilke F. Primary non-Hodgkin’s lymphomas of the female breast. Cancer 1992; 69(3): 725 – 35. 30. Gholam D, et al. Primary breast lymphoma. Leuk Lymphoma 2003; 44(7): 1173 – 8. 31. Farinha P, et al. High frequency of MALT lymphoma in a series of 14 cases of primary breast lymphoma. Appl Immunohistochem Mol Morphol 2002; 10(2): 115 – 20. 32. Domchek SM, et al. Lymphomas of the breast: primary and secondary involvement. Cancer 2002; 94(1): 6 – 13. 33. Dixon JM, et al. Primary lymphoma of the breast. Br J Surg 1987; 74(3): 214 – 6. 34. Cohen PL, Brooks JJ. Lymphomas of the breast. A clinicopathologic and immunohistochemical study of primary and secondary cases. Cancer 1991; 67(5): 1359 – 69. 35. Au WY, et al. Lymphoma of the breast in Hong Kong Chinese. Hematol Oncol 1997; 15(1): 33 – 8. 36. Ariad S, et al. Breast lymphoma. A clinical and pathological review and 10-year treatment results. S Afr Med J 1995; 85(2): 85 – 9. 37. Arber DA, et al. Non-Hodgkin’s lymphoma involving the breast. Am J Surg Pathol 1994; 18(3): 288 – 95. 38. Carbone PP, et al. Report of the committee on Hodgkin’s disease staging classification. Cancer Res 1971; 31(11): 1860 – 1. 39. Isaacson P, Wright DH. Extranodal malignant lymphoma arising from mucosa-associated lymphoid tissue. Cancer 1984; 53(11): 2515 – 24. 40. Ferguson DJ. Intraepithelial lymphocytes and macrophages in the normal breast. Virchows Arch A Pathol Anat Histopathol 1985; 407(4): 369 – 78. 41. Akbari CM, Welch JP, Pastuszak W. Primary lymphoproliferative disorders of the breast. Conn Med 1995; 59(11): 651 – 5. 42. DeCosse J, et al. Primary lymphosarcoma of the breast: a review of 14 cases. Cancer 1962; 15: 1264 – 8. 43. Jernstrom P, Sether J. Primary lymphosarcoma of the mammary gland. JAMA 1967; 201: 506. 44. Mambo NC, Burke JS, Butler JJ. Primary malignant lymphomas of the breast. Cancer 1977; 39(5): 2033 – 40. 45. Dao AH, Adkins RB Jr, Glick AD. Malignant lymphoma of the breast: a review of 13 cases. Am Surg 1992; 58(12): 792 – 6.
199
46. Telesinghe PU, Anthony PP. Primary lymphoma of the breast. Histopathology 1985; 9(3): 297 – 307. 47. Freeman C, Berg JW, Cutler SJ. Occurrence and prognosis of extranodal lymphomas. Cancer 1972; 29(1): 252 – 60. 48. de Souza LJ, Talvalkar GV, Morjaria JH. Primary malignant lymphoma of the breast. Indian J Cancer 1978; 15(4): 30 – 5. 49. Lawler MR Jr, Richie RE. Reticulum cell sarcoma of the breast. Cancer 1967; 20(9): 1438 – 46. 50. Murata T, et al. Primary non-Hodgkin malignant lymphoma of the male breast. Jpn J Clin Oncol 1996; 26(4): 243 – 7. 51. Tanino M, et al. Lymphosarcoma of the male breast. Breast 1984; 10: 13 – 5. 52. Tanaka T, et al. Primary malignant lymphoma of the breast. With a review of 73 cases among Japanese subjects. Acta Pathol Jpn 1984; 34(2): 361 – 73. 53. Pullen CM, Cass AJ. Bilateral primary lymphoma of the breast. Aust N Z J Surg 1996; 66(12): 845 – 7. 54. Cohen Y, et al. Primary breast lymphoma. Harefuah 1993; 125(1 – 2): 24 – 6, 63. 55. Latteri MA, et al. Primary extranodal non-Hodgkin lymphomas of the uterus and the breast: report of three cases. Eur J Surg Oncol 1995; 21(4): 432 – 4. 56. Poulsen LO, et al. Immunologic observations in close relatives of two sisters with mammary Burkitt’s lymphoma. Mammary Burkitt’s lymphoma in sisters. Cancer 1991; 68(5): 1031 – 4. 57. Kuper-Hommel MJ, et al. Treatment and survival of 38 female breast lymphomas: a population-based study with clinical and pathological reviews. Ann Hematol 2003; 82(7): 397 – 404. 58. Vignot S, et al. Non-Hodgkin’s lymphoma of the breast: a report of 19 cases and a review of the literature. Clin Lymphoma 2005; 6(1): 37 – 42. 59. Bobrow LG, et al. Breast lymphomas: a clinicopathologic review. Hum Pathol 1993; 24(3): 274 – 8. 60. Prevot S, et al. Primary non-Hodgkin’s malignant lymphoma of the breast. Anatomopathologic diagnosis of 14 cases. Bull Cancer 1990; 77(2): 123 – 36. 61. Tan PH, Sng IT. Breast lymphoma – a pathologic study of 14 cases. Ann Acad Med Singapore 1996; 25(6): 783 – 90. 62. Burke JS. Are there site-specific differences among the MALT lymphomas – morphologic, clinical? Am J Clin Pathol 1999; 111(1 Suppl. 1): S133 – 43. 63. Aozasa K, et al. Malignant lymphoma of the breast. Immunologic type and association with lymphocytic mastopathy. Am J Clin Pathol 1992; 97(5): 699 – 704. 64. Rooney N, et al. Primary breast lymphoma with skin involvement arising in lymphocytic lobulitis. Histopathology 1994; 24(1): 81 – 4. 65. Lammie GA, et al. Sclerosing lymphocytic lobulitis of the breast – evidence for an autoimmune pathogenesis. Histopathology 1991; 19(1): 13 – 20. 66. Schwartz IS, Strauchen JA. Lymphocytic mastopathy. An autoimmune disease of the breast? Am J Clin Pathol 1990; 93(6): 725 – 30. 67. Tomaszewski JE, et al. Diabetic mastopathy: a distinctive clinicopathologic entity. Hum Pathol 1992; 23(7): 780 – 6. 68. Giedsing Hansen T, et al. Primary non-Hodgkin’s lymphoma of the breast (PLB): a clinicopathological study of seven cases. Apmis 1992; 100(12): 1089 – 96. 69. Ott G, et al. The t(11;18)(q21;q21) chromosome translocation is a frequent and specific aberration in low-grade but not highgrade malignant non-Hodgkin’s lymphomas of the mucosa-associated lymphoid tissue (MALT-) type. Cancer Res 1997; 57(18): 3944 – 8. 70. Raghoebier S, et al. Essential differences in oncogene involvement between primary nodal and extranodal large cell lymphoma. Blood 1991; 78(10): 2680 – 5. 71. Anania G, et al. Primary non-Hodgkin’s T-cell lymphoma of the breast. Eur J Surg 1997; 163(8): 633 – 5. 72. Slanetz PJ, Whitman GJ. Non-Hodgkin’s lymphoma of the breast causing multiple vague densities on mammography. AJR Am J Roentgenol 1996; 167(2): 537 – 8. 73. Liberman L, et al. Non-Hodgkin lymphoma of the breast: imaging characteristics and correlation with histopathologic findings. Radiology 1994; 192(1): 157 – 60.
200
BREAST CANCER
74. el-Ghazawy IM, Singletary SE. Surgical management of primary lymphoma of the breast. Ann Surg 1991; 214(6): 724 – 6. 75. Eskelinen M, et al. Lymphoma of the breast. Ann Chir Gynaecol 1989; 78(2): 149 – 52. 76. Zinzani PL, et al. Bilateral primary breast lymphoma: a case of local recurrence. Leuk Lymphoma 2003; 44(4): 737 – 8. 77. Andre JM, et al. Malignant lymphomas and other hematosarcomas with initial breast localization. Retrospective study of 20 cases. Bull Cancer 1983; 70(5): 401 – 9. 78. Cook PD, et al. Follicular lymphoma adjacent to foreign body granulomatous inflammation and fibrosis surrounding silicone breast prosthesis. Am J Surg Pathol 1995; 19(6): 712 – 7. 79. Gaudet G, et al. Breast lymphoma associated with breast implants: two case-reports and a review of the literature. Leuk Lymphoma 2002; 43(1): 115 – 9. 80. Sahoo S, et al. Anaplastic large cell lymphoma arising in a silicone breast implant capsule: a case report and review of the literature. Arch Pathol Lab Med 2003; 127(3): e115 – 8. 81. Meyer JE, Kopans DB, Long JC. Mammographic appearance of malignant lymphoma of the breast. Radiology 1980; 135(3): 623 – 6. 82. Meyer JE, et al. Large-core breast biopsy to obtain tissue for tumor markers in breast lymphoma. AJR Am J Roentgenol 1994; 162(6): 1500. 83. Pameijer FA, et al. Non-Hodgkin’s lymphoma of the breast causing miliary densities on mammography. AJR Am J Roentgenol 1995; 164(3): 609 – 10. 84. Darnell A, et al. Primary lymphoma of the breast: MR imaging features. A case report. Magn Reson Imaging 1999; 17(3): 479 – 82. 85. Demirkazik FB. MR imaging features of breast lymphoma. Eur J Radiol 2002; 42(1): 62 – 4. 86. Naganawa S, et al. MR lmaging of the primary breast lymphoma: a case report. Breast Cancer 1996; 3(3): 209 – 13. 87. Bakheet SM, et al. F-18 FDG positron emission tomography in primary breast non-Hodgkin’s lymphoma. Clin Nucl Med 2001; 26(4): 299 – 301. 88. The Non-Hodgkin’s Lymphoma Pathologic Classification Project. National cancer institute sponsored study of classifications of nonHodgkin’s lymphomas: summary and description of a working formulation for clinical usage. Cancer 1982; 49(10): 2112 – 35.
89. Shipp M, Mauch P, Harris N. Non-Hodgkin’s lymphomas. In DeVita V, Hellman S, Rosenberg S (eds) Cancer: Principles and Practice of Oncology. Lippincott-Raven, 1997: 2165 – 2220. 90. Isaacson PG. Gastrointestinal lymphoma. Hum Pathol 1994; 25(10): 1020 – 9. 91. Radaszkiewicz T, Dragosics B, Bauer P. Gastrointestinal malignant lymphomas of the mucosa-associated lymphoid tissue: factors relevant to prognosis. Gastroenterology 1992; 102(5): 1628 – 38. 92. Wotherspoon AC, et al. Helicobacter pylori-associated gastritis and primary B-cell gastric lymphoma. Lancet 1991; 338(8776): 1175 – 6. 93. Miller TP, et al. Chemotherapy alone compared with chemotherapy plus radiotherapy for localized intermediate- and high-grade nonHodgkin’s lymphoma. N Engl J Med 1998; 339(1): 21 – 6. 94. Coiffier B, et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 2002; 346(4): 235 – 42. 95. Lattes R. Sarcomas of the breast. Int J Radiat Oncol Biol Phys 1978; 4(7 – 8): 705 – 8. 96. Carbone A, et al. Primary lymphoblastic lymphoma of the breast. Clin Oncol 1982; 8(4): 367 – 73. 97. Durodola JI. Burkitt’s lymphoma presenting during lactation. Int J Gynaecol Obstet 1976; 14(3): 225 – 31. 98. Jones DE, et al. Burkitt’s lymphoma: obstetric and gynecologic aspects. Obstet Gynecol 1980; 56(4): 533 – 6. 99. Kay S. Lymphosarcoma of the female mammary gland. AMA Arch Pathol 1955; 60(5): 575 – 9. 100. Shepherd JJ, Wright DH. Burkitt’s tumour presenting as bilateral swelling of the breast in women of child-bearing age. Br J Surg 1967; 54(9): 776 – 80. 101. Tweeddale DN, Mahr MM. Secondary lymphosarcoma of the breast in pregnancy. Report of a case. Obstet Gynecol 1964; 24: 584 – 6. 102. Brustein S, et al. Malignant lymphoma of the breast. A study of 53 patients. Ann Surg 1987; 205(2): 144 – 50. 103. The International Non-Hodgkin’s Lymphoma Prognostic Factors Project. A predictive model for aggressive non-Hodgkin’s lymphoma. N Engl J Med 1993; 329(14): 987 – 94.
Section 4 : Breast Cancer
16
Male Breast Cancer
Ian K. Komenaka, Kathy D. Miller and George W. Sledge, Jr
INTRODUCTION Although breast cancer is an uncommon malignancy in men, it has been recognized since antiquity. The earliest reference to breast cancer appears in the Edwin Smith Surgical Papyrus dating from 3000 to 2500 B.C. and refers to a man.1 The first clinical description is attributed to the 14th-century English surgeon John of Aderne who warned a priest with a large breast mass, that treatment by a barber “would bring him to death”.2 The subsequent scattered case reports were compiled by Williams in the late 19th century,3 but an exhaustive and detailed review of the basic characteristics of the disease did not appear until 1927.4 Knowledge of many of the relevant aspects of the disease and appropriate therapy remains limited. Large series of male breast cancer (MBC) are rare, retrospective, and cover extended time periods, generally at single institutions, during which methods of diagnosis, staging, and treatment may have changed dramatically. Prospective, randomized trials are not available. Treatment for men has therefore been based on the known biology of MBC and the knowledge gained from controlled clinical studies performed in women.
EPIDEMIOLOGY AND BIOLOGY The American Cancer Society estimated that in 2005 approximately 1690 men would be diagnosed with breast cancer in the United States, and that 460 patients would die from the disease.5 This accounts for 0.8% of all breast cancer cases, and only 0.23% of all malignancies in men. In the United States, breast cancer is responsible for 0.16% of all cancer deaths in men. The incidence of MBC, once thought to be relatively stable, now seems to be increasing. Incidence of MBC increased significantly from 0.86 to 1.06 per 100 000 population over the last 26 years.6 Rates of in situ tumors increased most rapidly up by 123% among men. Invasive localized disease rates were found to have increased by a more modest 37%. Conversely, the disease rates of invasive regional and distant breast carcinoma declined for men and women during the 1980s and 1990s. For example, the distant disease rate was found to have decreased 41% among men from 0.09 per 100 000 man-years during 1975–1980 to
0.05 per 100 000 man-years during 1997–2001.7 Advances in screening mammography could not account for the dramatic increases noted in early-stage breast carcinoma among men, given that men are not routinely screened for breast carcinoma. However, a heightened awareness of male breast carcinoma might result in the earlier detection of “symptomatic” in situ and invasive localized tumors because of the easier detection of lesions in men with a small breast volume.7 The prevalence of MBC increases with age, from 0.1 cases per 100 000 at 30–34 years to 6.5 cases per 100 000 at 85 years and older.8 The disease is extremely rare before the age of 30, although two children have been reported with the disease.9 The mean age at diagnosis is approximately 63–71 years, nearly 10 years older than the corresponding mean age for women.8,10 – 12 The reason for this age difference is not known. As with female breast cancer, the incidence of MBC varies with geographic location. The lowest rates are reported in Finland and Japan,13 and the highest incidence occurs in several African nations.14 In Zambia, nearly 15% of patients diagnosed with breast cancer are men.15 Regions with the highest incidence of MBC coincide with areas of increased incidence of liver disease, suggesting a possible association with higher estrogen levels. The cause of MBC remains elusive. Many etiologic factors have been proposed but supporting evidence is generally thin and based on small numbers of patients (see Table 1). One commonly reported factor is a hormonal imbalance between estrogen and androgens. The association between estrogen levels and breast cancer in men is of interest because estrogen-related risk factors have been strongly implicated in female breast cancer. The imbalance between estrogen and androgen may be the result of either estrogen excess or androgen deficiency. Conditions associated with MBC that increase levels of circulating estrogens in the body include chronic liver disease,7,16 obesity early in life,17,18 and pharmacologic estrogen therapy, particularly in prostate cancer.19 In obese men, estrogen production, metabolism, and bioavailability are enhanced. Levels of circulating estrogens may be increased with conversion of androgens to estradiol and
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
202
BREAST CANCER
Table 1 Factors implicated in the etiology of male breast cancer.
Estrogen – androgen imbalance Exogenous estrogen Klinefelter’s syndrome Treatment for prostate cancer Chronic liver disease Obesity Testicular abnormalities Orchiectomy Mumps orchitis Cryptorchidism Excess heat exposure Increased prolactin Drugs Head trauma Ionizing radiation Family History Genetic syndromes BRCA2
estrone in peripheral adipose tissue.20 Treatment required for male to female transexuality has been implicated in MBC. Surgical and chemical castration and prolonged administration of large doses of female hormones produce the estrogen–androgen imbalance.21,22 Similarly, estrogen therapy used in prostate cancer could lead to hormonal imbalance. Kanhai et al. demonstrated that after chemical castration for prostate cancer, there was moderate acinar and lobular formation of breast tissue.19 Disorders of testicular function result in hormonal imbalance by decreasing androgen production. A number of conditions have been implicated, including mumps orchitis, orchiectomy, undescended testes, and testicular injury.23 Several case series have found an unusually high association with Klinefelter’s syndrome, where patients display a eunuchoid habitus, gynecomastia, and small, firm testes. Patients have high levels of gonadotropins and low levels of androgens.24 Klinefelter’s syndrome has been documented by karyotype in up to 7% of MBC patients.25,26 Occupational environmental exposure to high temperatures has been associated with increased risk of breast cancer in men. The increased temperature presumably results in testicular damage with altered androgen and estrogen levels. Among the affected are men working in steel and rolling mills, those involved in machinery repair and motor vehicle manufacturing, and blast furnace workers.27,28 Exposure to carcinogens such as polycyclic aromatic hydrocarbons, nitrosamines, and metal fumes may also play a role.29 Hyperprolactinemia, whether precipitated by drugs, brain trauma, or skull fractures, may increase the risk of developing breast cancer.30 Radiation-induced breast cancers have also been reported.31,32 The relationship between gynecomastia, a proliferation of normal breast tissue under estrogenic stimulation, and MBC remains controversial. Microscopic evidence of ductal hyperplasia has been found in 40% of men with breast cancer.33 In addition, autopsy studies have shown gynecomastia to be prevalent in as many as 50% of all MBC patients.34 However, reviews of healthy men have found clinical gynecomastia to be present in 35–40%,35 and therefore the rates of gynecomastia in MBC patients may be similar to those in the general population.36
A possible familial association has been recognized for many years, with increased risk for close male and female relatives of MBC patients.37 Data from the Surveillance, Epidemiology, and End Results (SEER) program showed that men with a positive family history have an odds ratio of 3.98 for developing breast cancer.38 Advances in molecular analysis over the last decade have allowed the identification of breast cancer susceptibility genes, BRCA1 and BRCA2. Although most studies of BRCA1 carriers do not seem to be associated with an increased risk of MBC,39 one study demonstrated a 10.5% (8 of 76) incidence of BRCA1 mutations in Ashkenazi Jewish men with breast cancer.40 BRCA2 mutations, on the contrary, appear to predispose men to breast cancer.24 Wooster et al. localized the gene to chromosome 13q12–13 and described multiple cases of MBC linked to this area.41 The prevalence of BRCA2 mutations in men with breast cancer remains unclear. In a study of 237 families affected by breast cancer including at least one affected male, 76% were because of BRCA2.42 The frequency of BRCA mutations in a population-based series of 54 male patients without a strong family history found only two (4%) with BRCA2 mutations.43 PTEN tumor suppressor gene and CHEK2 kinase are two genes associated with female breast cancer that have also been associated with MBC. In addition, AR gene and CYP17 gene have also been suspected in MBC but have no definite role in female patients.24
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS The clinical manifestations of MBC have been well described by several authors.10,28,44 The most common presentation, a painless, firm subareolar mass, is seen in 75–85% of patients. There is a slight predilection for the left breast (1.07 : 1) in most collective series, with bilateral disease being quite unusual.10,13,36,45 Other presenting symptoms include nipple discharge, ulceration or bleeding of the nipple, pain or swelling of the breast, or symptoms related to metastatic disease.46 In addition to the palpable mass, common physical examination findings include nipple retraction, inversion or fixation, nipple discharge, and mastitis. The rate of nipple involvement has been reported to be as high as 40–50% because of the sparsity of breast tissue and the central location of tumors.36 Signs of Paget’s disease occur in approximately 3–5% of patients.10,47 Paget’s disease in men is nearly always associated with an underlying invasive cancer.48 Clinically involved axillary nodes occur in 40–55% of patients at presentation.48,49 Palpable axillary adenopathy with an occult breast tumor, however, is uncommon in men.9 The frequency of locally advanced disease with skin ulceration has declined as the delay in diagnosis has decreased from 18 months in older series to less than 6 months in some series after 1981.9,50,51 The main differential diagnosis is between gynecomastia and breast cancer. Other benign tumors of the male breast include lipoma, inclusion cyst, lymph node, fat necrosis, and leiomyoma, but are exceedingly rare.52 In men suspected
MALE BREAST CANCER
of having breast cancer, the diagnostic evaluation should parallel that recommended for women. The distinction on physical examination may be difficult but a bloody nipple discharge strongly favors carcinoma.53,54 Several studies have demonstrated that mammography can reasonably differentiate gynecomastia and breast cancer.52,55,56 One study of 100 male patients demonstrated 90% accuracy in differentiating the two pathologies.55 An uncalcified mass is most commonly seen; microcalcifications are less frequent and tend to be more scattered with a coarse appearance.54,56 Ultrasonography has proven helpful in imaging lesions in the male breast with 80% of breast cancers having irregular margins.56 Complex cystic lesions, heterogeneous hypoechoic lesions, and increased vascularity are other ultrasonographic signs concerning malignancy.57 Fine needle aspiration biopsy with conventional cytologic examination has been demonstrated to be very accurate with high sensitivity and specificity.58 If cytology is inconclusive (5%) or specimen is insufficient (22%), an open biopsy is required and should be performed if any question remains following diagnostic imaging. Additional evaluation following cytologic diagnosis should include biochemical evaluation of renal and liver function, and a plain chest radiograph. Bone scans, abdominal computed tomography, and brain imaging should be reserved for patients with symptoms, laboratory abnormalities, or advanced disease. The values of serum tumor markers such as CEA, CA 15–3, and CA 27–29 have not been evaluated in MBC.
PATHOLOGY Virtually all known histologic types of breast cancer have been reported in men. As in women, infiltrating ductal carcinoma, with or without an intraductal component, predominates and accounts for more than 80% of cases.10 Ductal carcinoma in situ (DCIS) is found in about 2–17% of cases, although the true incidence is difficult to determine as many historical series eliminated those patients.59 Given the absence of screening mammography in men, the striking rise in DCIS seen in women is unlikely to be paralleled in men. Medullary, tubular, papillary, mucinous, and inflammatory variants have been reported in a minority of patients. Conventional wisdom suggested that lobular carcinoma did not occur in men because of the lack of terminal differentiation in the rudimentary male breast. Both invasive and in situ forms of lobular carcinoma, however, have been reported.60,61 Benign and malignant sarcomas have been well described and may account for up to 8% of male breast tumors.62 The breast is not often considered a site of metastatic disease; however, prostate cancer is the most common tumor to metastasize to the breast.63 Estrogen and progesterone receptors are more commonly positive in MBC than in the female counterpart. A study using data from the SEER database demonstrated 90% positivity in 680 MBC patients with known estrogen receptor status.6 Similarly, progesterone receptors were found to be positive in over 81% in those with known receptor status. Consistently high levels of estrogen (81%) and progesterone (74%) receptors are found in the largest literature review
203
of studies of 1301 MBC patients.36 These rates approximate those seen in older women, possibly because of the similarities in physiologic estrogen status. In contrast with female breast cancer, there is no correlation between age and receptor positivity. The prognostic utility of estrogen and progesterone receptors in men has not been established. The number of studies looking at the prognostic impact of hormone receptors in men is limited. Data from the SEER database demonstrated similar 5-year survival rates in both the hormone receptor-positive and -negative groups.6 In addition, two large studies demonstrate contrasting results. A study of 229 patients from the Princess Margaret Hospital did not find any significant difference in overall survival after adjustment for key factors such as size of tumor, lymph node status, and type of treatment.64 A study of 215 patients from Wisconsin, however, did find an improved overall survival after adjustment for tumor stage and lymph node status.45 One recent study found expression of androgen receptor in tumor tissue to be an independently adverse factor for both disease-free and overall survival.65 The role of Her-2/neu expression in MBC is also not well defined. A recent review of Her-2/neu by immunohistochemistry in pooled data of 511 male patients demonstrated 37% overexpression.36 However, the antibody preparations and definitions for positive staining varied significantly among the studies. More contemporary studies suggest that older studies may have overestimated the rate of Her-2/neu overexpression.66 Recently, 65 patients were studied by immunohistochemistry. Scoring was performed according to currently established guidelines and 9% (6/65) demonstrated 2+ or 3+ overexpression.67 Another study of 99 patients found 15.1% (15/99) demonstrated 2+ or 3+ overexpression.68 These cases were then tested for Her2/neu gene amplification by fluorescence in situ hybridization (FISH). Seven of seven 3+ staining samples demonstrated gene amplification, as well as four of eight 2+ staining samples. Her-2/neu gene amplification/protein overexpression did not correlate with tumor state, histological grade, estrogen/progesterone receptor status, or the axillary lymph node status. In addition, it appears that Her-2/neu positivity is lower than in female breast cancer patients. It is unknown if Her-2/neu overexpression in males has the same prognostic implications as in females; however, one study found decreased disease-free survival with Her-2/neu overexpression.67
PROGNOSIS The same Tumor-Node-Metastasis (TNM) system is used to stage both male and female breast cancer. Older series demonstrated higher rates of presentation with regional disease and correspondingly low 5-year survival rates.69 Scheike found an increase in patients presenting with stage I disease from 20% in 1943–1957, to 44% in 1958–1972.47 There was a similar decrease in patients with stage IV disease, although the percentage of patients with stage II disease remained constant. In comparison, recent National Cancer Institute (NCI) SEER data demonstrate that 41% of male patients present with disease localized to the breast.
204
BREAST CANCER
In addition, 37% present with regional node involvement.7 According to American Joint Committee on cancer (AJCC) staging, these patients presented as stage 0 – 10%, stage I – 29%, stage II – 38%, stage III – 7%, and stage IV – 8%. Despite increases in the number of patients diagnosed with localized disease, the number of patients presenting with distant metastasis has not changed greatly. In the past, approximately 10% of both male and female patients presented with overt metastasis, however, more recently this number was approximately 7–8%.7 The risk of contralateral second breast cancer in men with a diagnosis of carcinoma of the breast was greatly increased in 1788 men in the SEER database. This risk was greater in men diagnosed with the first breast cancer before the age of 50 and was not associated with the type of treatment.70 Stage and axillary nodal status continue to be the most important prognostic indicators in MBC (see Table 2). One report correlated survival with the number of pathologically involved nodes, Guinee et al. found 5-year survivals of 90, 73, and 55% for patients with zero, one to three, and four or more positive nodes, respectively.49 Similarly, Lartigau et al. reported 10-year survivals of 84, 44, and 14% for patients with zero, one to three, and four or more involved lymph nodes.71 Tumor size appears to be an important prognostic factor with 5-year survival of 74% for T1, 53% for T2, and 37% for T3 lesions. Histologic grade also correlates well with survival; 5-year survival decreased from 74% for grade I tumors to 53% for those with grade III histologic features.6 The prognostic significance of ancillary studies used in female breast cancer has been poorly evaluated in men. DNA ploidy, percentage of cells in S-phase, Ki67, Cathespin D, and p53 have not demonstrated a reproducible correlation with prognosis. The relative prognosis of men and women with breast cancer has been the subject of much debate. Early reports suggested a much worse survival for men compared with women. In studies in which male and female patients were matched for age and stage, however, survivals were similar.49,74,75 The previously observed differences in prognosis were thought to result from delays in diagnosis rather than a different biology or tumor aggressiveness. Comparisons that further stratify for the number of axillary nodes involved confirm similar survival.49 Thus, while stage for
stage the prognosis is equivalent in men and women, the tendency for more advanced disease at presentation and older median age seen in men may result in a lower overall survival. These differences are most likely because of the older age of male patients, death from other causes, and the lower life expectancy of men in the general population.6,75
TREATMENT Localized Disease The traditional approach most often used in series prior to 1960 was the Halstedian radical mastectomy. The pattern of surgical treatment has been based on the extent of disease at presentation and the standard of care for contemporary women. While randomized trials have proven the safety and efficacy of less aggressive surgical procedures in women with breast cancer, similar data does not exist for men. The finding that modified radical mastectomy is equivalent to radical mastectomy in prospective randomized trials in women, has led to the use of modified radical mastectomy in men. Studies that have compared radical mastectomy and modified radical mastectomy in men have found no difference in local recurrence rates and in overall survival rates.50,76,77 Most modern series use modified radical mastectomy as the standard local therapy for men.36 Breast cosmesis is not a primary consideration for most men. Therefore, there has been little enthusiasm for investigating breast-conserving treatments. Successful breast conservation with lumpectomy and radiation therapy, however, has been described in a small series of patients with small breast cancers.78 The use of the sentinel node procedure has replaced axillary node dissection in women with a clinically negative axilla. There are many large reviews demonstrating its efficacy and accuracy in women. It is also the subject of two ongoing large prospective, randomized trials. In men, however, there have been a very limited number of patients evaluated with the sentinel node procedure. One trial of 18 patients using only colloid human albumin labeled with 99 Tc demonstrated successful identification of the node in all patients. Six of 18 patients (33%) had positive sentinel nodes.79 Two other series used the combination of blue dye and 99m Tc-radiolabeled colloidal albumin. All patients in the
Table 2 Overall survival based on pathologic axillary nodal status.
Five-year overall survival (%) Author
Years 69
Ramantanis et al. Yap et al.72 Heller et al.33a Donegan et al.45 Cutuli et al.10,73 Guinee et al.49c Erlichman et al.46 Borgen et al.51,74
1937 – 1974 1945 – 1975 1949 – 1976 1953 – 1995 1960 – 1986 1965 – 1986 1967 – 1981 1975 – 1990
NR, not reported; LN, lymph node. a 10-year survival rate. b Survival calculated for operable patients only. c Disease-specific survival.
Total patients
All
LN negative
LN positive
138 87 97 156 397 335 89 104
32.5 42 40 50 65 NR NR 85
56.5 77 79 73.6b 82 90 77 100
30.8 37 11 64.7b 61 65 37 60
MALE BREAST CANCER
two series had successful identification of sentinel nodes (nine and seven patients).80,81 Five of nine patients (56%) and one of seven patients (14%) had positive sentinel nodes, respectively. Routine axillary dissections were not performed in either trial and therefore false-negative rates could not be determined. On the basis of this limited data, it appears that the sentinel node procedure in MBC patients is successful in the hands of experienced surgeons, avoids unnecessary removal of uninvolved lymph nodes, and reduces length of hospital stay.48 The role of adjuvant radiotherapy to the chest wall and axilla is poorly defined. The use of radiotherapy has varied widely, with some institutions recommending radiation only to those patients with inoperable tumors,71 while others recommend radiation to nearly all patients.49 Several retrospective reviews have found that postmastectomy radiation does reduce the risk of local recurrence but does not improve overall survival.10,45,82 Given the lack of definitive data, it seems reasonable to consider local radiation in patients with risk factors similar to those described in women, i.e. those with large primary tumors, chest wall invasion, or multiple positive lymph nodes. Radiation to the axilla can be reduced especially after complete axillary lymph node dissection, when the risk of axillary recurrence is less than 1%.83 Although uncommon, when breast conservation is chosen in men, adjuvant radiation therapy is indicated, as analogous in women. The significant benefit of adjuvant hormonal therapy in women combined with the high frequency of estrogen receptor expression in MBC has sparked interest in the use of adjuvant hormonal therapy to prevent systemic recurrence. Ribeiro and Swindell reported the results of an unselected series of patients with stage II and III disease treated with adjuvant tamoxifen for 1 to 2 years.84 After a median follow-up of 49 months, the 39 treated patients had a 5-year actuarial survival of 61%, compared with 44% in a historical control group. Two other series have found an improved survival after treatment with tamoxifen.44,64 Although these patients reported few serious side effects, the experience at the Memorial Sloan–Kettering Cancer Center is quite different. Between 1990 and 1993, 24 patients received tamoxifen as adjuvant therapy for MBC. More than half the patients reported at least one side effect – most commonly decreased libido, weight gain, hot flushes, and mood alterations or depression. Five patients (20.8%) discontinued therapy within 1 year.85 Although the side effects reported by Anelli et al. should not be taken lightly, they are hardly life-threatening in severity and should not discourage the use of adjuvant tamoxifen in men. Further, while the standard duration of therapy with tamoxifen in women is 5 years, most reviews in men have been for less than 2 years of tamoxifen. Therefore, the benefit may be underestimated if proportional benefit parallels duration of treatment as in women. The finding of increased diseasefree survival in postmenopausal women with Anastrozole has not yet resulted in significant reports of the adjuvant use of third-generation aromatase inhibitors in men.86 In addition, the improved disease-free survival in women switching to exemestane or letrozole after 2 to 3 years or 5 years,
205
respectively, of treatment with tamoxifen has also not been reported in men.87,88 Recommendations for adjuvant chemotherapy have been largely based on studies performed in women. The NCI completed a phase II study of 24 men with stage II disease and histologically proven nodal involvement. All patients received cyclophosphamide, methotrexate, and 5–fluorouracil (CMF) for a total of 12 months beginning within 4 weeks of definitive surgery. None of the patients received adjuvant radiation therapy. After a median followup of 46 months, the projected 5-year survival of 80% compares well with historical controls. Two of the four patients with recurrent disease were disease-free for more than 60 months, raising the question as to whether adjuvant therapy prevents or merely delays recurrence.89 In a similar study of stage II and III patients using doxorubicin-based chemotherapy, Patel et al. found a 5-year survival of over 85%, with 64% of patients remaining disease-free.90 Other authors have found improved outcomes with adjuvant chemotherapy.36 The rarity of MBC precludes randomized trials of adjuvant therapy in this population. The limited data available suggest that adjuvant hormonal or chemotherapy may improve disease-free and overall survival in patients with node-positive disease compared to that achieved with local therapies alone. Furthermore, adjuvant therapy should be considered in patients with tumors greater than 1 cm. The treatment of men with pure DCIS requires special mention. Cutuli et al. reviewed 31 men treated at 19 French regional cancer centers over a 22-year period.73 All patients were treated with surgical excision; six underwent lumpectomy, and 25 underwent mastectomy. Axillary dissection in 19 patients found no evidence of nodal involvement. After a median follow-up of 83 months, four patients had a local recurrence, including three of those initially treated with lumpectomy. One patient’s disease remained in situ; three had developed invasive disease but were salvaged with radical surgical excision. One patient developed contralateral DCIS. One patient developed metastasis and died 30 months after local recurrence.73 As in women, DCIS is primarily a local disease with an excellent prognosis. Simple mastectomy without axillary node dissection is the treatment of choice.
Metastatic Disease The median survival from the time of presentation with metastatic disease is about 26.5 months,77 although the range is large and long-term survivors have been reported.91 Since approximately 8% of men present with metastatic disease and many of those treated locally will recur at some point during the course of their illness, the need for effective treatment cannot be overemphasized. Strategies for treatment of men with disseminated disease have mirrored those developed for women. Hormonal therapy, either ablative or additive, and systemic chemotherapy have both been used with some success; however, controlled studies are lacking. Hormonal therapy has been the mainstay of treatment since Farrow and Adair first reported healing of skeletal metastasis after orchiectomy in 1942.92 In a review of 70 men treated with orchiectomy, Meyskens et al. found a collective response rate of 67% with a median duration of
206
BREAST CANCER
response of 22 months.93 In addition, one review found that patients who responded to orchiectomy were more likely to respond to second-line ablative therapies, and responding patients had improved survival.94 Jaiyesimi et al. reviewed ablative therapies in 447 patients and found response rates of 55% for orchiectomy, 80% for adrenalectomy, and 56% for hypophysectomy.95 Investigators at the Roswell Park Memorial Institute confirm the effectiveness of adrenal ablation. Eight of 10 patients responded with a median duration of response of 15 months. Of these eight patients, five had previously responded to orchiectomy and three had had no prior response.96 Despite excellent response rates to hormonal ablation, these treatments have rarely been used as first-line therapy since the advent of effective additive hormonal treatments. Additive hormonal therapies eliminate the psychological opposition to orchiectomy and the substantial surgical morbidity and side effect risk associated with adrenalectomy and hypophysectomy while preserving response. The overall response rates of additive therapies have been reported as 75% for androgens, 57% for antiandrogens, 40% for aminoglutethimide, and 49% for tamoxifen.36 The efficacy of tamoxifen in metastatic MBC has been documented in sporadic case reports and small series. Patterson et al. found 31 patients in 16 collected reports with a cumulative response rate of 48%.97 Ribeiro treated 24 patients with advanced disease and reported a response rate of 38%.98 Median duration of response was 21 months with a range of 8 to 60 months. While these response rates are somewhat lower than those reported with orchiectomy, the percentage of patients with estrogen receptor-positive tumors in each study is not known. The relationship between steroid receptor status and response to tamoxifen is clear. In both reports, more than 80% of known receptor-positive patients responded while there were no documented responses in those patients known to be receptor negative. The widespread use of aromatase inhibitors in postmenopausal women has not yet been exported to men with breast cancer. Success with the early aromatase inhibitor aminoglutethimide has been limited in men. In women, the third-generation aromatase inhibitors Anastrozole, letrozole, and exemestane have shown good results compared with tamoxifen as first-line agents. In the only reported series (in men) to date, three out of five men responded to Anastrozole, with a mean duration of response of 8 months.99 A recent case reported an ongoing clinical complete response of 12 months with the use of letrozole.100 Other hormonal manipulations have also documented effectiveness in small studies. Medroxyprogesterone, cyproterone acetate, the combination of the gonadotropin-releasing hormone agonist analog buserelin and the antiandrogen flutamide have had variable success. Other agents that have produced temporary regressions include estrogens, prednisone, and androgens. Systemic chemotherapy has not been rigorously studied in MBC. As many patients benefit from hormonal manipulations, chemotherapy is generally reserved as second-line therapy for those patients whose disease has become refractory to hormonal agents. Lopez et al. directly compared hormonal therapy and chemotherapy in a small series of
14 patients, and higher response rates were observed after hormonal therapy.98 At least two studies have suggested that though the response to chemotherapy is often faster in men than in women with metastatic breast cancer, the duration of response is shorter.12,101 Reported response rates for the various regimens include 67% for 5-flurouracil, doxorubicin, cyclophosphamide; 55% for other doxorubicin containing regimens, and 33% for CMF-like regimens.36
AUTHORS’ RECOMMENDATIONS Carcinoma of the male breast is uncommon, but not rare. The limited numbers of patients and the inability to conduct cooperative studies have prevented careful research. Accordingly, conclusions must be based on the available literature. The following recommendations find some support in the available data but cannot and should not be regarded as definitively proven. 1. Carcinoma of the male breast is analogous to carcinoma of the female breast. The lone exception is the higher rate of estrogen receptor positivity and response to hormonal therapy seen in men. 2. Genetic factors such as BRCA2 increase the risk of developing MBC but overall account for a minority of patients. 3. A modified radical mastectomy is the surgical procedure of choice for most patients. Total mastectomy and sentinel node biopsy can be considered by experienced surgeons. 4. Radiation therapy may reduce the risk of local recurrence in patients with large tumors and/or multiple positive lymph nodes but has no impact on survival. 5. Adjuvant hormonal therapy with tamoxifen and adjuvant chemotherapy may improve survival in patients with axillary nodal involvement. Treatment decisions need to be individualized. 6. Metastatic disease is highly responsive to hormonal therapy; this should be the initial mode of treatment in men with hormone receptor-positive cancers. Ablative therapies may have a higher response rate but are generally much less acceptable to patients than tamoxifen. Many patients will respond to sequential hormonal therapies. 7. Response rates to cytotoxic chemotherapy are similar to those seen in women and may provide palliation to patients in whom hormonal therapy is no longer effective.
REFERENCES 1. Breasted JH. The Edwin Smith Surgical Papyrus. Chicago, Illinois: University of Chicago Press, 1930: 403 – 406. 2. Holleb AI, Freeman HP, Farrow JH. Cancer of male breast. I. N Y State J Med 1968; 68: 544 – 53. 3. Williams W. Cancer of the male breast, based on records of 100 cases; with remarks. Lancet 1889; 2: 261 – 3. 4. Wainwright J. Carcinoma of the male breast. Arch Surg 1927; 14: 846 – 52. 5. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55(1): 10 – 30. 6. Giordano SH, et al. Breast carcinoma in men: a population-based study. Cancer 2004; 101(1): 51 – 7. 7. Anderson WF, Devesa SS. Breast carcinoma in men. Cancer 2005; 103(2): 432 – 3; author reply 433.
MALE BREAST CANCER 8. Donegan W. Cancer of the male breast. In Donegan W (ed) Cancer of the Breast, 3rd ed. Philadelphia, Pennsylvania: W.B. Saunders, 1988: 716 – 727. 9. Crichlow RW. Carcinoma of the male breast. Surg Gynecol Obstet 1972; 134(6): 1011 – 9. 10. Cutuli B, et al. Male breast cancer: results of the treatments and prognostic factors in 397 cases. Eur J Cancer 1995; 31A(12): 1960 – 4. 11. Anderson WF, et al. Is male breast cancer similar or different than female breast cancer? Breast Cancer Res Treat 2004; 83(1): 77 – 86. 12. Ribeiro G. Male breast carcinoma – a review of 301 cases from the Christie Hospital & Holt Radium Institute, Manchester. Br J Cancer 1985; 51(1): 115 – 9. 13. Ewertz M, et al. Incidence of male breast cancer in Scandinavia, 1943 – 1982. Int J Cancer 1989; 43(1): 27 – 31. 14. El-Gazayerli M, Abdel-Aziz A. On bilharziasis and male breast cancer in Egypt. Br J Cancer 1963; 17: 556 – 71. 15. Bhagwandeen SB. Carcinoma of the male breast in Zambia. East Afr Med J 1972; 49(2): 89 – 93. 16. Sorensen HT, et al. Risk of breast cancer in men with liver cirrhosis. Am J Gastroenterol 1998; 93(2): 231 – 3. 17. Ballerini P, et al. Hormones in male breast cancer. Tumori 1990; 76(1): 26 – 8. 18. Ewertz M, et al. Risk factors for male breast cancer – a case-control study from Scandinavia. Acta Oncol 2001; 40(4): 467 – 71. 19. Thellenberg C, et al. Second primary cancers in men with prostate cancer: an increased risk of male breast cancer. J Urol 2003; 169(4): 1345 – 8. 20. Hsing AW, et al. Risk factors for male breast cancer (United States). Cancer Causes Control 1998; 9(3): 269 – 75. 21. Symmers WS. Carcinoma of breast in trans-sexual individuals after surgical and hormonal interference with the primary and secondary sex characteristics. Br Med J 1968; 2(597): 82 – 5. 22. Kanhai RC, et al. Short-term and long-term histologic effects of castration and estrogen treatment on breast tissue of 14 male-to-female transsexuals in comparison with two chemically castrated men. Am J Surg Pathol 2000; 24(1): 74 – 80. 23. Thomas DB, et al. Breast cancer in men: risk factors with hormonal implications. Am J Epidemiol 1992; 135(7): 734 – 48. 24. Weiss JR, Moysich KB, Swede H. Epidemiology of male breast cancer. Cancer Epidemiol Biomarkers Prev 2005; 14(1): 20 – 6. 25. Evans DB, Crichlow RW. Carcinoma of the male breast and Klinefelter’s syndrome: is there an association? CA Cancer J Clin 1987; 37(4): 246 – 51. 26. Hultborn R, et al. Prevalence of Klinefelter’s syndrome in male breast cancer patients. Anticancer Res 1997; 17(6D): 4293 – 7. 27. Rosenbaum PF, et al. Occupational exposures associated with male breast cancer. Am J Epidemiol 1994; 139(1): 30 – 6. 28. Pollan M, Gustavsson P, Floderus B. Breast cancer, occupation, and exposure to electromagnetic fields among Swedish men. Am J Ind Med 2001; 39(3): 276 – 85. 29. Cocco P, et al. Case-control study of occupational exposures and male breast cancer. Occup Environ Med 1998; 55(9): 599 – 604. 30. Olsson H, Ranstam J. Head trauma and exposure to prolactin-elevating drugs as risk factors for male breast cancer. J Natl Cancer Inst 1988; 80(9): 679 – 83. 31. Eldar S, Nash E, Abrahamson J. Radiation carcinogenesis in the male breast. Eur J Surg Oncol 1989; 15(3): 274 – 8. 32. Ron E, et al. Male breast cancer incidence among atomic bomb survivors. J Natl Cancer Inst 2005; 97(8): 603 – 5. 33. Heller KS, et al. Male breast cancer: a clinicopathologic study of 97 cases. Ann Surg 1978; 188(1): 60 – 5. 34. Andersen JA, Gram JB. Male breast at autopsy. Acta Pathol Microbiol Immunol Scand [A] 1982; 90(3): 191 – 7. 35. Braunstein GD. Gynecomastia. N Engl J Med 1993; 328(7): 490 – 5. 36. Giordano SH, Buzdar AU, Hortobagyi GN. Breast cancer in men. Ann Intern Med 2002; 137(8): 678 – 87. 37. Olsson H, et al. Population-based cohort investigations of the risk for malignant tumors in first-degree relatives and wives of men with breast cancer. Cancer 1993; 71(4): 1273 – 8. 38. Rosenblatt KA, et al. Breast cancer in men: aspects of familial aggregation. J Natl Cancer Inst 1991; 83(12): 849 – 54.
207
39. Hall JM, et al. Linkage of early-onset familial breast cancer to chromosome 17q21. Science 1990; 250(4988): 1684 – 9. 40. Frank TS, et al. Clinical characteristics of individuals with germline mutations in BRCA1 and BRCA2: analysis of 10,000 individuals. J Clin Oncol 2002; 20(6): 1480 – 90. 41. Wooster R, et al. Localization of a breast cancer susceptibility gene, BRCA2, to chromosome 13q12-13. Science 1994; 265(5181): 2088 – 90. 42. Ford D, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1998; Mar; 62(3): 676 – 89. 43. Friedman LS, et al. Mutation analysis of BRCA1 and BRCA2 in a male breast cancer population. Am J Hum Genet 1997; 60(2): 313 – 9. 44. Ribeiro G, et al. A review of the management of the male breast carcinoma based on an analysis of 420 treated cases. The Breast 1996; 5: 141 – 6. 45. Donegan WL, et al. Carcinoma of the breast in males: a multiinstitutional survey. Cancer 1998; 83(3): 498 – 509. 46. Erlichman C, Murphy KC, Elhakim T. Male breast cancer: a 13-year review of 89 patients. J Clin Oncol 1984; 2(8): 903 – 9. 47. Scheike O. Male breast cancer. 5. Clinical manifestations in 257 cases in Denmark. Br J Cancer 1973; 28(6): 552 – 61. 48. Gennari R, et al. Male breast cancer: a special therapeutic problem. Anything new? (Review). Int J Oncol 2004; 24(3): 663 – 70. 49. Guinee VF, et al. The prognosis of breast cancer in males. A report of 335 cases. Cancer 1993; 71(1): 154 – 61. 50. Gough DB, et al. A 50-year experience of male breast cancer: is outcome changing? Surg Oncol 1993; 2(6): 325 – 33. 51. Borgen PI, et al. Current management of male breast cancer. A review of 104 cases. Ann Surg 1992; 215(5): 451 – 7; discussion 457 – 9. 52. Appelbaum AH, et al. Mammographic appearances of male breast disease. Radiographics 1999; 19(3): 559 – 68. 53. Amoroso WL, Robbins GF, Treves N. Serous and serosanguinous discharge from the male nipple. Arch Surg 1956; 73(2): 319 – 29. 54. Dershaw DD, et al. Mammographic findings in men with breast cancer. AJR Am J Roentgenol 1993; 160(2): 267 – 70. 55. Evans GF, et al. The diagnostic accuracy of mammography in the evaluation of male breast disease. Am J Surg 2001; 181(2): 96 – 100. 56. Gunhan-Bilgen I, et al. Male breast disease: clinical, mammographic, and ultrasonographic features. Eur J Radiol 2002; 43(3): 246 – 55. 57. Yang WT, et al. Sonographic features of primary breast cancer in men. AJR Am J Roentgenol 2001; 176(2): 413 – 6. 58. Joshi A, Kapila K, Verma K. Fine needle aspiration cytology in the management of male breast masses. Nineteen years of experience. Acta Cytol 1999; 43(3): 334 – 8. 59. Hittmair AP, Lininger RA, Tavassoli FA. Ductal carcinoma in situ (DCIS) in the male breast: a morphologic study of 84 cases of pure DCIS and 30 cases of DCIS associated with invasive carcinoma – a preliminary report. Cancer 1998; 83(10): 2139 – 49. 60. Sanchez AG, Villanueva AG, Redondo C. Lobular carcinoma of the breast in a patient with Klinefelter’s syndrome. A case with bilateral, synchronous, histologically different breast tumors. Cancer 1986; 57(6): 1181 – 3. 61. Nance KV, Reddick RL. In situ and infiltrating lobular carcinoma of the male breast. Hum Pathol 1989; 20(12): 1220 – 2. 62. Visfeldt J, Scheike O. Male breast cancer. I. Histologic typing and grading of 187 Danish cases. Cancer 1973; 32(4): 985 – 90. 63. Green LK, Klima M. The use of immunohistochemistry in metastatic prostatic adenocarcinoma to the breast. Hum Pathol 1991; 22(3): 242 – 6. 64. Goss PE, et al. Male breast carcinoma: a review of 229 patients who presented to the princess Margaret hospital during 40 years: 1955 – 1996. Cancer 1999; 85(3): 629 – 39. 65. Kwiatkowska E, et al. BRCA2 mutations and androgen receptor expression as independent predictors of outcome of male breast cancer patients. Clin Cancer Res 2003; 9(12): 4452 – 9. 66. Bloom K, et al. Male breast carcinomas do not show amplification of the her-2/neu gene (Abstract). Breast Cancer Res Treat 2000; 64(1): 127.
208
BREAST CANCER
67. Wang-Rodriguez J, et al. Male breast carcinoma: correlation of ER, PR, Ki-67, Her2-Neu, and p53 with treatment and survival, a study of 65 cases. Mod Pathol 2002; 15(8): 853 – 61. 68. Rudlowski C, et al. Her-2/neu gene amplification and protein expression in primary male breast cancer. Breast Cancer Res Treat 2004; 84(3): 215 – 23. 69. Ramantanis G, Besbeas S, Garas JG. Breast cancer in the male: a report of 138 cases. World J Surg 1980; 4(5): 621 – 3. 70. Auvinen A, Curtis RE, Ron E. Risk of subsequent cancer following breast cancer in men. J Natl Cancer Inst 2002; 94(17): 1330 – 2. 71. Lartigau E, et al. Male breast carcinoma: a single centre report of clinical parameters. Clin Oncol (R Coll Radiol) 1994; 6(3): 162 – 6. 72. Yap HY, et al. Male breast cancer: a natural history study. Cancer 1979; 44(2): 748 – 54. 73. Cutuli B, et al. Ductal carcinoma in situ of the male breast. Analysis of 31 cases. Eur J Cancer 1997; 33(1): 35 – 8. 74. Borgen PI, et al. Carcinoma of the male breast: analysis of prognosis compared with matched female patients. Ann Surg Oncol 1997; 4(5): 385 – 8. 75. El-Tamer MB, et al. Men with breast cancer have better diseasespecific survival than women. Arch Surg 2004; 139(10): 1079 – 82. 76. Ouriel K, Lotze MT, Hinshaw JR. Prognostic factors of carcinoma of the male breast. Surg Gynecol Obstet 1984; 159(4): 373 – 6. 77. Digenis AG, et al. Carcinoma of the male breast: a review of 41 cases. South Med J 1990; 83(10): 1162 – 7. 78. Vetto J, et al. Stages at presentation, prognostic factors, and outcome of breast cancer in males. Am J Surg 1999; 177(5): 379 – 83. 79. De Cicco C, et al. Sentinel node biopsy in male breast cancer. Nucl Med Commun 2004; 25(2): 139 – 43. 80. Goyal A et al. ALMANAC Trialists Group. Sentinel lymph node biopsy in male breast cancer patients. Eur J Surg Oncol 2004; 30(5): 480 – 3. 81. Albo D, et al. Evaluation of lymph node status in male breast cancer patients: a role for sentinel lymph node biopsy. Breast Cancer Res Treat 2003; 77(1): 9 – 14. 82. Zabel A, et al. External beam radiotherapy in the treatment of male breast carcinoma: patterns of failure in a single institute experience. Tumori 2005; 91(2): 151 – 5. 83. Cutuli B. The impact of loco-regional radiotherapy on the survival of breast cancer patients. Proc Eur J Cancer 2000; 36(15): 1895 – 902. 84. Ribeiro G, Swindell R. Adjuvant tamoxifen for male breast cancer (MBC). Br J Cancer 1992; 65(2): 252 – 4.
85. Anelli TF, et al. Tamoxifen administration is associated with a high rate of treatment-limiting symptoms in male breast cancer patients. Cancer 1994; 74(1): 74 – 7. 86. Howell A et al. ATAC Trialists’ Group. Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years’ adjuvant treatment for breast cancer. Lancet 2005; 365(9453): 60 – 2. 87. Coombes RC et al. Intergroup Exemestane Study. A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N Engl J Med 2004; 350(11): 1081 – 92. 88. Goss PE, et al. A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med 2003; 349(19): 1793 – 802. Epub 2003 Oct 9. 89. Bagley CS, et al. Adjuvant chemotherapy in males with cancer of the breast. Am J Clin Oncol 1987; 10(1): 55 – 60. 90. Patel HZ II, Buzdar AU, Hortobagyi GN. Role of adjuvant chemotherapy in male breast cancer. Cancer 1989; 64(8): 1583 – 5. 91. Siddiqui T, et al. Cancer of the male breast with prolonged survival. Cancer 1988; 62(8): 1632 – 6. 92. Farrow J, Adair F. Effect of orchiectomy on skeletal metastases from cancer of the male breast. Science 1942; 95: 654. 93. Meyskens FL Jr, Tormey DC, Neifeld JP. Male breast cancer: a review. Cancer Treat Rev 1976; 3(2): 83 – 93. 94. Kantarjian H, et al. Hormonal therapy for metastatic male breast cancer. Arch Intern Med 1983; 143(2): 237 – 40. 95. Jaiyesimi IA, et al. Carcinoma of the male breast. Ann Intern Med 1992; 117(9): 771 – 7. 96. Kennedy BJ, Kiang DT. Hypophysectomy in the treatment of advanced cancer of the male breast. Cancer 1972; 29(6): 1606 – 12. 97. Patterson JS, Battersby LA, Bach BK. Use of tamoxifen in advanced male breast cancer. Cancer Treat Rep 1980; 64(6 – 7): 801 – 4. 98. Lopez M, et al. Hormonal treatment of disseminated male breast cancer. Oncology 1985; 42(6): 345 – 9. 99. Giordano SH, et al. Efficacy of anastrozole in male breast cancer. Am J Clin Oncol 2002; 25(3): 235 – 7. 100. Zabolotny BP, Zalai CV, Meterissian SH. Successful use of letrozole in male breast cancer: a case report and review of hormonal therapy for male breast cancer. J Surg Oncol 2005; 90(1): 26 – 30. 101. Lopez M, et al. Chemotherapy in metastatic male breast cancer. Oncology 1985; 42(4): 205 – 9.
Section 4 : Breast Cancer
17
Phyllodes Tumor of the Breast Ian Ellis, Elinor J. Sawyer, Raj Rampaul and Carlos G. Pineda
INTRODUCTION AND HISTORICAL BACKGROUND Phyllodes tumors are a rare entity in the breast. They make up less than 1% of all breast tumors. They form part of the spectrum of fibroepithelial tumors of the breast, which includes the common benign fibroadenoma. It was first described by Johannes M¨uller in 1838, who used the term cystosarcoma phyllodes, as phyllode described the leaflike projection of the stroma into cystic spaces while sarcoma referred to the “fleshy” consistency of the tumor. The term cystosarcoma phyllodes is a misleading description, as the tumors are rarely cystic and the majority follow a benign clinical course. In total, more than 60 synonyms have been reported1 but the preferred term used by most centers and by the World Health Organization is “phyllodes tumor”. They display a broad range of clinical and pathological behavior; phyllodes tumors should be regarded as a spectrum of fibroepithelial neoplasms rather than a single disease entity. At one extreme, malignant phyllodes tumors, if inadequately treated, have a propensity for rapid growth and metastatic spread. In contrast, benign phyllodes tumors on clinical, radiological, and cytological examination are often indistinguishable from fibroadenomas and can be cured by local surgery. With the nonoperative management of fibroadenomas widely adopted, the importance of phyllodes tumors lies in the need to differentiate them from other benign breast lesions and to adopt a considered approach to management.
EPIDEMIOLOGY & BIOLOGY Epidemiology Phyllodes tumors are rare and account for 1% of all mammary neoplasms and 2.5% of fibroepithelial lesions in the breast2 in contrast to fibroadenomas, which are the most common benign breast lesion. Like fibroadenomas, they are composed of stromal and epithelial elements, but what distinguishes them from fibroadenomas is their stromal overgrowth and propensity to form epithelial lined clefted spaces. The etiological relationship between fibroadenomas and phyllodes
tumors is at present unclear. Many patients develop both lesions either synchronously or metachronously, and histological features of both fibroadenomas and phyllodes tumors have been identified in some tumors;3 however, whether phyllodes tumors develop from fibroadenomas or arise de novo remains to be defined. The peak incidence occurs between 41 and 50 years of age (10–20 years later than the peak incidence of fibroadenomas). They are more prevalent in the Latin-American white and Asian populations.1 Although very rare, phyllodes tumors can also occur in young women. Rajan reviewed 45 cases that occurred in women aged 10–24 years, 34 benign, 11 malignant, and 5 with associated fibroadenomas, they behaved no differently from tumors in older women.4 There is a reported association between malignant phyllodes tumors and nulliparity.5 Phyllodes tumors can be bilateral in approximately 1% of cases.6 Phyllodes tumors are predominantly breast neoplasms but similar lesions can occur in the prostate gland7 and in supernumerary breast tissue in the vulva and axilla.8,9 Phyllodes tumors are thought not to occur in the male breast because of the absence of a lobulo-alveolar structure; however, there have been reports of fibroadenomas and phyllodes tumors arising in men with gynecomastia with lobule development.10
Biology Most phyllodes tumors are benign, but they do show a spectrum of behavior from benign to malignant. The incidence of malignant tumors varies widely in different reported series (Table 1). One of the reasons for this variation is the lack of standard interpretation of the histological features used to define benign and malignant phyllodes tumors. It is generally the stromal component that becomes malignant and can metastasize, resembling sarcoma. The epithelial component of the tumor has been identified in metastases from a phyllodes tumor in only one case.11 The stromal component of these tumors arises from the intralobular fibroblasts of the breast.12 The incidence of metastases varies among the reported series. From Table 1, it can be seen that 5–50% of the
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
210
BREAST CANCER
Table 1 Distribution of benign, borderline, and malignant tumors in different published series and the number of local recurrences and distant metastases in each histological group.
Tumor classification Series Treves and Sunderland Lester and Stout Pietruszka and Barnes Murad et al. Salvadori et al. Bartoli et al. Cohn-Cedermark et al. Grimes Moffat et al. Stebbing et al. Reinfuss et al. Yamada et al. Rajan et al. Mokbel et al. a b
Year of Size of publication series 1951 1953 1978 1988 1989 1990 1991 1992 1995 1995 1996 1997 1998 1999
77 58 42 25 81 106 77 100 32 33 170 118 45 30
Benign 41 28 18 15 28 92 42 51 23 24 92 110 34 21
(54%) (48%) (43%) (60%) (35%) (87%) (55%) (51%) (72%) (73%) (54%) (94%) (75%) (70%)
Number of recurrences
Borderline Malignant
Benign
18 (23%) 10 (17%) 5 (12%) –a 32 (39%) 12 (11%) –a 22 (22%) 4 (13%) 6 (18%) 19 (11%) 4 (3%) 8 (18%) 2 (7%)
4 1 4 4 1 6
18 20 19 10 21 2 35 27 5 3 59 4 3 7
(23%) (35%) (45%) (40%) (26%) (2%) (45%) (27%) (17%) (9%) (35%) (3%) (7%) (23%)
Borderline Malignant
(10%) 4 (4%) 1 (22%) 1 (27%) (4%) 10 (6%) 0 b
14 6 5 4 6 4 2
(27%) (26%) (21%) (4%) (5%) (12%) (10%)
Number of metastases
(22%) (10%) (20%) – (31%)
– 7 (32%) 0 3 (50%) 3 (16%) 0 0 0
10 1 1 6 3 0
(54%) (5%) (5%) (60%) (14%)
7 1 0 7 2 2 3
(26%) (20%)
b
(12%) (50%) (67%) (43%)
Benign
Borderline Malignant
0 1 (4%) 0 0 0 0 4 (10%) 0 0 0 4 (4%) 0 0 0
0 1 (10%) 1 (20%) – 1 (3%) 0 – 3 (14%) 0 0 2 (11%) 0 0 0
9 2 4 4 1 0 12 6 0 0 19 1 1 1
(50%) (10%) (21%) (40%) (5%) (34%) (22%)
(32%) (25%) (33%) (14%)
These studies did not use a borderline category. This data not available from the publication.
malignant tumors in the quoted series have developed metastases. It is also interesting to note that occasionally benign or borderline tumors can become metastatic, and metastases have been reported to occur up to 12 years after initial diagnosis. Metastatic spread is, as for sarcomas of other sites, via the hematogenous route. The commonest sites involved are the lung, pleura, and bone; axillary lymph nodes are rarely involved. Recurrence of these tumors does occur (Table 1) and is thought to relate more to inadequate excision rather than histologic subtype.2 Hajdu showed that 18% of benign and 9% of malignant phyllodes tumors recurred and that recurrence was dependent on the type of surgery performed. This series also showed that recurrence was due to intracanalicular or intracystic extension in 50% of these recurrent cases and recommended that phyllodes should be excised with a generous margin.13 Most recurrent tumors are histologically similar to the primary neoplasms but frequently are more cellular, with focally atypical periductal areas. In Hadju’s series, 2 of 28 recurrences showed malignant transformation (the original tumors appeared entirely benign). This transformation from benign to malignant has not been confirmed by all authors.
Molecular Biology The stroma of phyllodes tumors is generally thought to be the neoplastic element, in view of the progression to sarcoma in some cases. In the past, the epithelium within the tumors was thought to be entirely innocent; however, epithelial hyperplasia is common in benign phyllodes tumors, and carcinoma in situ has been described as well as invasive carcinoma. More recently, both the stroma and epithelium have been shown to contain distinct molecular changes, suggesting that both could contribute to the neoplastic process, at least in the early stages.14 The issue of the clonal origins of the epithelium and stroma in these tumors is intriguing. A cytogenetic study found that both the epithelium and stroma contained the
same chromosomal changes, suggesting that they arise from the same neoplastic clone.15 However, the data on clonality showed that in phyllodes tumors the stroma is monoclonal, in contrast to the majority of fibroadenomas in which the stroma is polyclonal. The epithelium appeared to be polyclonal in both types of lesions.16,17 However, occasionally the stroma in fibroadenomas can be monoclonal, and this suggests that there are lesions initially classified on morphological grounds as fibroadenomas, which evolve into phyllodes tumors. This fits with the clinical finding that some phyllodes tumors arise in preexisting fibroadenomas. The finding of genetic changes in the epithelium and the conclusion from the clonality studies may seem at odds with each other. However, a similar situation has also been described in breast carcinomas, in which the surrounding stroma has been found to harbor genetic changes, some of which are similar to those in the adjacent carcinoma and some of which are not.18 These findings suggest that the interplay between stroma and epithelium in the breast is complex and important in the development of both stromal and epithelial malignancies. This is supported by observations in phyllodes tumors that show that density of stromal mitotic figures correlate with proximity to the epithelium.19 These findings imply that the epithelium exerts some control over proliferation of the stroma. Studies that have looked at growth factor expression in phyllodes tumors suggest that platelet-derived growth factors,20 the Wnt family21 and the fibroblast growth factors22 and their respective receptors are important in this cross talk between the epithelium and the stroma. The insulin-like growth factors, IGF-I and IGF-II are another set of growth factors that have been shown to be highly overexpressed within the stroma of phyllodes tissue and appear to be acting in an autocrine manner.23 There are case reports of phyllodes tumors secreting insulin. These same growth factor pathways also appear to be important in fibroadenomas and from the molecular pathology point of view suggest that fibroadenomas and phyllodes tumors are part of the same biological continuum.
PHYLLODES TUMOR OF THE BREAST
The genetic changes that cause a phyllodes tumor to become malignant are yet to be defined. Studies looking at ploidy show that diploid phyllodes tumors tend to be biologically indolent, whereas most aneuploid tumors are associated with a poor clinical outcome.24 Cytogenetic studies have shown that gain of 1q and loss of 3p are the most frequent chromosomal changes. Gain of 1q material was significantly associated with histologically defined stromal overgrowth and was associated with recurrence.25 Studies of p53 expression have shown that abnormal expression does occur in some borderline and malignant phyllodes tumors and is associated with known negative prognostic factors (tumor grade, prominent stromal overgrowth, prominent stromal nuclear pleomorphism, high stromal mitotic count, and an infiltrative tumor margin). Phyllodes tumors are associated with Li-Fraumeni syndrome, caused by germline p53 mutations, along with breast carcinoma, soft tissue sarcomas, osteosarcoma, brain tumors, adrenocortical carcinoma, and Wilm’s tumors. Although breast carcinoma and sarcomas are numerically most frequent in this syndrome, Birch found that the greatest increases relative to the general population rates were in adrenocortical carcinoma and phyllodes tumor.26 C-myc and c-KIT over-expression have been found to be associated with malignant phyllodes tumors. In some but not all cases, C-myc over-expression was due to gene amplification.27 In conclusion, one can tentatively postulate the following model of phyllodes tumor development. In benign tumors, stromal proliferation to a certain extent is under the control of the epithelium. Although the epithelium promotes stromal growth, it may also limit it in benign tumors, because any excessive stromal growth alters the epithelium: stroma ratio. Epithelial growth, sometimes manifest as hyperplasia, may in turn be promoted by the stroma. Once the stroma of the tumor acquires specific, as yet unknown mutations and becomes malignant, the stromal proliferation becomes autonomous and no longer requires a mitogenic stimulus from the epithelium. This results in a reduction in the epithelium: stroma ratio, as is typically seen in malignant phyllodes tumors, in which the epithelium may be very hard to find or is present only as a single layer of luminal epithelial cells, in contradistinction to the hyperplasia seen in many benign phyllodes tumors.
211
Figure 1 The cut surface appearance of a phyllodes tumor showing the characterized clefted leaf bud-like appearance.
Larger tumors frequently exhibit cystic spaces and foci of hemorrhage, and necrosis may be present.
Histologic Appearances Microscopically, phyllodes tumors are composed of two major elements, clefts lined by epithelial cells and an associated cellular stroma. The epithelial element consists of the usual two layers of myoepithelial and secretory epithelial cells. Focal epithelial hyperplasia is not uncommon, and although this may occasionally be florid, malignant change is extremely rare. Both lobular neoplasia and ductal carcinoma in situ have been recorded;29,30 associated invasive carcinoma is an even greater rarity.31,32 The appearances of the stromal element vary considerably from case to case. Benign Phyllodes Tumors
The stroma is characteristically more cellular than in a fibroadenoma (see Figure 2), the spindle cells do not exhibit
PATHOLOGY Macroscopic Appearances Phyllodes tumors, by their very nature, grow radially, compressing the adjacent breast parenchyma and creating a pseudocapsule through which tongues of phyllodes stroma may protrude and grow into adjacent breast tissue. Macroscopically, phyllodes tumors typically form a firm lobulated mass and vary in size from less than 2 cm to more than 10 cm in diameter, with an average size of 5 cm,2,28 although tumors of more than 20 cm are recorded in many series. They are usually well circumscribed, often with a bosselated contour. The cut surface has a characteristic whorled pattern, resembling a compressed leaf bud (see Figure 1), with visible clefts.
Figure 2 The histologic appearance of a benign phyllodes tumor showing the combination of clefted epithelial-lined spaces and hypercellular stroma.
212
BREAST CANCER
pleomorphism, and mitotic counts were less than 10 per 10 fields (field area 0.152 mm). Malignant tumors (five cases, 16%) had infiltrative margins, marked stromal overgrowth, increased cellularity and pleomorphism, with mitotic counts greater than 10 per 10 fields. Five additional cases with intermediate features were classified as borderline. None of these histologic factors were useful in predicting local recurrence, which was strongly related to completeness of local excision. All recurrences in benign tumors (four cases) occurred in patients treated initially with local excision and were controlled with complete local reexcision or mastectomy. There were no recurrences or metastases in malignant tumors treated initially with mastectomy but there were uncontrolled chest wall recurrences in one patient treated with local excision. Figure 3 A malignant phyllodes tumor showing stromal overgrowth with stromal cell atypia and increased mitotic frequency.
pleomorphism, and mitoses are infrequent. The presence of occasional bizarre stromal giant cells does not indicate malignant change.33 Malignant Phyllodes Tumors
At the other end of the spectrum, a minority of tumors will show frankly sarcomatous change, characterized by stromal overgrowth and hypercellularity (see Figure 3), nuclear atypia, and an increased mitotic count. In particular, specific patterns such as rhabdomyosarcoma, chondrosarcoma, and osteosarcoma are clear indicators of malignancy. Borderline Phyllodes Tumors
There remains an intermediate group of tumors with appearances that pose problems for the pathologist and clinician in predicting the likelihood of local recurrence and metastatic malignant potential. A number of studies2,13,34 – 37 have addressed this question since Norris and Taylor38 first suggested that an infiltrative rather than a pushing margin, cellular atypia, increased mitotic count and large size all favored malignancy. No coherent pattern emerges from these studies because of the small number of cases assessed, the varied database (most series being derived from secondary or tertiary referral centers), and the division into two groups, benign and malignant, by some and three groups, benign, borderline, and malignant by others. However, the majority of studies have confirmed the importance of the criteria proposed by Norris and Taylor38 and refined by Pietruszka and Barnes.2 In one of the few studies based on a community population rather than referred cases,28 and using semiquantitative criteria derived from those proposed by Pietruszka and Barnes,2 Ward and Evans36 divided phyllodes tumors into three groups: benign, borderline, and malignant. The histologic characteristics used included its margin (pushing or infiltrating), stromal cellularity (mild, moderate, or severe), stromal overgrowth (mild, moderate, or severe), tumor necrosis (present or absent), cellular atypia (mild, moderate, or severe), and the number of mitoses per high-power field. In benign tumors (22 cases, 68%), the margins were pushing, there was minimal stromal overgrowth, cellularity and
Differential Diagnosis In most cases, the distinction from fibroadenoma is straightforward, but difficulties may be encountered with small lesions. A leaflike pattern and hypercellular stroma favor a diagnosis of phyllodes tumor, but size alone cannot be used as a distinguishing feature. Fibromatosis, primary sarcoma, and metaplastic carcinoma of the breast, although very rare, must be distinguished from malignant phyllodes tumors. Again, the mixture of elements, with the leaflike epithelial clefts, is a feature of phyllodes tumor not seen in the other lesions. Metaplastic carcinoma may be distinguished by positive immunostaining of the spindle cells with epithelial markers such as broad-spectrum cytokeratins.
Sarcomatous Overgrowth Sarcomatous differentiation is a rare but important event in phyllodes tumors.39 Guerrero has reviewed the literature back to 1979 and identified a total of 30 documented cases.39 The most common histologic type is liposarcoma, accounting for a total of 23 cases (adipose differentiation ranged from mature fat to liposarcoma). Malignant Fibrous Histiocytoma (MFH) was identified in three cases, rhabdomyosarcoma and chondrosarcoma in one case each, and by multiple histologic types. Powell et al.40 studied 14 patients with adipose differentiation and concluded that malignant phyllodes tumor with adipose differentiation can be graded histologically on the basis of this component and that despite the high-grade histology in some instances, the patients had an excellent prognosis when the tumors were completely excised. Because of the rarity of this event, little is known about the clinical behavior, treatment, and prognosis of such tumors. However, others have observed that most clinically malignant phyllodes tumors reported in the literature that have metastasized have had overgrowth of an obvious sarcomatous element on histologic examination.41 This malignant element has often been something other than low-grade fibrosarcoma (e.g. high-grade sarcoma, liposarcoma, rhabdomyosarcoma). Therefore, close examination of the stroma with multiple sections is mandatory. The truly malignant phyllodes tumor may be present only in a portion of the phyllodes tumor.
Use of Biological Markers The usefulness of markers (Ki67, p53, epidermal growth factor receptor families, c-KIT, platelet-derived growth factor,
PHYLLODES TUMOR OF THE BREAST
and numerous other markers) as predictors of clinical behavior in phyllodes tumors also has been studied but none has been able to predict unfavorable outcomes.41 At most, these markers correlate with histologic grade, which we already know is faulty at predicting which patients are at lethal risk, and at worst the markers are expressed in all phyllodes tumors, with little association with clinical behavior. At present, strict histologic assessment and generous sampling of these tumors provide the most clinically useful information.
CLINICAL PRESENTATION AND DIAGNOSIS Presentation Phyllodes tumors account for 0.3–0.9% of all breast tumors, up to 3% of fibroepithelial lesions and occurs over a wide age range (median age 45). These lesions (both benign and malignant) typically present as a symptomatic breast (or rarely axillary) lump. Of the 40 patients reviewed by Mangi et al.,42 37 presented with a palpable mass, 7 complained of pain, 3 were detected on mammographic screening. Similar to this study, many other series reflect the majority of patients presenting with a palpable mass. On examination, they are smooth, mobile, and well circumscribed (often diagnosed as a fibroadenoma). A clue to it being a phyllodes tumor may be suspected if there is a recent history of rapid increase in size. This feature of rapid growth may lead to stretched or even ulcerated overlying skin. It is important to note that such “suspicious” presenting signs do not necessarily predict the behavior of the tumor. Nonetheless, as is the case with all breast lumps, they should undergo a triple assessment to rule out malignant change. Phyllodes tumors are found more commonly in the upper outer quadrant with an equal propensity to occur in either breast. Fixation to the skin or the pectoralis muscles has been reported, but ulceration is uncommon, even in patients with histologically malignant tumors. Pressure necrosis of the overlying skin can occur. A significant proportion of patients have previously had a fibroadenoma and in a minority these have been multiple. Occasionally, fibroadenomas present simultaneously with a phyllodes tumor and synchronous bilateral phyllodes tumors have been reported. The diagnosis of a phyllodes tumor should be considered in all women, particularly over the age of 35, who present with a rapidly growing but clinically benign breast lump. Malignant phyllodes tumor is not known to metastasize to lymph nodes but rather hematogenously. Thus, unlike carcinomas, evaluation of the axilla beyond clinical examination is not necessary. Palpable lymph nodes may be elicited on presentation in up to 20% of patients but this is thought to be a feature of tumor necrosis. Carcinoma arising within a phyllodes tumor has been reported and in one case report there was metastasis of the lymph node but it was from the carcinomatous component.43 Even though the male breast lacks the lobulo-alveolar complex, there have been isolated case reports of phyllodes tumor arising within the male breast. The literature is scant, with such data limited to isolated case reports; thus it is
213
difficult for us to draw definitive conclusions on its natural history in males.
Diagnosis The major diagnostic challenge in making a preoperative imaging diagnosis is that both mammography and ultrasonography do not reliably distinguish phyllodes tumors from fibroadenomas.1 In mammography, both will appear as welldefined masses with smooth occasionally lobulated borders. A radiolucent “halo” may be seen around the lesion, due to compression of the surrounding breast stroma. Such features can even mimic a well-circumscribed carcinoma. No mammographic indicators have been identified that allow differentiation between benign and malignant tumors. Preoperative diagnosis using ultrasonography is also known to be disappointing. Phyllodes tumors often show smooth contours with low-level homogeneous internal echoes, intramural cysts, and the absence of posterior acoustic enhancement. Certain features could raise suspicion of the mass being a phyllodes tumor such as smooth walls, low-level internal echoes and echo-free margins. Recently, modalities such as magnetic resonance imaging (MRI), color flow Doppler and even vascular embolization have been employed with varying degrees of success. The data supporting their use are limited to either case reports or small series.44 Many studies have employed fine needle aspiration cytology (FNAC) as part of their triple assessment.45,46 However, in the assessment of these lesions FNAC has significant limitations. Both phyllodes tumors and fibroadenomas belong to a spectrum of fibroepithelial lesions, hence accurate cytological diagnosis of phyllodes tumors by fine needle aspiration can be difficult. Cytologically, it is often easier to differentiate benign from malignant phyllodes tumors than to separate benign phyllodes tumors from fibroadenomas. The use of core or excision biopsy invariably provides a superior result for tissue diagnosis as the architecture of the dual epithelia and stromal components as well as their cytomorphological characteristics can be assessed. Core biopsy is now the nonoperative sampling method of choice. It should be borne in mind that the histologic characteristic of a phyllodes tumor can vary in different regions and a core biopsy sample, which is of limited size, may not include all parts of the lesion. If there is clinical suspicion of a phyllodes tumor, formal excision is recommended to allow comprehensive assessment of its characteristics.
TREATMENT Surgery Margin of excision, especially incomplete excision, is the principal cause of local recurrence.28,47 For benign lesions, surgery should be guided by the size of the lesion and the breast. Small lesions can safely undergo wide local excision with a 1- to 2-cm margin. As some of these patients are young, cosmesis should be an important aspect of the surgical decision-making process. Oncoplastic procedures such as partial mastectomy and latissimus dorsi flaps and envelope mastectomy with reconstruction have shown excellent
214
BREAST CANCER
results. Mastectomy should be reserved for larger lesions where the breast is too small or achieving an acceptable clearance margin provides a poor cosmetic outcome (i.e. size of lesion vs size of breast). There is at present no role for axillary staging sampling or clearance, and we do not envisage a role for sentinel node biopsy. Because excision of the tumor and not surrounding breast tissue is the goal, this provides an opportunity, even in large tumors, for extensive skin and nipple sparing. If breastconserving surgery is used and an adequate margin is not achieved, some authors argue against immediate reexcision.48 Timely reexcision of breast recurrences is more appropriate in that situation, requiring close clinical surveillance. However, overwhelmingly, the published data clearly show that achieving a tumor-free margin is imperative to achieve the best local regional control. Moffat et al. showed that in 32 women with a median follow-up of 135 months, 6 had local recurrence and all possessed involved margins.28 Mangi et al.42 reported on 40 patients and again showed a significant correlation between local recurrence and involved margins (p < 0.05). The average interval between the diagnosis of primary benign and recurrent benign phyllodes tumors is 2 years,13 but several years may lapse between apparent recurrences. Incompletely treated multiplicity is a likely cause of so-called recurrence in some settings, as it is for fibroadenoma.41 In the advent of local recurrence for benign phyllodes, a reexcision with wider margins is recommended. High-grade (malignant) phyllodes tumors show a strong propensity to recur or metastasize if margins are involved. Kapiris et al.49 studied 45 women with Phyllodes tumor. Of 24 patients treated conservatively, local recurrence occurred in 40% and distant metastasis in 27%. Both events were related to margin status.49 Although the trend for so-called malignant phyllodes tumor has been to obtain mastectomy, in Barth’s review47 there was no difference in survival for women with histologically malignant phyllodes tumor based on type of surgical treatment. He concluded that survival was not impaired by breast-conserving surgery. In summary, clear margin status is associated with a low local recurrence rate regardless of histologic features of the lesion.50 We recommend ensuring wide local excision even if this means mastectomy, especially in cases of malignant phyllodes tumor. These recommendations are similar to the algorithm proposed by Buchanan51 for management of phyllodes tumor consisting of needle biopsy, tumor excision with adequate margins and testing for margin involvement and simple mastectomy for tumors greater than 5 cm or those found to be borderline or malignant.
Radiotherapy There are no reported randomized controlled trials looking at the use of radiotherapy in phyllodes tumors because of the rarity of the disease. However, there are case reports of phyllodes tumors responding to radiotherapy. Following wide local excision, postoperative radiotherapy is currently not recommended routinely, as it has not been shown in a retrospective series to reduce recurrence rates.37,50 There
is a prospective phase II study presently recruiting in the United States to try and answer this question in patients with borderline and malignant phyllodes tumors and negative margins. Radiotherapy may have a role in preventing local recurrence if negative margins cannot be achieved.52 We would therefore recommend offering radiotherapy to those patients where clear margins cannot be achieved surgically, as these patients are at high risk of local recurrence. For patients with malignant, inoperable local recurrence or metastases, radiotherapy may offer some palliation and occasional complete responses are documented.53,54
Chemotherapy As in more conventional sarcomas, the efficacy of chemotherapy in these tumors is limited but can be useful for treating symptomatic metastases. Chemotherapy is based on the guidelines for the treatment of sarcomas rather than breast carcinomas. The most commonly used drugs are those used for the treatment of soft tissue sarcomas outside the breast, adriamycin and ifosfamide.54,55 The role of routine postoperative chemotherapy in malignant phyllodes tumors is not established. Because of the rarity of the disease, no large randomized controlled trials have been performed. The only randomized study looking at this question randomized 28 patients with malignant phyllodes tumors to receive adjuvant doxorubicin and dacarbazine or observation. Adjuvant chemotherapy did not impact on survival.56 On the basis of this data, we at present do not recommend adjuvant chemotherapy for malignant phyllodes tumors.
Hormonal and Biological Therapy In the past, when biochemical assays were performed to assess estrogen and progesterone receptor status, it was thought that phyllodes tumors were progesterone receptor positive. This led to the suggestion that hormonal therapy may be useful in the treatment of advanced, malignant phyllodes tumors. However, four reports57 have all shown that estrogen and progesterone receptors are expressed only in glandular elements of phyllodes tumors and not in the stroma. It is therefore unlikely that these types of drugs will be useful in the treatment of malignant phyllodes tumors. Recent increased understanding of the molecular biology of phyllodes tumors brings the potential for use of targeted biological therapy. c-KIT is a proto-oncogene that encodes a tyrosine kinase receptor (CD117) and is a marker for gastrointestinal stromal tumors (GIST). The success of the therapeutic agent imatinib (Glivec) targeted at this receptor for GIST has led to examination of CD117 expression in other tumors including phyllodes tumor. Increased expression of CD117 has been identified in mammary phyllodes tumors and increasing c-KIT expression found with increasing degree of malignancy, up to 46% in malignant cases58 and has led the authors to speculate that the therapeutic agent, imatinib (Glivec), may be a potentially useful drug for its management.
PHYLLODES TUMOR OF THE BREAST
Characteristics and Management of Recurrent Disease Parker et al. have reviewed the characteristics associated with local recurrence.1 Local recurrence rates range from 10 to 40%, with an average of 15%; local recurrence appears to be related to the extent of the initial surgery and should be regarded as a failure of primary surgical treatment. Whether malignant tumors have an increased risk of recurrence is unclear but when it does occur it is invariably seen earlier than with benign tumors; local recurrence usually occurs within the first few years of surgery and histologically resembles the original tumor. Occasionally, recurrent tumors show increased cellularity and more aggressive histologic features than the original lesion. In most patients, local recurrence is isolated and is not associated with the development of distant metastases; in a minority of patients repeated local recurrence occurs over a prolonged period with no survival disadvantage and this is often seen irrespective of either the histologic type of the tumor or the extent of the specimen margins. They concluded that local recurrence can usually be controlled by further wide excision and mastectomy is not invariably required but should be considered for local recurrence after local surgery for borderline or malignant tumors. Occasionally, aggressive local recurrence can result in widespread chest wall disease with direct invasion of the underlying lung parenchyma, and isolated reports of good palliation in this situation with radiotherapy have been published.
Characteristics and Management of Metastatic Disease Parker et al. have also recently reviewed the characteristics associated with metastatic disease.1 Overall, less than 10% of patients with phyllodes tumors develop distant metastases, and this phenomenon is more common (approximately 20%) in patients with histologically malignant tumors. Most distant metastases develop without evidence of local recurrence. The commonest sites for distant metastases are the lung, bone, and abdominal viscera. These often occur in the absence of lymph node metastases and histologically contain only the stromal element. The risk of metastatic disease does not appear to be influenced by the extent of the initial surgery and appears to be predetermined by tumor biology. Few reports have been published about distant metastases after excision of a benign phyllodes tumor, except when histologically malignant local recurrence occurs. Metastatic phyllodes tumors have a poor prognosis and no longterm survival has been reported. Isolated reports have been published about palliation of metastatic disease from singleagent or combination chemotherapy, but the exact role of chemotherapy in metastatic phyllodes tumors remains to be defined.
PROGNOSIS Most malignant phyllodes tumors do not recur or metastasize while some histologically benign tumors can show a usually aggressive clinical course. Consequently, it has been
215
suggested that all phyllodes tumors should be regarded with malignant potential. In view of this relatively indolent behavior of the majority, radical surgery for all phyllodes tumors is unnecessary and attempts have been made to identify clinical and histopathological prognostic factors. Little consensus exists regarding their relative importance, and different factors appear to be important in predicting local recurrence and metastatic spread. No reliable clinical prognostic factors other than incompleteness of excision have been identified that predict local recurrence. Patient age does not appear to be important but tumors presenting in adolescence do seem to be less aggressive irrespective of their histologic type.1 The role of tumor size is unclear. While most series have reported low local recurrence rates with tumors less than 2 cm in diameter, no correlation between tumor size and the risk of local recurrence has been demonstrated.1 The risk of local recurrence is increased in incompletely excised lesions. Conversely, low recurrence rates have been reported in patients in whom histologically clear margins have been assured. Whether the risk of local recurrence is increased in histologically malignant phyllodes tumors is unclear. Most distant metastases develop from borderline or malignant tumors. Unlike in local recurrence, tumor size does appear to be an important factor in predicting metastatic spread. Many histologic prognostic factors have been evaluated but no clear consensus exists for their role in the assessment of prognosis (see “Pathology” section above). Barth47 performed a meta-analysis and demonstrated that strict adherence to histologic criteria showed no distant metastases in lesions that were categorized as histologically benign or borderline. The use of a borderline category allows placement of tumors that usually do not act in a malignant fashion but may be more likely to recur locally.41 Of great importance is that local recurrences are unlikely to evolve into malignancy if this feature was not present in the primary tumor.28,41 But there is some evidence that any large or second recurrence may be life threatening.13 However, most local recurrences can be treated successfully with surgical reexcision. These series emphasize the need for clinical vigilance in follow-up of the preserved breast and for mandatory careful histologic sampling of these tumors. Recently, Carter and Page41 have supported the use of three histologic categories. Some phyllodes tumors, although rare, are clinically malignant at presentation and have the ability to metastasize. These usually have a clear focus of easily recognizable sarcoma on histologic exam. The vast majority of phyllodes tumors are benign and potentially curable by local excision when completely removed. A third subset of phyllodes tumors of indeterminate biology, some of which recur and a very few of which metastasize, also exists. It is this third group that probably makes up at least a slight majority of those tumors that were previously diagnosed as malignant but do not behave as such. In light of this knowledge, it seems unjustifiable to give a fully malignant diagnostic term to lesions that only occasionally or rarely kill without first recurring. They emphasize that the importance of the borderline malignant category is to prevent overdiagnosis of malignancy.41 By placing some phyllodes tumors in a borderline malignant category, it will
216
BREAST CANCER
guide surgical colleagues to the need to achieve clearance and reduce the risk of recurrence in a particular patient.
KEY POINTS AND AUTHORS’ RECOMMENDATIONS • Phyllodes tumors are defined as fibroepithelial lesions of the breast, which show epithelial-lined clefted spaces with surrounding hypercellular stroma. • They can be classified as benign, borderline, or malignant on the basis of histologic features. • Phyllodes tumors account for less than 1% of breast neoplasms in contrast to fibroadenomas which are common. • There is no distinct boundary between fibroadenoma and phyllodes tumor at the benign end of the spectrum and both can be excised with breast conservation. • No imaging techniques appear to be able to differentiate benign and malignant lesions. • Core biopsy is the preferred nonoperative sampling technique. • The majority of phyllodes tumors behave in a completely benign fashion in that they do not have metastatic potential, local recurrence being the only real concern. • Local recurrence per se is not an indicator of malignancy because it has been described in benign, borderline, and malignant phyllodes tumors. • Clear margin status is associated with a low local recurrence rate regardless of histologic features of the lesion. We recommend ensuring wide local excision even if this means mastectomy, especially in cases of borderline and malignant phyllodes tumor. • Histologic type is the most important predictor for metastatic spread; in particular, the presence of high-grade sarcoma, which even in limited proportions appears predictive of aggressive behavior. • Twenty percent of patients with malignant tumors develop distant metastases, the most common sites being the lung, bone, and abdominal viscera. • The 5-year survival for benign, borderline, or malignant tumor is 96%, 74%, and 66% respectively. • Wide excision or mastectomy should be performed, ensuring histologic clear margins. • Axillary nodal dissection is not indicated in the management of phyllodes tumors. • The role of adjuvant radiotherapy and hormonal therapy remains to be defined. • Following development of metastatic disease, palliation may be obtained with single-agent or combination chemotherapy (anthracycline-based regimens).
REFERENCES 1. Parker SJ, Harries SA. Phyllodes tumours. Postgrad Med J 2001; 77(909): 428 – 35. 2. Pietruszka M, Barnes L. Cystosarcoma phyllodes: a clinicopathologic analysis of 42 cases. Cancer 1978; 41(5): 1974 – 83. 3. Treves N, Sunderland DA. Cystosarcoma phyllodes of the breast: a malignant and a benign tumor; a clinicopathological study of seventyseven cases. Cancer 1951; 4(6): 1286 – 332.
4. Rajan PB, Cranor ML, Rosen PP. Cystosarcoma phyllodes in adolescent girls and young women: a study of 45 patients. Am J Surg Pathol 1998; 22(1): 64 – 9. 5. Lindquist KD, et al. Recurrent and metastatic cystosarcoma phyllodes. Am J Surg 1982; 144(3): 341 – 3. 6. Rosenfeld JC, DeLaurentis DA, Lerner H. Cystosarcoma phyllodes. Diagnosis and management. Cancer Clin Trials 1981; 4(2): 187 – 93. 7. De Raeve H, et al. Cystosarcoma phyllodes of the prostate with rhabdomyoblastic differentiation. Pathol Res Pract 2001; 197(10): 657 – 62. 8. Chulia MT, et al. Phyllodes tumor in ectopic breast tissue of the vulva. Int J Surg Pathol 2001; 9(1): 81 – 3. 9. Oshida K, et al. Phyllodes tumor arising in ectopic breast tissue of the axilla. Breast Cancer 2003; 10(1): 82 – 4. 10. Ansah-Boateng Y, Tavassoli FA. Fibroadenoma and cystosarcoma phyllodes of the male breast. Mod Pathol 1992; 5(2): 114 – 6. 11. Kracht J, Sapino A, Bussolati G. Malignant phyllodes tumor of breast with lung metastases mimicking the primary. Am J Surg Pathol 1998; 22(10): 1284 – 90. 12. Atherton AJ, et al. Dipeptidyl peptidase IV expression identifies a functional sub-population of breast fibroblasts. Int J Cancer 1992; 50(1): 15 – 9. 13. Hajdu SI, Espinosa MH, Robbins GF. Recurrent cystosarcoma phyllodes: a clinicopathologic study of 32 cases. Cancer 1976; 38(3): 1402 – 6. 14. Sawyer EJ, et al. Molecular analysis of phyllodes tumors reveals distinct changes in the epithelial and stromal components. Am J Pathol 2000; 156(3): 1093 – 8. 15. Dietrich CU, et al. Cytogenetic findings in phyllodes tumors of the breast: karyotypic complexity differentiates between malignant and benign tumors. Hum Pathol 1997; 28(12): 1379 – 82. 16. Noguchi S, et al. Clonal analysis of fibroadenoma and phyllodes tumor of the breast. Cancer Res 1993; 53(17): 4071 – 4. 17. Kuijper A, et al. Analysis of the progression of fibroepithelial tumours of the breast by PCR-based clonality assay. J Pathol 2002; 197(5): 575 – 81. 18. Kurose K, et al. Genetic model of multi-step breast carcinogenesis involving the epithelium and stroma: clues to tumour-microenvironment interactions. Hum Mol Genet 2001; 10(18): 1907 – 13. 19. Sawhney N, et al. Epithelial – stromal interactions in tumors. A morphologic study of fibroepithelial tumors of the breast. Cancer 1992; 70(8): 2115 – 20. 20. Feakins RM, et al. p53 expression in phyllodes tumours is associated with histological features of malignancy but does not predict outcome. Histopathology 1999; 35(2): 162 – 9. 21. Sawyer EJ, et al. The Wnt pathway, epithelial-stromal interactions, and malignant progression in phyllodes tumours. J Pathol 2002; 196(4): 437 – 44. 22. La Rosa S, et al. Expression of acidic fibroblast growth factor (aFGF) and fibroblast growth factor receptor 4 (FGFR4) in breast fibroadenomas. J Clin Pathol 2001; 54(1): 37 – 41. 23. Sawyer EJ, et al. Beta-catenin abnormalities and associated insulin-like growth factor overexpression are important in phyllodes tumours and fibroadenomas of the breast. J Pathol 2003; 200(5): 627 – 32. 24. Niezabitowski A, et al. Prognostic evaluation of proliferative activity and DNA content in the phyllodes tumor of the breast: immunohistochemical and flow cytometric study of 118 cases. Breast Cancer Res Treat 2001; 65(1): 77 – 85. 25. Lu YJ, et al. Phyllodes tumors of the breast analyzed by comparative genomic hybridization and association of increased 1q copy number with stromal overgrowth and recurrence. Genes Chromosomes Cancer 1997; 20(3): 275 – 81. 26. Birch JM, et al. Relative frequency and morphology of cancers in carriers of germline TP53 mutations. Oncogene 2001; 20(34): 4621 – 8. 27. Sawyer EJ, et al. Malignant phyllodes tumours show stromal overexpression of c-myc and c-kit. J Pathol 2003; 200(1): 59 – 64. 28. Moffat CJ, et al. Phyllodes tumours of the breast: a clinicopathological review of thirty-two cases. Histopathology 1995; 27(3): 205 – 18. 29. Knudsen PJ, Ostergaard J. Cystosarcoma phyllodes with lobular and ductal carcinoma in situ. Arch Pathol Lab Med 1987; 111(9): 873 – 5.
PHYLLODES TUMOR OF THE BREAST 30. Grove A, Deibjerg Kristensen L. Intraductal carcinoma within a phyllodes tumor of the breast: a case report. Tumori 1986; 72(2): 187 – 90. 31. Grimes MM. Cystosarcoma phyllodes of the breast: histologic features, flow cytometric analysis, and clinical correlations. Mod Pathol 1992; 5(3): 232 – 9. 32. Leong AS, Meredith DJ. Tubular carcinoma developing within a recurring cystosarcoma phyllodes of the breast. Cancer 1980; 46(8): 1863 – 7. 33. Powell CM, Cranor ML, Rosen PP. Multinucleated stromal giant cells in mammary fibroepithelial neoplasms. A study of 11 patients. Arch Pathol Lab Med 1994; 118(9): 912 – 6. 34. Hart WR, Bauer RC, Oberman HA. Cystosarcoma phyllodes. A clinicopathologic study of twenty-six hypercellular periductal stromal tumors of the breast. Am J Clin Pathol 1978; 70(2): 211 – 6. 35. Murad TM, et al. Histopathological and clinical correlations of cystosarcoma phyllodes. Arch Pathol Lab Med 1988; 112(7): 752 – 6. 36. Ward RM, Evans HL. Cystosarcoma phyllodes. A clinicopathologic study of 26 cases. Cancer 1986; 58(10): 2282 – 9. 37. Cohn-Cedermark G, et al. Prognostic factors in cystosarcoma phyllodes. A clinicopathologic study of 77 patients. Cancer 1991; 68(9): 2017 – 22. 38. Norris HJ, Taylor HB. Relationship of histologic features to behavior of cystosarcoma phyllodes. Analysis of ninety-four cases. Cancer 1967; 20(12): 2090 – 9. 39. Guerrero MA, Ballard BR, Grau AM. Malignant phyllodes tumor of the breast: review of the literature and case report of stromal overgrowth. Surg Oncol 2003; 12(1): 27 – 37. 40. Powell CM, Rosen PP. Adipose differentiation in cystosarcoma phyllodes. A study of 14 cases. Am J Surg Pathol 1994; 18(7): 720 – 7. 41. Carter BA, Page DL. Phyllodes tumor of the breast: local recurrence versus metastatic capacity. Hum Pathol 2004; 35(9): 1051 – 2. 42. Mangi AA, et al. Surgical management of phyllodes tumors. Arch Surg 1999; 134(5): 487 – 92; discussion 492 – 3. 43. Parfitt JR, et al. In-situ and invasive carcinoma within a phyllodes tumor associated with lymph node metastases. World J Surg Oncol 2004; 2(1): 46. 44. Rao CR, et al. Cystosarcoma phyllodes. Diagnosis by fine needle aspiration cytology. Acta Cytol 1992; 36(2): 203 – 7.
217
45. Chao TC, et al. Phyllodes tumors of the breast. Eur Radiol 2003; 13(1): 88 – 93. 46. Kinoshita T, et al. Magnetic resonance imaging of benign phyllodes tumors of the breast. Breast J 2004; 10(3): 232 – 6. 47. Barth RJ Jr. Histologic features predict local recurrence after breast conserving therapy of phyllodes tumors. Breast Cancer Res Treat 1999; 57(3): 291 – 5. 48. Zurrida S, et al. Which therapy for unexpected phyllode tumour of the breast? Eur J Cancer 1992; 28(2 – 3): 654 – 7. 49. Kapiris I, et al. Outcome and predictive factors of local recurrence and distant metastases following primary surgical treatment of high-grade malignant phyllodes tumours of the breast. Eur J Surg Oncol 2001; 27(8): 723 – 30. 50. Chaney AW, et al. Primary treatment of cystosarcoma phyllodes of the breast. Cancer 2000; 89(7): 1502 – 11. 51. Buchanan EB. Cystosarcoma phyllodes and its surgical management. Am Surg 1995; 61(4): 350 – 5. 52. Chaney AW, et al. Adjuvant radiotherapy for phyllodes tumor of breast. Radiat Oncol Investig 1998; 6(6): 264 – 7. 53. Eich PD, et al. Diagnostic radiation oncology: malignant cystosarcoma phylloides. Strahlenther Onkol 2000; 176(4): 192 – 5. 54. Paulsen F, et al. Cystosarcoma phyllodes malignum: a case report of a successive triple modality treatment. Int J Hyperthermia 2000; 16(4): 319 – 24. 55. Hawkins RE, et al. Ifosfamide is an active drug for chemotherapy of metastatic cystosarcoma phyllodes. Cancer 1992; 69(9): 2271 – 5. 56. Morales-Vasquez F, et al. A randomized study of adjuvant chemotherapy for malignant phyllodes tumours of the breast. Proceedings of 2005 ASCO annual meeting. J Clin Oncol 2005; 23(16S): Abstract 574. 57. Umekita Y, Yoshida H. Immunohistochemical study of hormone receptor and hormone-regulated protein expression in phyllodes tumour: comparison with fibroadenoma. Virchows Arch 1998; 433(4): 311 – 4. 58. Tse GM, et al. Increased c-kit (CD117) expression in malignant mammary phyllodes tumors. Mod Pathol 2004; 17(7): 827 – 31.
Section 4 : Breast Cancer
18
Carcinosarcoma of the Breast
B.T. Hennessy, M.Z. Gilcrease, G. Babiera, W. Yang, V. Valero and G.N. Hortobagyi
INTRODUCTION Carcinosarcoma of the breast is rare and accounts for less than 0.1% of all breast malignancies.1,2 Carcinosarcomas have traditionally been distinguished from other forms of metaplastic breast carcinoma because they are aggressive and have an ominous prognosis; they consist of intraductal or infiltrating carcinoma contiguous or subtly merged with a highly cellular, often mitotically active malignantappearing stroma (sarcomatous component).3 – 6 Their macroscopic appearance is dependent on the variety of possible epithelial and mesenchymal components. By traditional definition, the demarcation between carcinomatous and sarcomatous components should be distinct in all microscopic fields, although the term “carcinosarcoma” has been used inconsistently in the literature to describe both “classic” carcinosarcomas (see Figure 1) and other metaplastic breast sarcomatoid carcinomas including adenocarcinomas with varying degrees of sarcomatoid metaplasia.7,8 However, if there is obvious microscopic transition between the epithelial and sarcomatoid components, then the term “biphasic sarcomatoid carcinoma” has often been employed to distinguish the tumor from a “classic” carcinosarcoma. Despite this, tumors that show both carcinomatous and sarcomatous features occur in various anatomical sites and, in spite of differences in terminology and exact microscopic composition, evidence suggests these are all similar tumors developing through a peculiar phenotypic transformation of carcinoma cells into sarcoma.9 – 19 In the breast, the most popular theories regarding the histogenesis of the sarcomatous component of carcinosarcomas and sarcomatoid carcinomas propose the malignant transformation of myoepithelial cells or myofibroblastic metaplasia of malignant epithelial cells.7,8,20 – 27
HISTORICAL BACKGROUND The issues of the histogenesis and clonality of carcinosarcomas of the breast have been debated for almost 100 years.28 In fact, a search of the PubMed database reveals several articles dating back to the 1950s presenting case reports, reviewing the literature, and discussing the origins of this rare tumor.29 – 32 For years, a prevailing notion was that breast
carcinosarcomas, based on their microscopic appearance with demarcation between the epithelial and sarcomatous components, are derived from two distinct cells of origin reflecting two different malignant processes (the “collision” hypothesis). However, research over the past 10–15 years has unearthed rather compelling evidence to the contrary. Wargotz et al. demonstrated in the 1980s that the sarcomatous component of carcinosarcoma is immunoreactive for cytokeratin in more than 50% of cases, regardless of the presence or absence of overt transitional areas between sarcomatous and epithelial areas.4 In fact, they found that a majority of carcinosarcomas diagnosed using the traditional definition of a distinct demarcation between the carcinomatous and sarcomatous components contain, upon close inspection, foci of subtle transition between both components. These investigators were among the first to suspect a myoepithelial origin for these tumors, in part because of their demonstration of dual staining with epithelial and myoepithelial markers, including actin and S-100. A review of current ultrastructural, cytogenetic, and immunohistochemical data in fact supports a monoclonal origin for carcinosarcomas.
EPIDEMIOLOGY Carcinosarcoma of the breast is rare. Metaplastic breast cancers constituted 0.2% of all breast neoplasms seen at The University of Texas MD Anderson Cancer Center between 1948 and 1978. In the Surveillance, Epidemiology, and End Results (SEER) database, a population-based tumor registry sponsored by the American National Cancer Institute consisting of tumor registries that collect information on all newly diagnosed cancer cases that occur in persons residing in 11 SEER participating areas, of a total of 281 342 breast cancers registered between 1988 and 2001, there were 98 carcinosarcomas (0.03%; Table 1) and an additional 213 registered cases of metaplastic sarcomatoid carcinoma (0.08%). Of the 98 registered cases of breast carcinosarcoma, 83 occurred in Caucasians, 11 in African Americans, and 4 in other races. The etiology of breast carcinosarcomas is not known but chronic estrogenic stimulation is probably involved, as with other breast cancers. In the uterus, tamoxifen has
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
CARCINOSARCOMA OF THE BREAST
(a)
(b)
(c)
(d)
219
Figure 1 (a,b) Low-power photomicrographs of two different carcinosarcomas demonstrating invasive carcinoma (solid arrows) and distinct high-grade sarcomatous component (open arrows). (c) High-power photomicrograph of same tumor in (b). (d) Carcinosarcoma with ductal carcinoma in situ (DCIS, solid arrow) and adjacent high-grade sarcomatous component (open arrow).
been implicated in the development of carcinosarcomas and adenocarcinomas.39 Carcinosarcoma of the breast has also been reported to occur in association with breast augmentation by liquid silicone injection.40
PATHOLOGY Azzopardi originally proposed the term “carcinosarcoma” of the breast for rare tumors containing discrete areas of carcinoma and sarcoma arising together within a fibroepithelial lesion.41 He assumed that most tumors with both carcinomatous and sarcomatous elements were sarcomatoid carcinomas, and that a strict definition of carcinosarcoma was required to correctly identify those tumors with true carcinomatous and sarcomatous components. To understand Azzopardi’s concept of carcinosarcoma, it is important to first have an appreciation of what constitutes fibroepithelial lesions of the breast.3 Fibroepithelial lesions are neoplasms that contain both epithelial and stromal components. They include fibroadenomas and the full range of benign, borderline, and malignant phyllodes tumors. In fibroadenomas, the epithelial and stromal components together form a wellcircumscribed mass lesion without malignant potential. Both epithelial and stromal elements lack cytologic atypia; there is only mild to moderate stromal cellularity, with an absence of infiltrative growth. Phyllodes tumors similarly contain an admixture of ductal and stromal elements, but the stromal component is more prominent and forms a characteristic architectural pattern as a result of stromal expansion and invagination of contiguous ductal structures. The distinction between benign, borderline, and malignant phyllodes tumors depends on the relative cellularity, cytologic atypia, mitotic rate, and infiltrative growth pattern of the stromal component.
Phyllodes tumors are true biphasic lesions, yet the epithelial component generally does not undergo malignant transformation. The stromal component of malignant phyllodes tumors undergoes sarcomatous transformation and becomes the dominant component, eventually obscuring the ductal elements. Only in rare cases do the interspersed ducts develop ductal carcinoma in situ (DCIS) or invasive carcinoma. In these rare cases, the carcinomatous and sarcomatous components of the tumor represent distinct clonal proliferations arising from the separate components of the preexisting biphasic lesion. Azzopardi referred to these rare “collision” tumors as “carcinosarcomas”. Only in the background of a biphasic lesion did he accept the premise that carcinoma and sarcoma could arise together in a single tumor. Tumors containing areas of carcinoma admixed with sarcomatous-appearing elements are more frequently observed in the absence of a preexisting fibroepithelial lesion. Such tumors have been regarded most often as carcinomas with areas of sarcomatous metaplasia, or sarcomatoid carcinomas. As is the case with tumors at other anatomic sites, most such breast tumors are, on close examination, found to contain areas of transition between carcinomatous and sarcomatous components. Focally, the ductal cells appear to merge with areas of spindle cells, suggesting a transformation of ductal cells into the mesenchymal-appearing cells. The sarcomatoid cells often maintain expression of cytokeratin, an intermediate filament characteristic of epithelial cells but generally not expressed by either benign or malignant stromal cells. Moreover, molecular analysis of some biphasic tumors have demonstrated monoclonality, supporting the concept that the keratin-positive sarcomatous-appearing elements are derived from the coexisting carcinomatous component.20 – 27
Cases
98
70
50
27
26
Study
SEER*
Wargotz*4
Gutman33
Rayson34
Kaufman35
Mass (24) MMG (5) Pain (1) Nipple discharge (1) Mass(26) Skin fixation (9) Chest wall fixation (6) Ulceration (1)
Mass (70) Skin fixation (9) Chest wall fixation (2) Skin/nipple ulceration (7) Nipple retraction (3) No information
No information
Symptom at diagnosis
Mean 54 (27 – 80)
59 (39 – 90)
Mean 50 (25 – 76)
56 (29 – 95)
60 (33 – 91)
Median age years (range)
No information
2.4-year median DFS time
Median 27 months (4 – 240 months)
Mean 4.9 years
Patients registered from 1988 to 2001
Follow up
Mean 46
50 (9 – 190)
T1-16 T2-53 T3/4-20 TX-9
9/44 (20.5%)
26%
14 (14%) but unknown in 24 (24%)
ALN involvement at diagnosis
RM/MRM (15) SM (8) Excision (3)
47 (12 – 130)
5/20 (25%)
RM (1) 34 (5 – 70) 13% MRM (17) Lumpectomy (1) WLE/AD (6) SM (2)
WLE (10) SM/MRM (31) RM (9); with AD (44)
Mastectomy/AD (50) Mastectomy (9) WLE/AD (23) WLE (10) None (6) MRM (26) RM (28) SM (11) WLE (3) Excision (2)
Surgery (number cases)
Median T size mm (range)
20 patients: all FAC
2 patients: CMF (1) ChlorMF (1)
No information
?
4 patients
13% + 9 patients: AC (4) CMF (3) CAF (2)
12.5% +
1/11+ (9%)
4/68+ (6%)
Adjuvant chemotherapy ER/PR (cases)
12
5
17
11
31
Adjuvant radiotherapy (patients)
0 (0)
1 (4)
3 (6)
0 (0)
6 (6)
Cases with distant metastases at diagnosis (%)
Distant (11) Local (4)
Lung (10) Local (4)
Distant (19) Local (9)
Distant (24) Local (15)
No information
Site relapse (cases)
(continued overleaf )
40% 5-year OS Mean OS 65 months
43% 5-year OS 32% 5-year DFS 40% 3-year DFS 70% 3-year OS
49% 5-year OS Median RFS 10 months (range 1 month – 6.3 years)
60% 5-year OS
Outcome
Table 1 Clinical studies of patients with carcinosarcoma of the breast and data extracted from the SEER (Surveillance, Epidemiology, and End Results) population database. Only two studies and the SEER data included only breast carcinosarcoma using the traditional definition (*), while the others included all metaplastic high-grade sarcomatoid carcinomas of the breast. In the Gutman study, although the term carcinosarcoma was used, the tumors were defined as all epithelial carcinomas of the breast with “malignant sarcomatoid metaplasia”.
220 BREAST CANCER
No information
8
6
4
2
Oberman (1)36
Gersell38
Ferrara23
Gogas*1 30, 67
53 (39 – 81)
68 (55 – 75)
64 (51 – 89)
64 (46 – 82)
52 (32 – 88)
Median age years (range) Surgery (number cases)
–
Median 1.75 years
RM (2) SM (1) MRM (1) MRM (2)
Median 3.3 years MRM (5) RM (5) TM (4) Lumpectomy (1) Minimum Mastectomy (11) 12 years Axillary dissection (6) Median 2.5 years MRM (5) TM (2) RM (1) 2 months – TM (3) 11 years Excision (2) RM (1)
Follow up
65, 50
37 (30 – 60)
50 (20 – 90)
50 (35 – 90)
52 (22 – 100)
40 (15 – 190)
Median T size mm (range)
Both patients
0 (0%)
0/4 (0%)
1/6 (17%)
1 (8%)
2/12 (17%)
ALN involvement at diagnosis
1/2+
?
?
0
?
1/5+ (20%)
Both patients
1 patient
No information
None
No information
None
Adjuvant chemotherapy ER/PR (cases)
None
None
No information
None
No information
1
Adjuvant radiotherapy (patients)
0 (0)
1 (25 bone)
1 (17)
0 (0)
0 (0)
0 (0)
Cases with distant metastases at diagnosis (%)
Lung (4) Bone (1)
No information
Site relapse (cases)
2 relapses and 1 death Recurred at 3 months + 6 years
2 relapses and 4 dead
Lung, bone (1) Liver, brain (1)
Lung (1) Local (1)
Lung (1) Local (1)
5 relapses No and deaths information
4 (33%) relapses
7 relapses and 8 deaths
Outcome
Note: Numbers may add up to greater than the number of patients in the study if more than 1 factor occurred in some patients. The Oberman study examined 29 metaplastic breast cancer patients in total, of whom 8 had “spindle cell carcinoma”, 15 “carcinoma with pseudosarcomatous metaplasia”, and 6 squamous metaplasia. ALN, axillary lymph node; ER, estrogen receptor; PR, progesterone receptor; OS, overall survival; DFS, disease-free survival; RFS, relapse-free survival; RM, radical mastectomy; MRM, modified radical mastectomy; WLE/AD, wide local excision and axillary dissection; SM, simple mastectomy; TM, total mastectomy; MMG, mammogram; AC, doxorubicin and cyclophosphamide; CAF, AC and 5-fluorouracil; CMF, cyclophosphamide, methotrexate, and 5-fluorouracil; ChlorMF, chlorambucil and MF.
Mass (2) MMG (1)
Mass (6), 1 discovered by physician Nipple discharge (1) Mass (3)
No information
12
Kurian37
No information
Symptom at diagnosis
15
Cases
Oberman (2)36
Study
Table 1 (continued).
CARCINOSARCOMA OF THE BREAST 221
222
BREAST CANCER
Azzopardi’s strict definition of carcinosarcoma has not been maintained in the pathology literature. In their popular series of articles on the classification of metaplastic carcinomas, Wargotz et al. referred to carcinosarcoma as a subtype of metaplastic carcinoma.4 They considered tumors with a combination of overt carcinomatous elements and high-grade mesenchymal-appearing areas as carcinosarcomas, without regard to whether these elements arose within a preexisting fibroepithelial lesion. In fact, in their series of carcinosarcomas, the sarcomatous component of these tumors was immunoreactive for cytokeratin in over 50% of cases, and a majority of their carcinosarcomas contained areas of subtle transition between both components. To other authors, these very characteristics are more in favor of sarcomatoid carcinoma than carcinosarcoma, but Wargotz et al. believed the combined carcinomatous and sarcomatous behavior of these tumors supported their designation as carcinosarcomas, regardless of whether the sarcomatous-appearing elements appeared to be derived from the carcinomatous component. Although no consistent definition of carcinosarcoma has been used in the literature, most reports of carcinosarcomas appear to meet the definition provided by Wargotz et al., as such tumors clearly arising within preexisting fibroepithelial lesions are rare. It should be recognized, however, that many sarcomatoid carcinomas reported in the literature are, in fact, the same tumors. The carcinomatous component of carcinosarcomas is generally similar in appearance to invasive ductal carcinomas (IDC), but it may be a poorly differentiated carcinoma or squamous carcinoma (see Figure 1).3 – 6 It may even be DCIS. The sarcomatous component is, by definition, intermediate to high grade and may have a variety of histologic appearances. It is most often composed of a highly cellular arrangement of spindle cells. The nuclei are sometimes only moderately atypical but may have a very high mitotic rate. Other tumors have marked nuclear pleomorphism. There may be varying amounts of myxoid matrix deposited among the tumor cells. The cells are sometimes arranged in bundles or fascicles, or they may appear haphazardly distributed. Sometimes they form a cartwheeling or storiform pattern characteristic of malignant fibrous histiocytomas. Foci with osseous or cartilaginous differentiation or both are frequently observed, sometimes forming the bulk of the sarcomatous component and resembling osteosarcoma or chondrosarcoma. The gross appearance of these tumors is often deceptively well demarcated, but infiltrative growth is characteristic microscopically.
Differential Diagnosis It is important not to equate carcinosarcomas with metaplastic carcinomas.3,4 Metaplastic carcinomas are often reported to be an aggressive form of breast cancer, but this is because the majority are tumors that contain at least focal high-grade mesenchymal-appearing areas, and these tumors are biologically very aggressive. Other forms of metaplastic carcinoma have not been shown to be more aggressive than typical IDCs, and some appear to be more indolent. In particular, it is very important to distinguish low-grade fibromatosis-like metaplastic tumors from carcinosarcomas.42 The former are
composed of a bland proliferation of spindle cells reminiscent of fibromatosis, but in contrast to fibromatosis, the spindle cells are immunoreactive for cytokeratin. Sometimes the spindle cells are accompanied by a small component (comprising less than 5% by definition) of overt carcinoma, either ductal or squamous. These tumors are regarded as the most indolent form of metaplastic carcinoma. They tend to recur locally, but most reported cases have not metastasized. None have involved regional lymph nodes, but rare examples 4 cm in size or greater have metastasized to the lungs. Various reports of spindle cell carcinomas of the breast appear to include a large proportion of fibromatosis-like tumors and, consequently, describe a relatively favorable prognosis. They may also include somewhat more cellular and atypical spindle cells admixed with a small epithelial component. These tumors, however, do not contain the high-grade sarcomatous-appearing areas observed in carcinosarcomas. Although not always explicitly stated, the term “spindle cell carcinoma” in the literature generally implies a low-intermediate grade tumor. Because carcinosarcomas share characteristics of both carcinomas and sarcomas, they should also be distinguished from malignant phyllodes tumors and more rare primary sarcomas of the breast.3 Sarcomas have a low incidence of axillary lymph node involvement and a high rate of hematogenous dissemination. Axillary lymph node surgery is usually not performed for sarcomas unless the nodes are clinically enlarged, and sarcomas do not respond well to chemotherapy and require innovative therapeutic strategies. Carcinosarcomas should also be distinguished from pure keratin-positive sarcomatoid tumors and those with only a minor (<5%) overt carcinomatous component, as these pure or almost pure sarcomatoid tumors (excluding low-grade fibromatosis-like tumors) have a biologic behavior similar to that of pure sarcomas.
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS Table 1 shows the clinical features associated with breast carcinosarcomas in a number of small published studies and in the SEER population database.1,4,23,33 – 38,43 As stated, the definition of carcinosarcoma varies somewhat from study to study. In the largest study of 70 breast carcinosarcoma patients by Wargotz et al., the definition used was “intraductal or infiltrating carcinoma contiguous or subtly merged with a highly cellular, mitotically active, pleomorphic spindle cell stroma”, while the Gutman et al. study of 50 carcinosarcoma patients included cases with “coexistent infiltrating ductal carcinoma and sarcomatoid metaplasia”.4,33 Generally, the majority of patients present with a palpable breast lump or mass. Fine needle aspiration (FNA) is usually positive for malignancy in the case of metaplastic breast cancer, although ductal and metaplastic elements are both found in smear material in just over half of cases allowing a definite diagnosis prior to definitive surgery.44 – 46 The tumor is usually relatively large and well circumscribed at diagnosis.4,7,20,21,23,33 – 38,43,47 As a result of large tumor size at presentation, skin or chest wall fixation, nipple retraction, and skin and nipple ulceration are relatively frequent.
CARCINOSARCOMA OF THE BREAST
Occasionally, carcinosarcomas may manifest as inflammatory carcinomas.48 Some studies have found tumor size to be the major predictor of patient outcome.33,36 In most series, the majority of breast carcinosarcomas and metaplastic sarcomatoid breast cancers have a high nuclear grade. Hormone receptors and Her-2/neu are usually negative. In most cases, the locoregional lymph nodes are not involved by tumor at presentation. Neither do most women have recognizable distant metastases at the time of diagnosis. Most reports suggest that hematogenous dissemination is more characteristic of breast carcinosarcoma than spread along lymphatic channels; this has traditionally been regarded as more in keeping with the sarcomatous rather than the carcinomatous phenotype. Consistent with this, certain investigators have noted that carcinosarcomas seem to behave biologically differently from conventional adenocarcinoma of the breast, with sarcomatoid characteristics dominating the clinical course.33 Generally, the prognosis is relatively poor, with most series suggesting that at least one-third of patients will die from the disease. This is in keeping with other carcinosarcomas such as uterine carcinosarcomas.49 Locoregional recurrence of breast carcinosarcoma is relatively frequent (see Table 1).
Imaging of Carcinosarcoma Carcinosarcomas and other metaplastic breast carcinomas are frequently visible (up to 90% of patients) by mammography. The characteristic mammographic appearance of metaplastic breast carcinoma is a high-density round, oval, or lobeshaped mass with indistinct, partially circumscribed, or circumscribed margins (see Figure 2a).50 – 54 Spiculated or partially spiculated margins can be present in up to 14% of the masses, and calcifications in up to 26% of the masses. Other tumors that may present with similar mammographic features of round or oval-shaped masses with circumscribed margins include the triad of circumscribed carcinomas (medullary, papillary, and mucinous), phyllodes tumors, and high-grade IDCs in women with genetic mutations.55 – 57 Carcinosarcoma is most frequently seen on sonograms as a solid, hypoechoic, hypervascular, oval mass with indistinct margins and posterior acoustic enhancement (see Figure 2b). Mixed solid and cystic masses have been described in 19–50% of the patients.51,53,54 On MR imaging, T2-weighted images reveal fairly well-defined oval and lobular-shaped masses with internal high signal–intensity necrotic or cystic components. On three-dimensional fast low-angle shot dynamic enhancement subtraction images, there is early enhancement and a delayed washout in a peripheral rim and nonenhancing internal components.58,59 Despite these characteristic imaging features of carcinosarcoma and other metaplastic breast carcinomas on mammography, sonography, and MR imaging, there is overlap with other malignant and benign tumors. Therefore, biopsy is necessary. As with breast adenocarcinoma, the additional role of sonography is in the evaluation of regional lymph node status.
223
PROGNOSTIC FACTORS AND PATIENT OUTCOMES In the SEER database, the majority of breast carcinosarcoma patients (54; 55%) presented with American Joint Committee for Cancer (AJCC Staging Manual, third edition, breast cancer staging) stage II disease. As in other studies, and with breast adenocarcinoma, patient outcome deteriorates as the stage of disease at presentation becomes more advanced (see Table 2).33,35 Some studies suggest that tumor size at diagnosis is the most important determinant of patient outcome, with a tumor size greater than 5 cm having the most deleterious effect on prognosis.4,33 Axillary nodal status generally has less impact on outcome, but this is probably confounded by the relatively low frequency of nodal metastases seen in most studies. Prognostic factors described by Wargotz et al. to be associated with a better prognosis in breast carcinosarcoma patients were complete microscopic circumscription of the tumor and the presence of an inflammatory infiltrate.4 In the study by Rayson et al., age greater than 60 years and no history of prior estrogen use were associated with a better outcome in a univariate analysis of 27 patients with metaplastic breast cancer.37 Table 1 indicates the generally poor prognosis associated with carcinosarcoma of the breast in the various reported studies. When the characteristics of breast carcinosarcomas and metaplastic sarcomatoid carcinomas in the SEER database are compared, it is apparent that clinical features and outcomes are very similar for both tumor types (see Table 3). In the Wargotz et al. study, the 5-year overall survival was 49%, and 43% in the Gutman et al. study.4,43 The median survival after recurrence is as low as 8–15 months, and the prognosis is particularly ominous in those with distant metastases.33,34 However, there is evidence that patients with only locoregional recurrence of disease may be cured, particularly with complete surgical resection.4,33,60 In one study, the median survival after solitary recurrence was 22 months in those who underwent surgical resection, compared to 10 months in other patients after recurrence. Wargotz et al. found a worse prognosis associated with carcinosarcoma compared to spindle cell carcinoma of the breast, with cumulative 5-year survival rates of 64% for 100 patients with the latter tumor type (p = 0.056), although the difference was less marked when survival was broken down by stage (see Table 4). It may be that the larger size of the tumor at diagnosis in their carcinosarcoma series (mean 6.3 vs 4.4 cm) partly accounts for this difference, although, as stated earlier, the term spindle cell carcinoma also implies a less aggressive tumor to many pathologists, and this was reflected in their series where many of the spindle cell carcinomas were of a lower grade compared to the carcinosarcomas.
TREATMENT OF BREAST CARCINOSARCOMAS Carcinosarcomas seem to behave biologically differently from conventional adenocarcinoma of the breast, with sarcomatoid characteristics dominating the clinical course. These biologic distinctions led one group to suggest that multimodality treatment including mastectomy and sarcomaoriented adjuvant chemotherapy and radiotherapy should be
224
BREAST CANCER
(a)
(b)
Figure 2 Radiographic appearance of a breast carcinosarcoma. (a) Right mediolateral oblique mammogram shows an oval high-density noncalcified mass with indistinct margins. (b) Right transverse ultrasound shows an oval heterogeneously hypoechoic solid mass with posterior acoustic enhancement.
Table 2 5-year and median (in months) overall (OS) and disease-specific survival (DSS) rates for 98 breast carcinosarcoma patients registered in the Surveillance, Epidemiology, and End Results (SEER) database between 1988 and 2001 by stage at diagnosis.
Stages I II III IV
Patient numbera
Median OS
5-year OS
5-year OS: 95% CI
Median DSS
5-year DSS
5-year DSS: 95% CI
20 54 7 6
NR 96 25 3
0.73 0.59 0.44 0.00
0.43 – 0.89 0.44 – 0.71 0.07 – 0.78 –
NR NR NR 3
0.87 0.72 0.53 0.00
0.57 – 0.97 0.56 – 0.83 0.07 – 0.86 –
CI, confidence interval; NR, not reached; AJCC, American Joint Committee for Cancer Staging Manual, third edition. a Outcome data was missing in 11 patients.
Table 3 5-year and median (in months) overall (OS) and disease-specific survival (DSS) rates for 213 metaplastic sarcomatoid breast cancer patients registered in the Surveillance, Epidemiology, and End Results (SEER) database between 1988 and 2001 by stage at diagnosis.
AJCC stages
Patient numbera
Median OS
5-year OS
5-year OS: 95% CI
Median DSS
5-year DSS
5-year DSS: 95% CI
45 122 27 14
NR 97 NR 5
0.81 0.59 0.67 0.18
0.56 – 0.93 0.46 – 0.69 0.40 – 0.84 0.03 – 0.43
NR NR NR 5
0.93 0.67 0.71 0.20
0.74 – 0.98 0.54 – 0.77 0.43 – 0.87 0.03 – 0.46
I II III IV
CI, confidence interval; NR, not reached; AJCC, American Joint Committee for Cancer Staging Manual, third edition. a Outcome data was missing in five patients.
Table 4 Carcinosarcoma and spindle cell carcinoma of the breast.
Stages Carcinosarcoma Spindle cell carcinoma Carcinosarcoma Spindle cell carcinoma Carcinosarcoma Spindle cell carcinoma
1 II III
Patient number
5-year overall survival (%)
2 19 32 42 31 33
100 84 63 55 35 55
Source: Reproduced from reference 4 by permission of Wiley-Liss, Inc. 1989.
considered in the treatment of this disease.33 Mastectomy with adjuvant chemotherapy or radiotherapy or both was found to be superior to mastectomy alone and to wide local excision with or without adjuvant therapy, particularly for patients with stage II carcinosarcoma. Certain investigators also found complete surgical resection of recurrent disease to result in better outcomes than any other modality.33,50 Conventional systemic chemotherapy appears to be less effective in these and other metaplastic breast cancers.34 As a result of this and the inherent aggressiveness of the disease, the median survival of patients with metastatic disease is as low
CARCINOSARCOMA OF THE BREAST
as 8 months.37 As with breast carcinosarcomas, carcinosarcomas at other sites including the uterus are highly aggressive neoplasms that frequently recur after surgical treatment and adjuvant chemoradiotherapy, and also respond poorly to salvage chemotherapy and irradiation.49 Some investigators therefore conclude that patients with metaplastic breast cancer including carcinosarcomas, particularly those with metastatic disease, are appropriate candidates for innovative therapeutic regimens.
Surgical Management Much of the literature groups breast carcinosarcomas with other subtypes of metaplastic carcinoma that may have different behaviors.33,37,61,62 Therefore, defining surgical treatment for carcinosarcoma patients is often difficult. This is in contrast to surgical treatment for patients diagnosed with IDC, in whom multiple randomized trials have demonstrated that breast conserving treatment (BCT) and mastectomy have equal efficacy.63,64 Two studies in the literature specifically address surgical management of carcinosarcoma of the breast.4,33 The study by Gutman et al. noted that overall and disease-free survival rates were not significantly different among patients treated by different surgical procedures (wide local excision – 10 patients, simple mastectomy or modified radical mastectomy – 31 patients, and radical mastectomy – 9 patients). However, when mastectomy followed by adjuvant therapy (radiation therapy and/or chemotherapy) was compared with mastectomy alone and with wide local excision with or without adjuvant therapy, the local recurrence rate was much lower in patients undergoing surgery followed by adjuvant therapy [45 vs 89% (p = 0.04) and 78% (p = 0.08), respectively]. Therefore, the authors recommended a multimodality approach that included mastectomy and sarcoma-oriented adjuvant chemotherapy and radiotherapy. In the study by Wargotz et al., the majority of patients underwent mastectomy (65/70 patients), and only a minority had a partial mastectomy (wide local excision – three patients, excisional biopsy – two patients).4 Only one patient underwent a partial mastectomy followed by adjuvant radiation therapy. Four of the five patients treated with partial mastectomy had a recurrence locally and were treated with subsequent mastectomy. One of the 65 patients who underwent some form of mastectomy experienced a local recurrence. Unfortunately, neither the tumor margin status nor the radiation therapy history of the specific patients who developed a local recurrence was clarified in the publication. In this study, those patients who developed a recurrence with distant metastases had an extremely poor prognosis compared with those patients experiencing only a local recurrence. Since the majority of patients underwent mastectomy, no conclusions could be drawn about the role of BCT. These authors also concluded by comparison with other subtypes of metaplastic carcinoma that carcinosarcoma seemed to be most aggressive. Because these two studies are now relatively old and surgical treatments for breast cancer have improved considerably over the past 15 years, it is virtually impossible to interpret the data published by Gutman et al. and Wargotz
225
et al. in the current era so as to make surgical treatment recommendations. In addition, the tumor margin status of those patients who underwent BCT (wide local excision, segmental mastectomy, partial mastectomy, or excisional biopsy, with or without radiation therapy) versus mastectomy was not assessed adequately. Both the tumor margin status and adjuvant treatment in patients undergoing BCT are extremely important to minimize local recurrence. Finally, the role of radiation therapy was not explored adequately in either study, specifically with regard to its use following wide local excision or segmental mastectomy. Presently, approximately 50–70% of patients diagnosed with primary breast cancer undergo BCT and it has been suggested that wide local excision may also be appropriate for patients with primary breast sarcoma.65 – 67 Thus, on the basis of existing literature and considering current treatment strategies, it seems reasonable to suggest local excision of carcinosarcoma with “wide” margins (in the form of mastectomy, wide local excision, or partial mastectomy) followed by adjuvant chemoradiation. To actually determine an acceptable standard of care for patients with this rare tumor type, a multicenter prospective database needs to be established under the guidance of a multidisciplinary group of oncologists. The role of surgery to the regional nodes is more apparent. Since the axillary lymph nodes will be involved in up to 20–25% of breast carcinosarcoma patients at presentation (see Table 1), and such involvement may be associated with a poorer prognosis, it is reasonable to suggest some form of axillary evaluation at the time of the primary breast surgery.33 However, there are no current data on the role of sentinel lymph node biopsy, a relatively new surgical procedure, in the evaluation of regional lymph nodes in patients with breast carcinosarcoma.
Chemotherapy and Radiotherapy Conventional systemic chemotherapy appears to be less effective in carcinosarcomas and other metaplastic breast cancers.34 This is supported by a recent (as yet unpublished) study at The University of Texas MD Anderson Cancer Center experience of 100 patients with biphasic metaplastic sarcomatoid breast cancer. In this study, we found a pathologic complete response rate of 10% and a clinical partial and complete response rate of 26% among 21 patients treated with primary systemic chemotherapy, of which 15 had conventional anthracycline-containing breast cancer regimens, 5 anthracycline/taxane-based regimens, and 1 the sarcoma-type chemotherapy regimen doxorubicin and ifosfamide (AI). This is considerably lower than the response rates of breast adenocarcinoma to primary systemic chemotherapy, particularly when one considers that the majority of such metaplastic breast cancers are hormone receptor-negative and of high nuclear grade. In the study by Rayson et al., of 10 chemotherapy regimens used in 7 metaplastic breast cancer patients with metastatic disease, there was 1 partial response (to doxorubicin) and 2 cases of stable disease (with doxorubicin and cyclophosphamide/5-fluorouracil/actinomycin D).34 Four patients with metastatic disease were treated with tamoxifen and two had stable disease (12 and 18 months), although there was no information provided on the hormone receptor
226
BREAST CANCER
status of these tumors. The median survival after disease recurrence in this study was just 8 months. Wargotz et al. stated in their study that “chemotherapy for recurrence offered no significant advantage” but did not provide more detailed data to support this.4 In the study by Gutman et al., chemotherapy (mainly doxorubicin-based) and hormonal therapy for metastatic breast carcinosarcoma also had minimal activity.33 Because of the small patient numbers in the studies presented in Table 1 and the lack of randomized controlled data, it is currently unknown if adjuvant chemotherapy or radiotherapy benefit patients with breast carcinosarcoma or other sarcomatoid breast cancers. However, in the Gutman et al. study, 14 patients with stage II disease given adjuvant chemotherapy, radiotherapy, or both had longer disease-free and overall survivals than 6 patients with stage II disease who did not receive adjuvant therapy. However, the survival benefit for 17 patients with stage III disease treated with adjuvant therapy failed to reach statistical significance, probably because of small numbers and low power. Overall, of 17 patients given adjuvant radiotherapy and 29 patients not irradiated, there were 2 and 9 local recurrences, respectively (p = 0.13). Similarly, of 19 patients treated with adjuvant chemotherapy and 27 patients not so treated, there were 5 and 11 distant recurrences, respectively (p not significant). These statistical analyses are confounded by low patient numbers and also by lack of control for initial risk (one would expect patients chosen for adjuvant therapy to be at higher risk of recurrence). Most patients treated with adjuvant therapy in this study had 5-fluorouracil, doxorubicin, and cyclophosphamide (FAC) and/or “standard portal and standard dose radiotherapy”. This group concluded that sarcoma-type chemotherapy combined with radiation therapy should be considered in the adjuvant treatment of patients with breast carcinosarcomas. There is very little in the literature concerning the use of the currently popular sarcoma regimen AI (doxorubicin, ifosfamide) in patients with breast carcinosarcoma. In our recent analysis of 100 biphasic metaplastic sarcomatoid breast cancer patients, three patients given adjuvant AI chemotherapy were alive and recurrence-free at a median follow-up of 55 months, in comparison to one of eight patients treated with adjuvant cyclophosphamide, methotrexate, and 5-fluorouracil (CMF), 50% of 50 patients given adjuvant anthracyclinebased therapy (usually FAC) and 54% of 13 patients given adjuvant anthracycline/taxane-based therapy. Although limited statistical power again precludes definitive interpretation of these data, it is reasonable to consider at least an anthracycline-based regimen in the adjuvant treatment of breast carcinosarcoma patients.
MOLECULAR PATHOLOGY AND POTENTIAL NEW MARKERS AND TARGETS The monoclonality of carcinosarcomas is no longer in dispute. In one study of a breast carcinosarcoma in which the demarcation between the carcinomatous and sarcomatous components was distinct in all microscopic fields, immunohistochemical analysis was negative for epithelial membrane
antigen and keratin in the sarcomatous component and negative for desmin in the carcinomatous component.8 However, polymerase chain reaction studies revealed an identical pattern of X-chromosome inactivation and an identical TGT–> TTT (T, thymidine; G, guanidine) transversion in codon 275 of the p53 gene in both components, strongly supporting the hypothesis that the tumor was derived from a single (totipotent?) stem cell. Loss of heterozygosity (LOH) analysis of six cases of breast carcinosarcoma and sarcomatoid carcinoma revealed identical clonality of the carcinomatous and spindle-cell components in all cases, identical to a focus of DCIS present in one case.68 Additional and differential areas of LOH were found in the sarcomatous component of the tumors, leading the authors to suggest that further genetic changes were required in the carcinomatous components to allow “transdifferentiation”. Recently, the use of gene profiling has allowed us to identify a number of subtypes of breast cancer, each with differences in prognosis and therapy responsiveness.69 The basal form of breast cancer is associated with aggressive behavior and poor survival. This type of breast cancer is believed to arise from basal progenitor or myoepithelial cells in the ductal epithelium of the breast.70,71 There are many similarities in the clinical behavior and pathologic characteristics of this breast cancer subtype and sarcomatoid breast cancers; indeed, it is plausible to suggest that carcinosarcomas may belong to this basal group of breast cancers, originating from a myoepithelial progenitor cell in the breast with a degree of maturational plasticity.7,8,20 – 27,72 The application of gene profiling and other new technologies to carcinosarcomas of the breast with comparison to more common breast cancer subtypes will help clarify the origins of these aggressive forms of breast cancer and their relationship to breast adenocarcinoma. Wargotz et al. were among the first to suspect a myoepithelial origin for carcinosarcomas following their demonstration of dual staining of both tumor cell types with epithelial and myoepithelial markers, including actin and S-100. As further support for a myoepithelial origin, a recent study found prominent p63 expression in sarcomatoid/metaplastic carcinomas of the breast.73 p63 is a p53 homolog whose gene is located on chromosome 3q27. Unlike the p53 gene, which encodes a unique 53-kDa protein, the p63 gene encodes six isoforms with differing C-terminal (α, β, and γ ) and N-terminal (transactivating and N isoforms) regions.74,75 While transactivating isoforms are able to promote the transcription of p53-reporting genes, N isoforms are unable to do so. N isoforms are consistently expressed in the nuclei of normal breast myoepithelial and basal cells, as well as in basal cells of several multilayered epithelia and may constitute a mechanism to overcome p53-driven cell cycle arrest and apoptosis.73 – 76 There is strong and consistent p63 expression in the nuclei of spindle and epithelioid neoplastic cells in both mono- and biphasic sarcomatoid tumors of the breast.73 Some p63-positive spindle cells also express cytokeratins. On the basis of double-immunostaining studies, a subset of p63-positive cells has been shown to coexpress smooth-muscle markers, while others do not.73,76 The p63-positive smooth-muscle marker-positive phenotype
CARCINOSARCOMA OF THE BREAST
may represent the usual immunophenotype of myoepithelial cells, while the p63-positive smooth-muscle marker-negative phenotype is probably expressed by basal cells.73 As stromal cells observed in normal breast, fibromatoses, fibroadenomas, and phyllodes tumors fail to express p63, it seems to be a good marker of the myoepithelial phenotype. However, p63 expression in normal breast tissue and other neoplasms of the breast is not a fully characterized phenomenon.73 Another study of carcinosarcomas demonstrated staining of tumor cells for basal cell-type cytokeratins and a combination of established (CD10, p63, smooth-muscle actin, and S-100) and novel myoepithelial markers.77 Thus, the study of breast carcinosarcomas may have the potential to assist us in better understanding some aspects of normal breast epithelial development and also the mechanisms of basal breast carcinogenesis. The loss of epithelial morphology and acquisition of mesenchymal characteristics (epithelial-mesenchymal transition, EMT) have been reported for some breast and other carcinoma cells in tumor progression.25 Although originally thought to be restricted to mesenchymal cells and sarcomatoid epithelial neoplasms such as carcinosarcomas, in human infiltrating ductal breast carcinomas, correlated upregulation of the mesenchymal markers tenascin-C and vimentin is frequently observed and associated with increased malignancy and invasiveness.78,79 Like vimentin, cytoplasmic tenascinC expression is present in the tumor cells of up to 20% of all human breast carcinomas but, in contrast, tenascinC expression is seen in significantly more tumor cells and tends to increase in the invasive components of tumors. Thus, it has been hypothesized that tenascin-C coexpression with vimentin is representative of cancer cells undergoing EMT. Tenascin-C and vimentin expression also correlate with downregulation of estrogen receptors and increased tumor grade in breast cancer. Epithelial tumor cells coexpressing vimentin and cytokeratins often have a spindle-shaped phenotype. Tenascin-C and vimentin may represent regulator genes involved in EMT during mammary carcinogenesis.25 Thus, a sarcomatoid phenotype associated with breast cancer may reflect myofibroblastic metaplasia or transdifferentiation of adenocarcinoma cells during EMT, a phenomenon possibly more likely to occur in a marked fashion when the tumor arises from cells with a myoepithelial phenotype. The coexpression of tenascin-C and vimentin is present in the breast carcinosarcoma cell line Hs578T, a cell line that possesses an activating H-ras mutation, in the stromal breast adenocarcinoma cell line MDA-MB-231, which possesses an activating K-ras mutation, in the Her-2/neu-amplified breast cancer cell line SKBR3, and in the immortalized breast myoepithelial cell line HBL100. Although several mechanisms have been implicated in EMT by studies, ras deregulation, by mutation or otherwise, may be one such mechanism, and indeed the transfection of transformed immortalized breast epithelial cells with a constitutively active form of ras can induce the acquisition of a spindle morphology with loss of E-cadherin and other molecular changes.80 E-cadherin, like N- and P-cadherin, is present in some areas of sarcomatoid breast cancers but is lost in the sarcomatous areas of carcinosarcomas.81,82 Because of the possible overlap
227
between carcinosarcomas and EMT, the expression of other molecules implicated in EMT, such as myc, twist, snail, and slug, should be studied in these rare aggressive breast tumors.83,84 A high staining intensity for COX2 is observed in one-third of uterine carcinosarcomas and is associated with significantly poorer outcome.49 c-KIT is expressed in approximately 17% and Her-2/neu in 30% of uterine carcinosarcomas. Some human breast cancers, notably those which are estrogen receptor-negative with high metastatic potential, produce high levels of prostaglandin E2.85 In some cell types, expression of the inducible COX2 isoform occurs in association with a ras gene mutation. Although evidence suggests the Her-2/neu gene is generally not amplified in metaplastic breast cancers, COX2 and c-KIT need to be evaluated in carcinosarcomas and sarcomatoid cancers of the breast and may constitute potential therapeutic targets. The roles played by mutations and other abnormalities in ras signaling and by p63 in the pathogenesis of metaplastic breast carcinogenesis also need to be investigated.
CONCLUSIONS AND AUTHORS’ RECOMMENDATIONS Carcinosarcomas of the breast have traditionally been defined as biphasic tumors with a clear light microscopic demarcation between epithelial and mesenchymal components but more recent studies have grouped them together with a more heterogeneous appearing group of breast adenocarcinomas with varying degrees of high-grade sarcomatoid metaplasia. Although this would likely offend the traditional pathology purists, it may more accurately reflect the biology of many of these lesions. Evidence now suggests these are similar tumors developing through a peculiar phenotypic transformation of carcinoma cells into sarcoma. These tumors are high-grade aggressive neoplasms that express mesenchymal, epithelial, and myoepithelial markers, and are usually negative for hormone receptors and Her-2/neu. In truth, the most reasonable conclusion to draw from available studies is that there are currently not enough data available to justify treating breast carcinosarcomas differently from other high-grade breast adenocarcinomas. It seems reasonable to suggest local excision of carcinosarcoma with “wide” margins (in the form of mastectomy, wide local excision, or partial mastectomy) followed by adjuvant chemoradiation. The study of Gutman et al. certainly suggests that adjuvant radiation therapy may have some activity in terms of local tumor control. Similarly, the data presented above suggest that, although responsiveness to cytotoxic chemotherapy may be limited, adjuvant anthracycline-based chemotherapy is likely to be associated with some improvement in patient outcomes. It is not known whether this chemotherapy should be based on a sarcoma-type regimen such as AI or whether it should incorporate a taxane as with conventional breast adenocarcinomas. Clearly, the stage of the tumor is a major driver of prognosis and should be taken into account when making a therapy decision. If indeed carcinosarcoma sensitivity to cytotoxic chemotherapy is limited, this probably reflects molecular changes; aberrations that
228
BREAST CANCER
need to be explored in this regard are ras activation, possibly by mutation, which is associated with chemoresistance in other tumors (e.g. lung, pancreatic), and p53 mutation, which is known to occur more commonly in the basal subtype of breast adenocarcinoma than in other breast cancer subtypes. In our opinion, better genomic and proteomic characterization of breast carcinosarcomas is required and will not only increase our knowledge of and ability to rationally target these rare tumors but may also help improve understanding of more common tumor types such as basal breast cancers and of the molecular mechanisms that drive EMT, an event currently felt to be responsible for the ability of many solid tumors to metastasize. New technologies such as comparative genomic hybridization and transcriptional profiling will assist us in this regard.
REFERENCES 1. Gogas J, et al. Carcinosarcoma of the breast: report of two cases. Eur J Gynaecol Oncol 2003; 24: 93 – 5. 2. Aritas Y, et al. Carcinosarcoma of the breast: clinicopathologic and radiologic findings in an unusual case. Breast J 2003; 9: 323 – 4. 3. Tavassoli FA. Pathology of the Breast, 2nd ed. Hong Kong, China: Appleton and Lange, 1999. 4. Wargotz ES, Norris HJ. Metaplastic carcinomas of the breast. III. Carcinosarcoma. Cancer 1989; 64: 1490 – 9. 5. Han AC, et al. Distinct cadherin profiles in special variant carcinomas and other tumors of the breast. Hum Pathol 1999; 30: 1035 – 9. 6. Harris M, Persaud V. Carcinosarcoma of the breast. J Pathol 1974; 112: 99 – 105. 7. Foschini MP, Dina RE, Eusebi V. Sarcomatoid neoplasms of the breast: proposed definitions for biphasic and monophasic sarcomatoid mammary carcinomas. Semin Diagn Pathol 1993; 10: 128 – 36. 8. Wada H, et al. Carcinosarcoma of the breast: molecular-biological study for analysis of histogenesis. Hum Pathol 1998; 29: 1324 – 8. 9. Nappi O, Wick MR. Sarcomatoid neoplasms of the respiratory tract. Semin Diagn Pathol 1993; 10: 137 – 47. 10. Colombi RP. Sarcomatoid carcinomas of the female genital tract (malignant mixed mullerian tumors). Semin Diagn Pathol 1993; 10: 169 – 75. 11. Iezzoni JC, Mills SE. Sarcomatoid carcinomas (carcinosarcomas) of the gastrointestinal tract: a review. Semin Diagn Pathol 1993; 10: 176 – 87. 12. Reuter VE. Sarcomatoid lesions of the urogenital tract. Semin Diagn Pathol 1993; 10: 188 – 201. 13. Siegal A, Freund U, Gal R. Carcinosarcoma of the stomach. Histopathology 1988; 13: 350 – 3. 14. Battifora H. Spindle cell carcinoma: ultrastructural evidence of squamous origin and collagen production by the tumor cells. Cancer 1976; 37: 2275 – 82. 15. Guarino M, et al. Sarcomatoid carcinomas: pathological and histogenetic considerations. Pathology 1996; 28: 298 – 300. 16. George E, et al. Malignant mixed mullerian tumours: an immunohistochemical study of 47 cases with histogenetic considerations and clinical correlation. Hum Pathol 1991; 22: 215 – 23. 17. Fromowitz FB, Bard RH, Koss L. The epithelial origin of a malignant mesodermal mixed tumor of the bladder: report of a case with long-term survival. J Urol 1984; 132: 978 – 81. 18. Guarino M. Epithelial-to-mesenchymal change of differentiation. From embryogenetic mechanism to pathological patterns. Histopathology 1995; 10: 171 – 84. 19. Sonoda Y, et al. Carcinosarcoma of the ovary in a patient with a germline BRCA2 mutation: evidence for monoclonal origin. Gynecol Oncol 2000; 76: 226 – 9. 20. Wargotz ES, Deos PH, Norris HJ. Metaplastic carcinomas of the breast. II. Spindle cell carcinoma. Hum Pathol 1989; 20: 732 – 40.
21. Sapino A, et al. Tumour types derived from epithelial and myoepithelial cell lines of R3230AC rat mammary carcinoma. Cancer Res 1992; 52: 1553 – 60. 22. Wargotz ES, Norris HJ. Metaplastic carcinomas of the breast. 1. Matrixproducing carcinoma. Hum Pathol 1989; 20: 628 – 35. 23. Ferrara G. Sarcomatoid carcinoma of the breast: pathology of four cases. Breast Dis 1995; 8: 283 – 94. 24. Balercia G, Bhan AK, Dickersin GR. Sarcomatoid carcinoma: an ultrastructural study with light microscopic and immunohistochemical correlation of 10 cases from various anatomic sites. Ultrastruct Pathol 1995; 19: 249 – 63. 25. Dandachi N, et al. Co-expression of tenascin-C and vimentin in human breast cancer cells indicates phenotypic transdifferentiation during tumour progression: correlation with histopathological parameters, hormone receptors, and oncoproteins. J Pathol 2001; 193: 181 – 9. 26. Teixeira MR, et al. Cytogenetic analysis shows that carcinosarcomas of the breast are of monoclonal origin. Genes Chromosomes Cancer 1998; 22: 145 – 51. 27. Yang GC, Yee HT, Waisman J. Metaplastic carcinoma of the breast with rhabdomyosarcomatous element: aspiration cytology with histological, immunohistochemical, and ultrastructural correlations. Diagn Cytopathol 2003; 28: 153 – 8. 28. Wick MR, Swanson PE. Carcinosarcomas: current perspectives and an historical review of nosological concepts. Semin Diagn Pathol 1993; 10: 118 – 27. 29. Tomasino RM, Verace V. Rare tumors of the female breast (carcinosarcoma) and sarcomatous transformations of intracanalicular fibroadenoma. [Article in Italian]. Arch De Vecchi Anat Patol 1967; 49: 401 – 18. 30. Donegan WL. Sarcomas of the breast. Major Probl Clin Surg 1967; 5: 245 – 72. 31. Kruger J. Carcinosarcoma of the breast. [Article in German]. Zentralbl Chir 1955; 80: 1238 – 41. 32. Smoilovskaia EIa, et al. Carcinosarcoma of the breast developing in monkeys after hyperestrinization and the use of radioactive silver (Ag110). [Article in Russian]. Vopr Onkol 1960; 6: 35 – 42. 33. Gutman H, et al. Biologic distinctions and therapeutic implications of sarcomatoid metaplasia of epithelial carcinoma of the breast. J Am Coll Surg 1995; 180: 193 – 9. 34. Rayson D, et al. Metaplastic breast cancer: prognosis and response to systemic therapy. Ann Oncol 1999; 10: 413 – 9. 35. Kaufman MW, et al. Carcinoma of the breast with pseudosarcomatous metaplasia. Cancer 1984; 53: 1908 – 17. 36. Oberman HA. Metaplastic carcinoma of the breast: a clinicopathologic study of 29 patients. Am J Surg Pathol 1987; 11: 918 – 29. 37. Kurian KM, Al-Nafussi A. Sarcomatoid/metaplastic carcinoma of the breast: a clinicopathological study of 12 cases. Histopathology 2002; 40: 58 – 64. 38. Gersell DJ, Katzenstein AA. Spindle cell carcinoma of the breast. Hum Pathol 1981; 12: 550 – 61. 39. Kloos I, et al. Tamoxifen-related uterine carcinosarcomas occur under/after prolonged treatment: report of five cases and review of the literature. Int J Gynecol Cancer 2002; 12: 496 – 500. 40. Cheung YC, et al. Sonographic features with histologic correlation in two cases of palpable breast cancer after breast augmentation by liquid silicone injection. J Clin Ultrasound 2002; 30: 548 – 51. 41. Azzopardi JG. Problems in breast pathology. In Bennington JL (ed) Major Problems in Pathology. Philadelphia, Pennsylvania: W.B. Saunders, 1979, Vol. 11. 42. Gobbi H, et al. Metaplastic breast tumours with a dominant fibromatosis-like phenotype have a high risk of local recurrence. Cancer 1999; 85: 2170 – 82. 43. Al-Nafussi AN. Spindle cell tumours of the breast: practical approach to diagnosis. Histopathology 1999; 35: 1 – 13. 44. Stanley MW, Tani EM, Skoog L. Metaplastic carcinoma of the breast: fine-needle aspiration cytology of seven cases. Diagn Cytopathol 1989; 5: 22 – 8. 45. Gupta RK. Cytodiagnostic patterns of metaplastic breast carcinoma in aspiration samples: a study of 14 cases. Diagn Cytopathol 1999; 20: 10 – 2.
CARCINOSARCOMA OF THE BREAST 46. Nogueira M, Andre S, Mendonca E. Metaplastic carcinomas of the breast – fine needle aspiration (FNA) cytology findings. Cytopathology 1998; 9: 291 – 300. 47. Eusebi V, et al. Sarcomatoid carcinomas of the breast: an immunocytochemical study of 14 cases. Prog Surg Pathol 1989; 10: 83 – 9. 48. Kuo S, et al. Metaplastic carcinoma of the breast-analysis of eight Asian patients with special emphasis on two unusual cases presenting with inflammatory-type breast cancer. Anticancer Res 2000; 20: 2219 – 22. 49. Raspollini MR, et al. COX-2, c-KIT and HER-2/neu expression in uterine carcinosarcomas: prognostic factors or potential markers for targeted therapies? Gynecol Oncol 2005; 96: 159 – 67. 50. Patterson SK, et al. Metaplastic carcinoma of the breast: mammographic appearances with pathologic correlation. AJR Am J Roentgenol 1997; 169: 709 – 12. 51. Samuels TH, et al. Squamous cell carcinoma of the breast. Can Assoc Radiol J 1996; 47: 177 – 82. 52. Brenner RJ, et al. Metaplastic carcinoma of the breast: report of three cases. Cancer 1998; 82: 1082 – 7. 53. Park JM, et al. Metaplastic carcinoma of the breast: mammographic and sonographic findings. J Clin Ultrasound 2000; 28: 179 – 86. 54. Gunhan-Bilgen I, et al. Metaplastic carcinoma of the breast: clinical, mammographic, and sonographic findings with histopathologic correlation. AJR Am J Roentgenol 2002; 178: 1421 – 5. 55. Sickles EA. Nonpalpable, circumscribed, noncalcified solid breast masses: likelihood of malignancy based on lesion size and age of patient. Radiology 1994; 192: 439 – 42. 56. Kaas R, et al. The significance of circumscribed malignant mammographic masses in the surveillance of BRCA 1/2 gene mutation carriers. Eur Radiol 2004; 14: 1647 – 53. 57. Liberman L, et al. Benign and malignant phyllodes tumors: mammographic and sonographic findings. Radiology 1996; 198: 121 – 4. 58. Velasco M, et al. MRI of metaplastic carcinoma of the breast. AJR Am J Roentgenol 2005; 184: 1274 – 8. 59. Chang YW, et al. Magnetic resonance imaging of metaplastic carcinoma of the breast: sonographic and pathologic correlation. Acta Radiol 2004; 45: 18 – 22. 60. Okushiba S, et al. A case of spindle cell carcinoma of the breast – long survival achieved by multiple surgical treatment. Breast Cancer 2001; 8: 238 – 42. 61. Chao TC, et al. Metaplastic carcinomas of the breast. J Surg Oncol 1999; 71: 220 – 5. 62. Christensen L, Schiodt T, Blichert-Toft M. Sarcomatoid tumours of the breast in Denmark from 1977 to 1987. A clinicopathological and immunohistochemical study of 100 cases. Eur J Cancer 1993; 29A: 1824 – 31. 63. Veronesi U, et al. Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N Engl J Med 2002; 347: 1227 – 32. 64. Jacobson JA, et al. Ten-year results of a comparison of conservation with mastectomy in the treatment of stage I and II breast cancer. N Engl J Med 1995; 332: 907 – 11. 65. Morrow M, et al. Factors predicting the use of breast-conserving therapy in stage I and II breast carcinoma. J Clin Oncol 2001; 19: 2254 – 62.
229
66. Staradub VL, Rademaker AW, Morrow M. Factors influencing outcomes for breast conservation therapy of mammographically detected malignancies. J Am Coll Surg 2003; 196: 518 – 24. 67. Zelek L, et al. Prognostic factors in primary breast sarcomas: a series of patients with long-term follow-up. J Clin Oncol 2003; 21: 2583 – 8. 68. Zhuang Z, et al. Identical clonality of both components of mammary carcinosarcoma with differential loss of heterozygosity. Mod Pathol 1997; 10: 354 – 62. 69. Sorlie T, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA 2001; 98: 10869 – 74. 70. Chang CC, et al. A human breast epithelial cell type with stem cell characteristics as target cells for carcinogenesis. Radiat Res 2001; 155: 201 – 7. 71. Nielsen TO, et al. Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res 2004; 10: 5367 – 74. 72. Korbling M, Estrov Z. Adult stem cells for tissue repair – a new therapeutic concept? N Engl J Med 2003; 349: 570 – 82. 73. Reis-Filho JS, Schmitt FC. p63 expression in sarcomatoid/metaplastic carcinomas of the breast. Histopathology 2003; 42: 94 – 5. 74. Reis-Filho JS, Schmitt FC. Taking advantage of basic research: p63 is a reliable myoepithelial and stem cell marker. Adv Anat Pathol 2003; 9: 280 – 9. 75. Yang A, McKeon F. p63 and p73: p53 mimics, menaces and more. Nat Rev Mol Cell Biol 2000; 1: 199 – 207. 76. Barbareschi M, et al. p63, a p53 homologue, is a selective nuclear marker of myoepithelial cells of the human breast. Am J Surg Pathol 2001; 25: 1054 – 60. 77. Leibl S, et al. Metaplastic breast carcinomas: are they of myoepithelial differentiation?: immunohistochemical profile of the sarcomatoid subtype using novel myoepithelial markers. Am J Surg Pathol 2005; 29: 347 – 53. 78. Lightner VA, Marks JR, McCachren SM. Epithelial cells are an important source of tenascin in normal and malignant human breast tissue. Exp Cell Res 1994; 210: 177 – 84. 79. Ishihara A, et al. Tenascin expression in cancer cells and stroma of human breast cancer and its prognostic significance. Clin Cancer Res 1995; 1: 1035 – 41. 80. Rao K, et al. Production of spindle cell carcinoma by transduction of H-Ras 61L into immortalized human mammary epithelial cells. Cancer Lett 2003; 201: 79 – 88. 81. Han AC, et al. Distinct cadherin profiles in special variant carcinomas and other tumors of the breast. Hum Pathol 1999; 30: 1035 – 9. 82. Peralta Soler A, et al. P-cadherin expression in breast carcinoma indicates poor survival. Cancer 1999; 86: 1263 – 72. 83. Karreth F, Tuveson DA. Twist induces an epithelial-mesenchymal transition to facilitate tumor metastasis. Cancer Biol Ther 2004; 3: 1058 – 9. 84. Valsesia-Wittmann S, et al. Oncogenic cooperation between H-Twist and N-Myc overrides failsafe programs in cancer cells. Cancer Cell 2004; 6: 625 – 30. 85. Gilhooly EM, Rose DP. The association between a mutated ras gene and cyclooxygenase-2 expression in human breast cancer cell lines. Int J Oncol 1999; 15: 267 – 70.
Section 4 : Breast Cancer
19
Tubular Carcinoma
Melinda E. Sanders, Ingrid A. Mayer and David L. Page
INTRODUCTION The morphologic features of tubular carcinoma were first described over a century ago by Cornil and Ranvier; but received little attention until the last 20 years because they are rarely symptomatic. Only recently with the advent of mammographic screening programs have these usually small, indolent, nonpalpable tumors regularly reached clinical attention.1 The orderly histologic pattern composed of small, angulated tubules resulted in the original diagnosis of “well-differentiated carcinoma”. The term “tubular carcinoma” is now preferred as it describes the lesional architecture of this “special type” of breast cancer. There has been considerable debate regarding the precise histologic criteria and the proportion of tubular structures required to establish the diagnosis of a “pure” tubular carcinoma.2 – 9 Several studies have now demonstrated that when >90% purity of pattern is observed, an excellent prognosis can be expected even in the presence of a positive lymph node.10 – 12 Patients with pure tubular carcinoma can expect survival rates similar to that of the general population.10
BIOLOGY AND EPIDEMIOLOGY Pure tubular carcinoma accounts for 1–2% of invasive cancers in the premammographic era series.5,6 The frequency of cases now reported by mammographic screening is 9–19%,1,13 – 18 with higher percentages noted in series of T1 lesions. It should also be noted that the definition for tubular carcinoma in some series was not as stringent as the 90% rule which others and we standardly require (see “Pathology” section). The biologic origin of tubular carcinoma is unknown; however, the resemblance of epithelial elements in some radial scars to cancer and the presence of cancer in some radial scars19 – 21 has prompted several authors to propose that radial scars represent an early stage in the evolution of invasive mammary carcinoma,22,23 which is usually of tubular histology but none of these authors have been able to provide definitive evidence to indicate that radial scars are in and of themselves premalignant. In addition, a recent study of a large cohort of women who underwent benign breast biopsies and were diagnosed with radial scar
were shown to have a modest elevation in subsequent breast cancer risk; however, this risk could be attributed to the category of coexistent proliferative disease.24
PATHOLOGY Grossly, no specific features, except small size, distinguish tubular cancer from other “no specific type”, often referred to as “ductal”, or mixed tumor types. Tubular carcinomas usually measure between 2.0 mm and 2.0 cm with the majority measuring less than 1.0 cm.6,7,25 Occasional cases can reach 3.0 cm in size. Histologically, tubular cancer is characterized by distinct tubular structures lined by a single layer of epithelial cells. The tubules may be round, oval, or “bent teardrop” shaped (see Figures 1 and 2). The epithelial cells lining the tubules are small and round with minimal pleomorphism and mitotic figures, if any, are rare (see Figure 3). Many cases have a minority of the tumor with more than a single cell layer forming central lumena, like those seen in invasive cribriform carcinoma. Apical snouts are seen in at least one-third of cases26 but are not specific (see Figure 4). Fine calcifications may be present in the lumena occasionally.25,27 Historically, two morphologic subtypes have been described, which do not have any clinical significance. The “pure type” has a dominantly stellate configuration with radiating fibrous arms containing neoplastic tubules located peripherally and stromal elastosis and hyalinized fibrosis present centrally. A consequence of this architecture is that the actual tumor volume of these lesions may be significantly less than that indicated by the overall tumor measurements mammographically and grossly. The “sclerosing type” is more diffuse and ill-defined, with a diffuse, haphazard infiltration of tubules within a desmoplastic stroma without central hyalinization.28,29 Lowgrade cribriform or micropapillary ductal carcinoma in situ and atypical ductal hyperplasia are usually found in association with tubular carcinoma and are usually centrally located within the lesion. Occasionally, tubular carcinoma is associated with atypical lobular hyperplasia or lobular carcinoma in situ.7 Tubular carcinoma is essentially always estrogen (see Figure 5) and progesterone receptor positive, has a lowgrowth fraction as demonstrated by Ki67 staining and rare if
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
TUBULAR CARCINOMA
231
any mitoses, and is Her-2/neu and epidermal growth factor negative (EGFR) negative.10,30 In some cases, the tumor may be seen in association with a radial scar or complex sclerosing lesion. In such cases, involvement is usually focal and recognized by the neoplastic tubules surrounding the normal structures also associated with the radial scars/complex sclerosing lesion, as well as extension of tubules into the adjacent adipose tissue (see Figure 6). Distinction of tubular carcinoma from adenosis or entrapped benign tubules in a sclerosing lesion may be facilitated by immunohistochemical staining for myoepithelial cells. In our experience, p63 (see Figure 7) and smooth muscle actin stains work best. These stains should be negative in tubular carcinoma but may also be negative in benign glands entrapped in sclerosis.
Figure 3 High-power photomicrograph showing the small, round tumor cells lining the carcinomatous tubules to have minimal pleomorphism and essentially absent mitotic activity.
Figure 1 Low-power photomicrograph of tubular carcinoma showing the characteristic angulated tubules surrounded by desmoplastic and sclerotic stroma and invading adipose tissue at the periphery.
Figure 4 High-power photomicrograph showing apical snouts which are present at the luminal aspect of the tumor cells in at least 30% of cases.
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS
Figure 2 The characteristic round or bent teardrop shaped tubules of tubular carcinoma are lined by a single layer of epithelial cells. The intervening stroma shows areas of elastosis and sclerosis.
Tubular carcinomas are usually detected by their spiculated appearance and occasionally by the presence of microcalcifications but may have subtle mammographic findings.25,27 They are usually incidental findings on screening mammography and not associated with a palpable mass on physical exam. In comparison with “no special type” carcinomas, “pure” tubular carcinomas are more likely to occur in postmenopausal patients and be smaller in size. Axillary node metastases occur less frequently, 0–13% of cases, and when observed, usually involve a single node.5,27,31,32 Multicentricity of tubular carcinoma in the ipsilateral breast has been reported,33 as well as an association with contralateral cancers occurring before and after the detection of the tubular cancer;5,33 however, these contralateral cancers are
232
BREAST CANCER
Figure 5 Immunohistochemical staining for estrogen receptor shows strong positivity in 100% of tubular carcinoma nuclei.
Figure 6 Low-power photomicrograph showing the neoplastic tubules of tubular carcinoma infiltrating the normal structures including usual type hyperplasia (center) and a papilloma (upper left) within a radial scar. Note the presence of a well-delineated myoepithelial layer surrounding the normal structures.
not restricted to tubular carcinomas.5 When several separate tubular carcinomas present in a single breast, they are often in the same segment and are linked by a common, usually low-grade ductal carcinoma in situ (DCIS). Because tubular areas are found focally in many breast cancers, specific criteria for the diagnosis of tubular carcinoma must be established so that the excellent prognosis applies. We reserve the diagnosis of “pure” tubular carcinoma for those tumors in which the classic tubules represent 90% or more of the tumor, and the remainder of the tumor shows the same well-differentiated morphology.8 This is in agreement with the current WHO classification.34 We regard those tumors containing between 70–89% tubules as tubular variants and tumors containing less than 70% of the classic tubules as “no specific type tumors” with tubular features.
Figure 7 p63 Immunohistochemical stain for basal cells shows positive nuclear staining of myoepithelial cells in normal ducts and ductules, as well as ducts with atypical ductal hyperplasia whereas the neoplastic tubules of tubular carcinoma show absence of staining.
This means that tubular variants may be regarded as a special, more orderly, subset of tumors that would be assigned a grade of 1 for the tubular component of the Nottingham grade. Justification for these classifications are multiple studies indicating that tubular histology is of prognostic value, and that the greater the purity of the pattern the more favorable the prognosis. Fifty-four women in a series reported by Cooper et al.5 with tumors composed purely of the characteristic low-grade, angulated tubules were still alive after 15 years follow-up, regardless of tumor size. In contrast, almost half of the patients whose carcinomas were a mixture of tubular and other types of carcinomas died within the 15-year followup period. This study showed the utility of recognizing tubular carcinoma by comparing outcome with that of other cases with similar but not precisely the same histologic characteristics. This study was one of the first to demonstrate the importance of subset analysis in the prognostication of breast cancer. A later study by Leibman et al. found no local or distant recurrences in 12 “pure” tubular carcinomas (defined as >90% tubules) after treatment by either lumpectomy to negative margins or mastectomy.27 A more recent study of women by Kader et al. in the Breast Cancer Outcomes Database (BCOD) maintained by the British Columbia Cancer Agency, Vancouver, British Columbia, compared outcomes in 171 “pure” tubular carcinomas, as defined using the current WHO criteria, to 386 low-grade ductal carcinomas and found a lower rate of local recurrence in tubular carcinomas (0.8% vs 4.5%) and a trend toward a lower rate of systemic relapse (4.3% vs 9.7%) but no difference in disease-specific survival (95.7% vs 94.7%) over a 6-year follow-up.35 This improved outcome was seen in the tubular group despite the fact that the low-grade ductal group was treated more aggressively. The women in the tubular group were almost twice as likely to have been treated with
TUBULAR CARCINOMA
breast-conserving therapy alone. Also, 38.6% of the lowgrade ductal group received chemotherapy or chemotherapy plus tamoxifen versus 21.1% of the tubular group. Several additional authors found the presence of a 75% tubular component to signify a more favorable prognosis than “no specific type” carcinomas,3,6,25 although not as well as those categorized as “pure”,34 justifying the recognition of the tubular variant category. Using these criteria, Elson found no patients with local or distant recurrence including four patients with positive lymph nodes at presentation.25 In contrast, Cabral et al. found no difference in presentation or outcome between pure tubular carcinomas and mixed tumors, using a definition of >95% tubularity for pure tubular carcinomas and 75–95% tubularity for mixed tumors. In this study of 22 pure and 22 mixed tumors, there were only two local recurrences in the pure tubular group, one managed by reexcision and another accompanied by systemic disease. These results are difficult to compare with those of other studies because their mixed group contained cases, such as those with >90% tubules and mixed tubular and cribriform carcinomas, that were placed in the pure category by other authors,8,36 as can be seen from the photomicrographs included in the paper.11 Inclusion of cases regarded as pure by current definition in their mixed group likely explains the lack of a difference in the outcome. An additional diagnostic consideration is the coexistence of mixed tubular and invasive cribriform carcinomas.8,36 Essentially, any combination of these two elements conveys the same excellent prognosis as that of a pure tubular carcinoma. This is an important consideration as many such cases may be assigned to the no specific type group by an inexperienced pathologist or clinician and this potentially useful prognostic information will be lost.
TREATMENT The importance of tubular carcinoma lies with therapeutic decisions for individual patients. With the increasing use of mammographic screening, the number of tubular cancers has increased significantly and their average size is about half of those found by palpation.37 Tubular cancers now represent a significant proportion of incidence cancers.38 The prognostic implications of tubular carcinoma are excellent and can be used to avoid over-treatment when strict criteria are used to recognize and carefully document the histologic features. In fact, patients with pure tubular carcinoma can expect survival rates similar to those of the general population.10 In most cases, complete excision of the lesion should be sufficient therapy for localized lesions. Baker stated in 1990 that pure tubular carcinomas less than 1 cm in size could be treated with excision alone.39 The risk of local recurrence is so low that adjuvant radiation is probably unnecessary. This is further supported by the fact that any later events in these patients are regularly of low grade, also estrogen receptor positive, and more easily detected in the unirradiated breast. Rare local recurrences can usually be easily managed by reexcision. When lymph nodes are involved at presentation, 8–17% of most series,5,31,32 they tend to be confined to one or two nodes and these patients still have
233
an excellent prognosis. Importantly, the presence of positive lymph nodes does not appear to affect disease-free survival (DFS) and therefore a full lymph node dissection is not helpful in providing prognostic information in most cases of tubular carcinoma10,12 and may introduce unnecessary morbidity. We regard the use of adjuvant radiation or chemotherapy, as well as axillary node dissection, unnecessary in the overwhelming majority of cases,10,30 especially since antiestrogen therapy is usually given to these women. Although a retrospective review of case pathology was not conducted, a series of 444 tubular cancers reported from the University of Texas Health Science Center in San Antonio, Texas, demonstrated 5-year DFS and 5-year overall survival (OS) rates of 95% and 91% for node-negative patients and rates of 94% and 92% for node-positive patients, respectively. In addition, a subset analysis of patients who received adjuvant endocrine (29%) or chemotherapy (10%) demonstrated no significant difference in DFS or OS rates regardless of nodal status. The only study of patients with tubular carcinoma treated with breast-conserving therapy with central pathologic review, reports 2 of 28 patients having a local recurrence and no distant failures at 10 years despite the fact that 17% of patients presented with positive lymph nodes.40 As a result of poor case definition, several studies have suggested that radiation and even chemotherapy may be necessary for some patients with tubular carcinoma given the propensity for distant metastasis.12,41,42 However, these studies did not have central pathology review, did not list the percentage of tubular differentiation used to diagnose tubular carcinoma, and several acknowledge the inclusion of intermediate-grade tumors.42,43 In addition, selection of cases spanned a 30- to 40-year period in these studies, during which criteria for diagnosis of tubular carcinoma have evolved considerably.12,41,42 Despite the bias of poor classification, Livi et al.12,41,42 showed no difference in locoregional failure between patients who had received adjuvant radiation therapy and those who had not.
PROGNOSIS When a strict definition is used, tubular carcinoma has an excellent long-term prognosis which in some series is similar to that of age-matched women without breast cancer.10 Recurrence after mastectomy or breast-conserving therapy is rare. Thurman et al.40 found no difference in site of first failure among cases of tubular, mucinous, or medullary carcinomas treated with breast-conservation therapy and followed up for >10 years; however, rates of recurrence were lowest among tubular carcinomas with only 2/28 patients with local recurrence that was easily managed by reexcision and no instances of distant recurrence despite the fact that 17% of cases were associated with 1–3 positive lymph nodes. Thus, disease-specific survival was 100% for tubular carcinomas. We can find no instances of death from breast cancer after treatment of a clearly documented “pure” tubular carcinoma less than 1.0 cm in diameter. We would argue that the three studies in the literature reporting deaths from tubular
234
BREAST CANCER
carcinoma do not strictly define their criteria or are cases, which are not “pure” by current definition. Cabral et al.11 describe a woman with a 7.0 mm tubular carcinoma (defined as >95% tubules but with no mention of tumor grade) treated with modified radical mastectomy and 16 negative axillary nodes, who developed a local recurrence and systemic disease, dying 87 months after her initial diagnosis. Winchester 1996 et al.43 also report a single death from a supposed 6.0 mm pure tubular carcinoma. Although this study did include central pathology review, their criteria for tubular carcinoma required >80% tubularity and cases with intermediate-grade nuclei were included. Cases were then subdivided into pure tubular carcinoma and mixed pattern lesions, although neither definition was further defined. Three of the four cases which had a recurrence were in the mixed pattern category and three of four cases were intermediate combined histologic grade, although the grades of individual tumors, including the supposed pure tubular carcinoma which resulted in death following metastasis to bone, cannot be determined from the data. In support of our contention, Peters et al. found distant metastases to occur only when no special type carcinoma constituted over 25% of the lesion.9
AUTHORS’ RECOMMENDATIONS For pure tubular carcinomas, the excellent prognosis indicates that conservative but complete surgical excision is adequate therapy for the overwhelming majority of cases. Since the addition of radiation or chemotherapy does not improve DFS or OS, they should not be advocated. Similarly, involvement of axillary lymph nodes is an uncommon finding and does not adversely affect outcome. We would advocate performance of a sentinel node biopsy at the time of definitive surgery only for large tumors, which may not have been adequately sampled at the time of core biopsy. Because pure tubular carcinomas are estrogen receptor positive, endocrine therapy with tamoxifen or aromatase inhibitors (for postmenopausal women) is usually considered, although it should be recognized that such therapy likely benefits most women by reducing the incidence of subsequent contralateral tumors rather than impacting their survival. On the basis of limited series and long-term follow-up data, there seems to be no role for adjuvant chemotherapy in the treatment of pure tubular cancers, regardless of tumor size or lymph node involvement.
REFERENCES 1. Patchefsky AS, et al. The pathology of breast cancer detected by mass population screening. Cancer 1977; 40(4): 1659 – 70. 2. Carstens PH. Tubular carcinoma of the breast. A study of frequency. Am J Clin Pathol 1978; 70(2): 204 – 10. 3. Carstens PH, et al. Tubular carcinoma of the breast. A long term followup. Histopathology 1985; 9(3): 271 – 80. 4. Fisher ER, et al. The pathology of invasive breast cancer. A syllabus derived from findings of the national surgical adjuvant breast project (protocol no. 4). Cancer 1975; 36(1): 1 – 85. 5. Cooper HS, Patchefsky AS, Krall RA. Tubular carcinoma of the breast. Cancer 1978; 42(5): 2334 – 42.
6. McDivitt RW, Boyce W, Gersell D. Tubular carcinoma of the breast. Clinical and pathological observations concerning 135 cases. Am J Surg Pathol 1982; 6(5): 401 – 11. 7. Oberman HA, Fidler WJ Jr. Tubular carcinoma of the breast. Am J Surg Pathol 1979; 3(5): 387 – 95. 8. Page DL, Anderson TJ. Diagnostic Histopathology of the Breast. Edinburgh, Scotland: Churchill Livingstone, 1987. 9. Peters GN, Wolff M, Haagensen CD. Tubular carcinoma of the breast. Clinical pathologic correlations based on 100 cases. Ann Surg 1981; 193(2): 138 – 49. 10. Diab SG, et al. Tumor characteristics and clinical outcome of tubular and mucinous breast carcinomas. J Clin Oncol 1999; 17(5): 1442 – 8. 11. Cabral AH, et al. Tubular carcinoma of the breast: an institutional experience and review of the literature. Breast J 2003; 9(4): 298 – 301. 12. Kitchen PR, et al. Tubular carcinoma of the breast: prognosis and response to adjuvant systemic therapy. ANZ J Surg 2001; 71(1): 27 – 31. 13. Feig SA, et al. Analysis of clinically occult and mammographically occult breast tumors. AJR Am J Roentgenol 1977; 128(3): 403 – 8. 14. Anderson TJ, Alexander FE, Forrest PM. The natural history of breast carcinoma: what have we learned from screening? Cancer 2000; 88(7): 1758 – 9. 15. Anderson TJ, et al. Influence of annual mammography from age 40 on breast cancer pathology. Hum Pathol 2004; 35(10): 1252 – 9. 16. Anderson TJ, et al. Comparative pathology of prevalent and incident cancers detected by breast screening. Edinburgh Breast Screening Project. Lancet 1986; 1(8480): 519 – 23. 17. Anderson TJ, et al. Comparative pathology of breast cancer in a randomised trial of screening. Br J Cancer 1991; 64(1): 108 – 13. 18. Rajakariar R, Walker RA. Pathological and biological features of mammographically detected invasive breast carcinomas. Br J Cancer 1995; 71(1): 150 – 4. 19. Douglas-Jones AG, Pace DP. Pathology of R4 spiculated lesions in the breast screening programme. Histopathology 1997; 30(3): 214 – 20. 20. Frouge C, et al. Mammographic lesions suggestive of radial scars: microscopic findings in 40 cases. Radiology 1995; 195(3): 623 – 5. 21. Sloane JP, Mayers MM. Carcinoma and atypical hyperplasia in radial scars and complex sclerosing lesions: importance of lesion size and patient age. Histopathology 1993; 23(3): 225 – 31. 22. Fisher ER, et al. A nonencapsulated sclerosing lesion of the breast. Am J Clin Pathol 1979; 71(3): 240 – 6. 23. Linell F, Ljungberg O, Andersson I. Breast carcinoma. Aspects of early stages, progression and related problems. Acta Pathol Microbiol Scand Suppl 1980; 272: 1 – 233. 24. Sanders M, et al. Interdependence of radial scar and proliferative disease with respect to invasive breast cancer risk in benign breast biopsies. Lab Invest 2002; 82(1): 50A. 25. Elson BC, et al. Tubular carcinoma of the breast: mode of presentation, mammographic appearance, and frequency of nodal metastases. AJR Am J Roentgenol 1993; 161(6): 1173 – 6. 26. Tavassoli F. Infiltrating carcinomas, common and familiar special types. In Tavassoli F (ed) Pathology of the Breast. Norwalk, Connecticut: Appleton and Lange, 1992: 293 – 294. 27. Leibman AJ, Lewis M, Kruse B. Tubular carcinoma of the breast: mammographic appearance. AJR Am J Roentgenol 1993; 160(2): 263 – 5. 28. Carstens PH, et al. Tubular carcinoma of the breast: a clinicopathologic study of 35 cases. Am J Clin Pathol 1972; 58(3): 231 – 8. 29. Parl FF, Richardson LD. The histologic and biologic spectrum of tubular carcinoma of the breast. Hum Pathol 1983; 14(8): 694 – 8. 30. Papadatos G, et al. Probability of axillary node involvement in patients with tubular carcinoma of the breast. Br J Neurosurg 2001; 88(6): 860 – 4. 31. Berger AC, et al. Axillary dissection for tubular carcinoma of the breast. Breast J 1996; 3: 204 – 8. 32. Deos PH, Norris HJ. Well-differentiated (tubular) carcinoma of the breast. A clinicopathologic study of 145 pure and mixed cases. Am J Clin Pathol 1982; 78(1): 1 – 7. 33. Lagios MD, Rose MR, Margolin FR. Tubular carcinoma of the breast: association with multicentricity, bilaterality, and family history of mammary carcinoma. Am J Clin Pathol 1980; 73(1): 25 – 30. 34. Tavassoli F, Devilee P (eds). Tumors of the Breast and Female Genital Organs, 1st ed. Lyon, France: IRC Press, 2003.
TUBULAR CARCINOMA 35. Kader HA, et al. Tubular carcinoma of the breast: a population-based study of nodal metastases at presentation and of patterns of relapse. Breast J 2001; 7(1): 8 – 13. 36. Elston CW, Ellis IO. The Breast, 3rd ed. London: Churchill Livingstone, 1998. 37. Lagios MD. Multicentricity of breast carcinoma demonstrated by routine correlated serial subgross and radiographic examination. Cancer 1977; 40(4): 1726 – 34. 38. Anderson WF, Chu KC, Devesa SS. Distinct incidence patterns among in situ and invasive breast carcinomas, with possible etiologic implications. Breast Cancer Res Treat 2004; 88(2): 149 – 59. 39. Baker RR. Unusual lesions and their management. Surg Clin North Am 1990; 70(4): 963 – 75.
235
40. Thurman SA, et al. Outcome after breast-conserving therapy for patients with stage I or II mucinous, medullary, or tubular breast carcinoma. Int J Radiat Oncol Biol Phys 2004; 59(1): 152 – 9. 41. Leonard CE, et al. Excision only for tubular carcinoma of the breast. Breast J 2005; 11(2): 129 – 33. 42. Livi L, et al. Tubular carcinoma of the breast: outcome and locoregional recurrence in 307 patients. Eur J Surg Oncol 2005; 31(1): 9 – 12. 43. Winchester DJ, et al. Tubular carcinoma of the breast. Predicting axillary nodal metastases and recurrence. Ann Surg 1996; 223(3): 342 – 7.
Section 5 : Thoracic Tumors
20
Thymoma and Thymic Carcinoma
Annette M. Moore, Christopher J. Sweeney, Mark R. Wick and Patrick J. Loehrer
HISTORICAL BACKGROUND The thymus is an enigmatic structure located in the anterior mediastinum; whose function is to facilitate the differentiation and maturation of T lymphocytes. As such, the thymus requires a complex microenvironment and is composed of specialized epithelial cells. Thus, the thymus gland serves an integral role in the immune process and alteration of its controlled function has been associated with a protean array of disease manifestations. Thymic epithelial tumors are rare and National Cancer Institute’s Surveillance, Epidemiology and End Results (SEER) tumor registry estimates the incidence of thymoma in the United States to be approximately 0.15 per 100 000 person-years.1 Thymomas, thymic carcinomas, thymic carcinoids, and thymolipomas contain epithelial components; however, only the first two are usually regarded as tumors exhibiting differentiation toward thymic epithelium. Accordingly, this chapter will solely concern itself with these lesions. Thymomas may be encapsulated tumors, which present as space-occupying lesions or alternatively they can be locally invasive or systemically disseminated at presentation. Despite this, they still have a bland cytological appearance at all stages. In contrast, thymic carcinomas have overtly malignant cytologic features. Thymic carcinomas can present in the same manner as thymomas, except that they usually present with invasive or metastatic disease and as such are associated with a much poorer prognosis.2,3 The traditional histologic classification of thymomas was based on the proportion of nonneoplastic lymphoid cells to neoplastic epithelial cells. Using this system, there did not appear to be a correlation with prognosis. More recently the World Health Organization (WHO) classification has been adopted by many pathologists. A discussion comparing and contrasting the WHO type, a clinicopathogical classification, and the more traditional descriptive terminology will be presented. Also in this chapter, thymomas and thymic carcinomas will be anatomically staged according to the system devised by Masaoka et al.3 (see Table 1). In this system, stage II and stage III tumors are further subdivided into those that are completely and incompletely resected. As a point of clarification, thymomas have been referred
to as “encapsulated” and “benign” thymomas if there is no invasion, and “malignant” if there is infiltration into other intrathoracic structures or systemic dissemination. Furthermore, the term “type I malignant thymoma” has been used to label locally invasive or metastatic thymoma, whereas thymic carcinomas have been termed “type II malignant thymomas.”2 However, we prefer the more simple descriptors “invasive thymoma” and “thymic carcinoma.”
ANATOMY In the first trimester the thymus gland arises form the ventral portion of the third pharyngeal pouch as a paired epithelial structure, in close association with the parathyroid glands, which develop posteriorly.4,5 The thymic epithelial cells are derived from the endoderm.6 The rudimentary thymus enlarges and descends into the anterosuperior mediastinum by week eight. The thymus gland lies adjacent to the pericardium (inferiorly) and the great vessels (anteriorly), and its superior aspect is in the root of the neck. Precursor lymphocytes migrate to the thymus during week eight, at the time when the microscopic architecture becomes apparent. This features a cortex, a medulla, and perivascular spaces buttressed by fibrous septa. Moreover, the lymphoid cells are intimately admixed with the epithelial cells but are separated by the perivascular spaces. This partitioning forms the blood–thymus barrier. Ultrastructural evaluation of the epithelial cells in the thymus has shown that there is a close association between different types of epithelial cells and the various steps in the maturation process of lymphocytes.6,7 Six different types of thymic epithelium have been identified to date.4 The outer cortex is the location for lymphoblastogenesis involving large immature lymphocytes, the inner cortex contains more mature thymocytes, and the medulla harbors the most mature T cells, which are similar to those in the peripheral blood.6 The thymus is the site of specialized lymphoid differentiation and “positive selection.”4,8 This gland generates large numbers of cortical thymocytes (lymphocytes) having both CD4 and CD8 molecules showing specificity for all major histocompatibility complex (MHC) molecules expressed in
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
238
THORACIC TUMORS
Table 1 Staging systems.
Masaoka staging: I II
III IVa IVb GETT Classification Stage I IA IB
Stage II Stage III IIIA IIIB Stage IV IVA IVB TNM T factor T1 T2
T3 T4 N factor N0 N1 N2 N3 M factor M0 M1
Macroscopically completely encapsulated and microscopically no capsular invasion (1) Macroscopic invasion into surrounding fatty tissue, mediastinal pleura, or both (2) Microscopic invasion into capsule Macroscopic invasion into neighboring organ, such as pericardium, great vessels, or lung Pleural or pericardial dissemination Lymphogenous or hematogenous metastasis.
Encapsulated tumor, totally resected Macroscopically encapsulated tumor, totally resected, but the surgeon suspects mediastinal adhesions and potential capsular invasion Invasive tumor, totally resected Invasive tumor subtotally resected Invasive tumor, biopsy Supraclavicular metastasis or distant pleural implants Distant metastasis
Macroscopically completely encapsulated and microscopically no capsular invasion Macroscopic adhesion or invasion into surrounding fatty tissue or mediastinal pleura or microscopic invasion into the capsule Invasion into neighboring organs such as the pericardium, great vessels, or lung Pleural or pericardial dissemination No lymph node metastasis Metastasis to anterior mediastinal lymph nodes Metastasis to intrathoracic lymph nodes except anterior mediastinal lymph nodes Metastasis to extrathoracic lymph nodes No hematogenous metastasis Hematogenous metastasis
the human species. The thymus actively culls a small proportion of T cells, which recognize “self MHC” molecules. These immature elements, which coexpress CD4 and CD8, are thought to bind to epithelial (self) cells and receive a protective signal, which allows them to survive and be exported. Other CD4+ and CD8+ cells die and the surviving cells differentiate into either CD4 or CD8 cells. Physiologically, the latter classes of lymphocytes recognize peptides as foreign if they do not have the same “self MHC” repertoire and tolerate particles that possess the same determinants. Some of these T cells bind very avidly to the epithelial (self) cells, are thus recognized as “autoaggressive” and subsequently undergo apoptosis.4,8 If not eliminated, they can attack “self” and produce autoimmune disease. In the neonate, the thymus has a maximal relative weight but it reaches a maximal absolute weight of roughly 35 g during puberty. The gland then undergoes gradual involution until it is only a small remnant structure in the adult, replaced
mostly by adipocytes. Ectopic thymic tissue has been recognized in many locations in the mediastinum and neck presumably due to aberrant or arrested migration.5 The most common sites include retrocarinal and the aortopulmonary window.
BIOLOGY AND EPIDEMIOLOGY Biology The thymic microenvironment has been implicated in the genesis of thymomas and thymic carcinomas. Fibronectin and laminin are proteoglycans and are part of the ground substance of the extracellular matrix. They interact with T lymphocytes, thymic epithelial cells, various thymic hormones, and other factors. The absence of these ground substances may be associated with invasion more strongly than histologic appearance alone, although they are more often lacking in “cortical” thymomas.9 The behavior of thymomas is determined by their epithelial components.10 Although patients with lymphocytepredominant tumors have a better survival compared with those who have epithelial-predominant tumors, lymphocyte content does not predict invasiveness.11 Thymomas retain some functions of the normal thymus and may be able to induce the differentiation and homing of lymphocytes.2 However, close observation reveals subtle differences.12 Minor phenotypic abnormalities in Leu-2 and Leu-3 antigen expression were seen in 3 of 15 tumors in one series and six other cases showed aberrations of cortical and medullary differentiation. A lack of class II major histocompatibility complex antigens was associated with decreased lymphoid content and diminished Leu-1 expression in cortical thymocytes.12 Moreover, a proliferation of the lymphocytes in thymomas has been documented with the Ki67 antibody; it is 35–80% higher than age-matched controls. It is thought that this process may play a role in the pathogenesis of autoimmune diseases associated with thymoma.13 Thymic carcinomas are less differentiated and functionally inert,2 possibly explaining why they are rarely associated with autoimmune conditions. Genetic abnormalities are poorly characterized. Several studies have shown loss of genetic material or loss of heterogeneity (LOH) in the long arm of chromosome 6.14 – 16 The only cytogenetic abnormality that has been observed is the presence of a translocation between chromosomes 15 and 19 in three cases of thymic carcinoma.17,18 Whether thymic carcinomas arise, at least sometimes, by clonal evolution from thymomas is under debate. A potential link is thymomas with modest cytologic atypia, which some investigators classify as “well-differentiated thymic carcinmoma” (WDTC). This entity was initially described by Kirchner et al.19 “WDTC” otherwise retains most histologic features of classic thymoma. This entity may represent the conceptual intermediary between thymoma and thymic carcinoma. Although the term “WDTC” has not been broadly accepted, proponents of this nomenclature argue that it is associated with invasive behavior in 83% of cases. On the other hand, critics point out that the frequency of paraneoplastic syndromes (predominantly myasthenia gravis [MG]), that is associated with “WDTC” is incongruent
THYMOMA AND THYMIC CARCINOMA
with a diagnosis of carcinoma. Moreover, we consider it to be a variant of thymoma because it retains the general morphologic features of this tumor. One might posit that there is a spectrum of genetic abnormality in thymic epithelial tumors, with marked karyotypic aberrancy conferring a greater degree of malignancy and a decreased likelihood of paraneoplasia. Anecdotal support for this notion comes from the observation that a small number of patients with biopsy proven thymoma may later develop thymic carcinoma.20 – 22 Although rare, thymic carcinoma may arise in any histologic type of thymoma, including the more benign histology of spindle cell thymoma. If necrosis is found in a thymoma, the pathologist should search for malignant changes such as specifically increased expression of epithelial membrane antigen (EMA), p53 protein, cytokeratin subtypes, or loss of CD99+ immature T lymphocytes.21
Epidemiology The cause of thymoma is unknown. Evidence implicating prior Epstein-Barr viral infection as a risk factor for thymic carcinoma includes the isolation of defective viral genomes in thymic carcinoma.23 – 25 It is thought that defective virus can disrupt Epstein-Barr virus (EBV) latency, bring about EBV reactivation and an increase in EBV antibody levels, and result in malignant progression of infected cells, such as those in the thymus.24 There may be geographical differences in this phenomenon, in analogy to EBV-associated nasopharyngeal cancer.26 EBV is felt to be specifically associated with thymic lymphoepithelioma-like carcinomas in young individuals.25 Support for this argument is built on the knowledge that nasopharyngeal carcinomas are also lymphoepithelial, can occur in younger patients and also arise from the primitive pharynx.27 This raises the possibility that EBV causes thymic carcinoma only in people who are infected at an early age. However, the data are limited and cannot confirm whether there is increased incidence in Asian countries. Nonetheless, there is some evidence that the incidence is higher in Asians/Pacific islanders and AfricanAmericans than in Caucasians.1 These differences may arise from genetic polymorphisms. The distribution of alleles at the human leukocyte antigen (HLA) locus on chromosome 6 varies across racial groups.28 Class I and class II HLA proteins are highly expressed on thymic epithelial cells and further studies need to be performed for a better understanding of a possible predisposition to thymoma.
239
keratinization and microcystification, yielding the structures known as “Hassall’s corpuscles.” Mast cells are also found in abundance throughout both the cortex and medulla, and these become more notable with aging as the lymphocyte content of the gland decreases. Indeed, the postpubertal thymus contains relatively few thymocytes, and instead is represented by a large amount of mature adipose tissue in which residual epithelial cells are embedded (see Figure 2).29,30 Thymic epithelium demonstrates some degree of morphologic variation, largely depending on whether they are located in one glandular subcompartment. Cortical epithelial cells have round-to-oval nuclear contours with vesicular chromatin and distinct small nucleoli, whereas those in the medulla more commonly assume a fusiform shape, contain dispersed chromatin, and manifest few, if any, nucleoli. In prepubescent individuals, thymocytes throughout the gland differ from the appearance of peripheral mature lymphocytes. The latter cells exhibit relatively enlarged nuclei with open chromatin patterns, discernible chromocenters, and folding
Figure 1 Normal thymic microarchitecture, showing peripheral, lymphocyte-rich cortical zone and a central medulla containing more numerous epithelial cells and Hassall’s corpuscles.
PATHOLOGY:HISTOPATHOLOGY OF NONNEOPLASTIC THYMUS AND PATHOLOGIC FEATURES OF THYMOMA The Nonneoplastic Thymus Microscopically, the thymus attains maturity during the first trimester of pregnancy. At that point, it has a multilobated appearance, with each lobule being composed of a cortex and a medulla (see Figure 1). The former of these two zones has a high lymphocyte-to-epithelial cell ratio, whereas the latter contains nearly an equal number of such elements. Clusters of epithelial cells in both subcompartments commonly undergo
Figure 2 Photomicrograph of thymic atrophy in an adult, demonstrating replacement of a large portion of the glandular parenchyma by adipose tissue.
240
THORACIC TUMORS
of the nuclear membranes. Mitoses are commonly seen in thymocytes as well.31 The structural relationship between intrathymic lymphocytes and the thymic epithelium is an intimate one, wherein elongated and branched cytoplasmic processes of epithelial cells are closely apposed to the plasmalemma of resident thymocytes. Because of the overall constituency of the cortex, there are relatively more epithelial cell extensions than karyons, with the reverse of that statement pertaining to the thymic medulla. Nonneoplastic morphologic abnormalities in the thymus principally are represented by: (i) hyperplasia, which almost always involves proliferation of intrathymic lymphocytes but not epithelial cells, and by (ii) dysplasia, where one observes only sparse thymocytes and abnormal aggregations of epithelium into rosettes or arborizing cords. The first of these two conditions – which also features the formation of lymphoid germinal centers – is closely associated with MG or Graves’ disease, and the second is linked to congenital immunodeficiency states.32
Thymoma Thymomas generally take the form of well-localized, nodular, multilobated masses in the anterosuperior mediastinum, usually with at least partial fibrous encapsulation (see Figure 3). However, they have been reported to rarely occur as primary lesions in other sites as well, both inside and outside the thorax. Such locations include the middle and posterior mediastinal compartments, the intrapulmonary or extrapulmonary pleura, and the neck.33 – 35 In fact, the term “SETTLE” (spindle cell epithelial tumors of thymiclike epithelium) has been named for intrathyroidal lesions with virtually all of the attributes of thymoma.36,37 The cut surfaces of thymomas show subdivision of the lesions by
Figure 3 Gross photograph of a resected, bisected, encapsulated thymoma. The parenchyma demonstrates numerous fibrous septa that subdivide the tumor into angular lobules.
fibrous bands, which generally intersect one another at acute angles. Spontaneous intralesional hemorrhage or necrosis is not usually apparent, but cystic changes may be prominent in selected examples. Indeed, such an alteration can be so striking in some instances – producing an image, which simulates that of unilocular or multilocular thymic cysts – that the pathologist must undertake exhaustive tissue sampling to document the presence of a neoplastic cell population.38 Typically, all of the cytoarchitectural features described above, in reference to the nonneoplastic thymus, pertain to variants of thymoma as well. By definition, thymomas are primary tumors in which the neoplastic epithelial cells are cytologically bland or, at most, biologically indeterminate. With that fact in mind, it follows that other epithelial thymic tumors manifesting cytologic malignancy must be classified differently (see below). Some thymomas are composed of epithelium, which resembles that of the nonneoplastic thymic cortex, and accordingly have been termed “cortical” thymomas by some observers (see Figure 4).39 On the other hand, others that are comprised of spindle cells with fusiform nuclei and dispersed chromatin have the attributes of so-called “medullary” thymomas (see Figure 5). Still another subset exhibits a mixture of these two cytologic morphotypes (“mixed” thymomas). However, it should be recognized that the microscopic variability of thymomas is considerable, perhaps second only to that of teratomas among all mediastinal neoplasms. Recognized secondary morphologic findings in the former group of lesions include a vast predominance of intratumoral lymphocytes; microcystic change; pseudoglandular formations; perivascular pseudorosettes; assumption of an organoid, endocrine-like substructure; hemangiopericytoma-like growth with branched stromal blood vessels; strikingly dense stromal vascularity with blood “lakes”, zones of loose lymphocytic aggregation (“medullary” differentiation),40 squamous metaplasia, storiform growth of spindle cells, rhabdomyoma-like differentiation, and focal nuclear atypia.41 With regard to the last of these possibilities, it must be acknowledged that selected
Figure 4 Photomicrograph of a “cortical” (lymphocyte predominant) thymoma in which larger thymic epithelial tumor cells are widely separated. These show slightly vesicular chromatin and small nucleoli.
THYMOMA AND THYMIC CARCINOMA
Figure 5 Histologic image of a “medullary” (epithelial-predominant/ spindle cell) thymoma. This lesion contains almost no lymphocytes, and the epithelial tumor cells have a fusiform appearance with dispersed nuclear chromatin and indistinct nucleoli.
cases strain the morphologic boundary between thymomas and thymic carcinomas (tumors with overt cytologic malignancy), and it is likely that at least a theoretical continuum exists between those entities. Nevertheless, it is our opinion that the “well-differentiated thymic carcinoma” (WDTC) of Kirchner et al.19 is more properly considered a form of thymoma with limited nuclear atypia. Because of the many morphologic differential diagnoses that are called to mind by the variations just cited, the pathologist may want to employ adjunctive diagnostic studies to
241
solidify an interpretation of thymoma. These typically center on the use of electron microscopy and immunohistochemistry, the results of which serve to define the presence of an epithelial tumor, which lacks – in almost all cases – any evidence of neuroendocrine differentiation.42 Histologic Subclassification of Thymomas (see Table 2) Histological classification of thymic tumors, historically and currently, remains one of the most controversial areas in pathology. Historically, the histologic subclassification of thymomas devised by Bernatz and colleagues49 divided thymomas into four groups based on their microscopic features: lymphocyte predominant (>66% lymphocytes), epithelial predominant (>66% epithelial cells), mixed lymphoepithelial (34–66% epithelial cells), and spindle cell. The last of these categories pertains to epithelial-predominant thymoma featuring a virtually pure population of fusiform cells. It merits reemphasis that thymoma must first be defined as a cytologically bland epithelial neoplasm for this scheme to have histopathologic utility. Its usefulness is in serving as a cue mechanism for well-defined histologic differential diagnostic problems concerning thymomas. With the exception of spindle cell thymomas, which typically pursue a benign course, the Bernatz system does not present prognostic information. In 1985, a construct proposed by Marino and MullerHermelink (the MMH classification)6 utilized the morphologic resemblance of neoplastic epithelial cells to subtypes of normal epithelial cells in the thymus.50 They presupposed that “medullary” thymomas carry a favorable prognosis, while “cortical” lesions have a relatively adverse evolution and “mixed cortical/medullary” thymomas have an intermediary behavior. As summarized recently by Shimosato and
Table 2 Histology.
Thymoma study
Patients, n 98
Verley and Hollmann
200
Lewis et al.60
283
Muller-Hermelink et al.39
Thymic carcinoma26,27 High grade Low grade WHO histologic typing44 – 48 A AB B1 B2 B3 C a
10-year survival (%). 15-year disease-specific survival. c Subgroup with invasion (%). d Median survival. e Percent invasive. b
58
96 237 122 269 90 92
Subgroups (%)
Clinical correlation
Type I: spindle and oval cell (30) Type II: lymphocyte rich (30) Type III: differentiated epithelial rich (33) Type IV: undifferentiated epithelial rich (equivalent to thymic carcinoma) (7) Predominantly lymphocytic (>66% lymphocytes) (25) Mixed lymphoepithelial (33 – 66% lymphocytes) (43) Predominantly epithelial (<33% lymphocytes) (25) Spindle cell (predominantly epithelial cells with prominent fusiform cells) (6) Cortical (43) Mixed: predominantly cortical (8) Mixed: common (36) Medullary (5) Mixed: predominantly medullary (8)
75a 75a 50a 0a 90b 80b 50b 100b 67c 0c 0c 0c 0c
Lymphoepithelial-like, small cell, large cell anaplastic, clear cell, sarcomatoid Keratinizing squamous, basaloid squamous, mucoepidermoid Predominant cells Epithelial cells (spindle) Epithelial/lymphocyte Lymphocyte Polygonal neoplastic epithelial cells with immature lymphocytes Neoplastic epithelial cells with minor portion of immature lymphocytes Thymic carcinoma
11.3 monthsd 25.4 monthsd 37%e 32%e 49%e 66%e 86%e 91%e
242
THORACIC TUMORS
Mukai41 it appears that the MMH system actually represents a derivative array of information that would already be available if the Bernatz scheme were used, albeit with dissimilar terminology. Also, significant problems with interobserver reproducibility of the MMH system have been noted.51 Because spindle cell (“medullary”) thymomas have long been known to behave innocuously, regardless of what specific adjectives are attached to them, it would seem worthwhile to reevaluate the claims of MMH proponents,52,53 after pure spindle cell neoplasms have been excluded from formal statistical analyses. In our personal experience with a large number of thymoma cases, the MMH system has not attained independent significance as a prognostic factor under those conditions. Although there is no means of pathologically classifying thymomas to correlate perfectly with biology (and hence prognosis), in 1999 the WHO agreed to a classification system based on the morphology of the epithelial cells, as well as the lymphocyte-to-epithelial cell ratio. The WHO classification separates the tumors into three types using letters A (“atrophic”: the thymic cells of adult life), B (“bioreactive”: the biologically active organ of the fetus and infant), and C (carcinoma). Type A Thymoma (‘‘Spindle Cell: Medullary’’) Type A thymoma corresponds to the spindle cell thymoma and the medullary thymoma of prior classifications. Most type A thymomas are encapsulated. Type AB Thymoma (Mixed) Another subset exhibits a mixture of two cytologic morphotypes: a lymphocyte-rich area (WHO type B thymoma) and a lymphocyte-poor area (WHO type A, spindle cell component). Type B Thymoma This tumor resembles the normal thymus. Type B thymomas are further subdivided (B1, B2, B3) based on the increasing epithelial/lymphocyte ratio and the emergence of atypia. Type C Thymoma [Pathology of primary thymic carcinomas (PTC)] Thymic carcinomas have been well recognized only in relatively recent times.20,54,55 Historically, it was often stated that the biologic potential of thymic tumors could not be predicted by histology56 but this opinion is only partially correct. It is true that conventional thymomas often demonstrate gross transcapsular invasion,57 – 59 and a small proportion show extrathoracic (but idiosyncratic) spread (“metastasizing thymoma”)60 as well. Nevertheless, these biologic events do not mandate a diagnosis of “carcinoma.” There is a distinctive subgroup of thymic epithelial neoplasms that exhibits obvious cellular anaplasia and aggressive behavior. This subgroup truly deserves the designation of “carcinomas.” Thymic carcinomas commonly lack the encapsulation or fibrous septation of thymomas. Their parenchyma is firm-to-hard and has a white-gray appearance, with frequent necrosis and hemorrhage. Basaloid squamous thymic carcinoma may associate itself with multilocular thymic cysts;61 another tumor variant, mucoepidermoid thymic carcinoma, has a gelatinous cut surface.62
The recognized subtypes of WHO type C thymoma (thymic carcinoma) include: basaloid squamous cell carcinoma (BSCC), keratinizing squamous cell carcinoma, nonkeratinizing squamous cell carcinoma, lymphoepithelioma-like squamous carcinoma, mucoepidermoid carcinomas (MECs), clear cell carcinoma, undifferentiated carcinoma, papillary carcinoma, and sarcomatoid carcinoma.43,63,64 These variants are discussed in some descriptive detail below, in light of their relative rarity and distinctive features. Basaloid Squamous Cell Carcinoma BSCC has the potential to involve the thymus primarily or by metastasis. Potential origins for secondary lesions include the oropharynx, hypopharynx, larynx, esophagus, lungs, and anorectal region.61 Thus, cautions should be exercised while concluding that BSCCs have originated in the mediastinum. Histologically, BSCC is composed of polygonal cells with high nucleocytoplasmic ratios, hyperchromatic round nuclei, and brisk mitotic activity. Nuclear “molding” is absent (see Figure 6). Moreover, this lesion may contain stromal mucin-containing pseudoglandular arrays resembling those of adenoid cystic carcinomas, as well as globular eosinophilic intercellular deposits of basement membrane material and areas of squamous differentiation with keratin “pearls.”55,61 The few cases reported as primary in the thymus had a tendency for association with multilocular thymic cysts (see Figure 7).61 Keratinizing Squamous Cell Carcinoma Keratinizing thymic squamous carcinoma (KTSC) is identical microscopically to its counterparts in other organs. It shows large polyhedral cells in nests and cords. Nuclei are vesicular or hyperchromatic, usually with obvious nucleoli (see Figure 8). Cytoplasm is eosinophilic, and incipient or well-formed keratin “pearls” are scattered throughout such lesions.41,63,65,66 A potential factor of consternation in cases of thymic KTSC concerns the concurrent presence of a histologic pattern that is more consonant with that of conventional
Figure 6 Microscopic photograph of primary small cell neuroendocrine carcinoma of the thymus. The neoplastic cells show dispersed nuclear chromatin, apoptosis, and “molding” of nuclear membranes on one another.
THYMOMA AND THYMIC CARCINOMA
243
Figure 8 Histologic image of keratinizing squamous cell carcinoma of the thymus. Interconnecting cords and nests of large tumor cells are present, with overtly atypical nuclei and foci of keratin deposition (left center).
Figure 7 (a) Photomicrograph of basaloid thymic carcinoma, exhibiting an association with a pre-existing multilocular thymic cyst (right of figure). (b) The tumor cells are arranged in lobules with prominent geographic necrosis.
thymoma. The two components may be seen in widely separated tissue blocks, or in admixture. The authors have even seen rare cases in which a gradual transition between them was evident. These observations imply that some cases of KTSC may develop through the “clonal evolution” (“dedifferentiation”) of thymomas. Nonkeratinizing Squamous Cell Carcinoma The nonkeratinizing variant of squamous thymic carcinoma differs only in its lesser level of differentiation.21,41,63 It is a tumor showing angular nests of polyhedral cells in a desmoplastic stroma. The distinct fibrous stromal septa of thymoma, however, are absent. The authors also include thymic neoplasms labeled by others as “large cell carcinomas”41 in the category of nonkeratinizing squamous carcinoma. Lymphoepithelioma-like Squamous Carcinoma “Lymphoepitheliomas” of the nasopharynx and other anatomic sites are now known to represent distinctive forms of squamous cell carcinoma, some of which are associated with infection by the EBV.67 Similar concepts apply to thymic neoplasia.68 Lymphoepithelioma-like thymic
Figure 9 Photomicrograph of LETC, demonstrating syncytia of polygonal tumor cells with vesicular nuclear chromatin and prominent nucleoli. Lymphocytes are intimately interspersed throughout the lesion.
carcinoma (LETC) has a peculiar histologic appearance unlike that of KTSC or nonkeratinizing squamous carcinoma. LETCs contain syncytia of polygonal cells with ill-defined boundaries, vesicular nuclei, eosinophilic nucleoli, and an admixture of mature lymphocytes (see Figure 9).41,68 Tumoral stroma is usually represented by narrow fibrovascular septa. Mucoepidermoid Carcinomas Rare examples of primary thymic carcinoma have been reported that imitated MEC of the salivary glands62 or adenosquamous carcinoma of the lung41,65,69 in that they exhibited a partially squamous constituency. Obviously, concerns regarding the origin of such lesions within the thymus must be addressed once more. In MEC, foci that resemble well-differentiated KTSC are admixed with goblet cell-type epithelium that surrounds mucinous material. On the other hand, adenosquamous carcinoma is a high-grade tumor that resembles nonkeratinizing squamous cancer histologically. The salient difference
244
THORACIC TUMORS
Figure 10 Microscopic photograph of clear-cell carcinoma of the thymus, showing pleomorphic, hyperchromatic nuclei, and lucent cytoplasm.
between those two lesions is the presence of glandular lumina in the adenosquamous carcinoma; these may contain material that is positive with the mucicarmine or periodic acid-Schiff (PAS)-diastase stains. Clear cell Carcinoma Few examples of thymic clear cell carcinoma (TCCC) have been documented.23,70 This lesion is uniformly composed of polyhedral cells with vesicular nuclei, distinct nucleoli, and clear cytoplasm (see Figure 10). In some cases, cellular lucency reflected the presence of abundant cytoplasmic glycogen; in others, hydropic change appeared to account for this finding. TCCC has a vaguely organoid growth pattern, inconspicuous vascularity, and an absence of “blood lakes” as expected in clear cell carcinomas in extrathymic sites such as the kidneys.70 Undifferentiated Thymic Carcinoma Two “pure” adenocarcinomas of the thymus have been documented by Shimosato and Mukai.41 One showed transition with an epithelial-predominant thymoma, as well as micropapillary growth and focal psammomatous calcification. The other lesion was also somewhat variable histologically, with foci of both micropapillary and solid growth. Sarcomatoid Carcinoma Only rare examples of primary sarcomatoid thymic carcinoma (STC) have been reported.54,66,71 At the risk of redundancy, it must again be said that metastases must be excluded rigorously before assigning a final diagnosis of primary STC. Microscopically, this neoplasm manifests irregular fascicles of pleomorphic spindle cells. Nuclei are hyperchromatic, nucleoli are obvious, and mitoses are numerous. Occasional examples have contained cohesive epithelioid cell nests amongst the spindle cells,54 and biphasic STCs with carcinoidal elements (“dedifferentiated” thymic carcinoids) have also been documented.72,73 Specialized Pathologic Characteristics of Thymic Carcinomas The major goals of ancillary pathologic studies in cases of suspected thymic carcinoma are to provide information that would exclude a nonepithelial lesion, and also to
secure – if possible – data that would tend to support a primary intrathymic nature for the mass in question. The cells of all thymic carcinomas are reactive for keratin and EMA; however, in sarcomatoid lesions these markers may be seen only focally.74 This fact leaves open the possibility that a small biopsy specimen might fail to demonstrate the true nature of the mass because of sampling artifact. The latter should be considered before accepting a final diagnosis of “mediastinal sarcoma.” Primary small cell neuroendocrine carcinoma of the thymus is distinguished from other forms of thymic carcinoma either by a distinctive pattern of intermediate filament expression, with perinuclear “globules” of keratin protein42 or by its potential reactivity for one of several neuroendocrine markers. These include chromogranin A (a matrix protein of neurosecretory granules), synaptophysin, Leu-7 (CD57 antigen), and selected neuropeptides such as adrenocorticotrophic hormone. There are, at present, no reliable markers that could be used to separate primary from secondary thymic small cell carcinomas. In that context, it is notable that several studies have shown the expression of CD5 by epithelial cells of nonsmall cell thymic carcinomas,75 – 77 but not thymomas. Other primary thymic malignancies, such as germ cell tumors and lymphomas, are CD5-negative, as are metastatic mediastinal carcinomas from other organs. This last fact obviously may be useful in excluding secondary tumors that might imitate thymic carcinomas histologically. At least some LETC cases show cellular EBV integration at a molecular level.27,68 Using in situ hybridization and riboprobes to Epstein-Barr virus early ribonucleic acid-1 (EBER-1), Wu and Kuo25 observed positivity in only one of five LETCs and none of the other 15 thymic carcinomas for EBER-1 transcripts, markedly limiting the differential diagnostic utility of this technique. Evaluations for mutant p53 protein, as undertaken by Tateyama et al., also show a broad range of expression by thymomas and PTC.78 Nonetheless, common abnormalities in the p53 gene in many other tumors preclude a meaningful role for that moiety in differential diagnosis. One particular challenge for the pathologist is the distinction between primary mediastinal synovial sarcoma and STC; most specialized features of both those lesions are largely similar. The demonstration of t(X;18) chromosomal translocations in synovial sarcoma by fluorescence in situ hybridization has now made it recognizable with diagnostic certainty.79 Molecular Markers in Thymoma and Thymic Carcinoma
Flow cytometric analysis has been done of the nuclear DNA in epithelial cells of thymomas. In one study of 25 patients, diploidy was associated with lower stage and better fiveyear survival although this was not true concerning the percentage of cells in S-phase.80 Another evaluation showed that mean nuclear DNA exhibited a continuum of low values in noninvasive thymomas but progressively higher DNA content was seen in invasive thymoma and thymic carcinomas.81 This study was small in scope and showed an overlap between noninvasive and invasive thymomas.
THYMOMA AND THYMIC CARCINOMA
However, a significant difference was found between the DNA content of thymic carcinomas and thymomas. Several studies have assessed other factors such as the proliferative activity (proliferating cell nuclear antigen [PCNA], Ki-67, mitotic figures) and the activation of matrix metalloproteinase-2 (MMP-2).82 Immunoreactivity for MMP-2 is low in WHO type AB thymoma; moderate in type B1; and high in types B2, BC, and C. These findings appear to correlate with the oncologic behavior of each thymoma type.83 Glycosylphosphatidyl inositol-anchored protein (GPI-80) gene expression is a secretory protein or cell surface protein that is also found in significantly higher concentrations in invasive thymoma (stage IV) versus earlier (stage I) thymoma.84 Interestingly, GPI-80 was not elevated in thymic carcinomas. The exact mechanism for the overexpression of this protein during thymoma progression is not known. The p53 protein has likewise been studied in thymic epithelial cells. There is some evidence that progression of thymic tumors may be associated with accumulation of mutant forms of this moiety.78,85 The protooncogene bcl-2 encodes a protein that inhibits apoptosis. It has been evaluated in 30 thymoma specimens.86 Medullary thymocytes exhibited positivity for this marker, as shown in normal thymic medulla as well. Such findings correspond to a relative lack of apoptosis in the medullary compartment and spindle cell thymomas, whereas cortical epithelial cells manifesting apoptosis did not stain. Similarly, “cortical” thymomas were bcl-2 -negative. These findings may support the concept that the neoplastic epithelial cells in thymomas are derived from different compartments and that the bcl-2 protein may play a role in “medullary” (spindle cell) differentiation. The MIC2 antibody 013 has also been investigated in this context.87 This marker of immature T cells, which is almost always seen in cortical thymocytes and only in 5% of medullary lymphocytes, may assist in the diagnosis of thymoma in the mediastinum, or ectopic and metastatic sites. Thymic carcinoma is 013-negative; however, it should be recognized with caution that the lymphoid cells of both lymphocyte-predominant thymomas and lymphoblastic lymphomas are labelled with this marker. Epidermal growth factor receptor (EGFR) is expressed in a high percentage of thymic epithelial tumors88 and may be associated with invasive and/or advanced-stage disease.89 There has been one case report of overexpression of mutated c-kit and a response to imatinib.90 It appears that c-kit is present in thymic carcinomas, but not thymomas and EGFR is present in thymomas, but not thymic carcinomas.91
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS Clinical Presentation Thymomas most commonly affect people between the ages of 40 and 60 years (with a mean age of 50 years) but they also may occur in the pediatric and geriatric populations and there is no sex predominance.3,11,52,60,92 – 99 The same demographic features have been observed for thymic
245
carcinoma.20,63,66,100,101 The latter neoplasms account for about 5% of thymic epithelial tumors.63 Manifestations associated with thymic epithelial tumors may be absent, they may be attributed to the tumor itself as a space-occupying lesion, or they may relate to systemic syndromes associated with such lesions. It has been observed that 80% of patients with thymic lesions that are found incidentally will have benign lesions or invasive thymoma.102,103 In contrast, 57% of patients with a malignant neoplasm will have definite symptoms and signs. Moreover, approximately 30% of patients with thymoma will be found to have a mediastinal mass on a chest x-ray, noted incidentally.60,93,94 In contrast, in 10–30% of patients with thymic carcinomas the lesions are found incidentally.20,66 Complaints occur because of growth of tumors into the surrounding tissues. These include chest pain by invading the chest wall, dyspnea, hemoptysis, stridor, cough, dysphagia, fatigue, dysphonia, the superior vena cava syndrome, cardiac arrhythmias, and Horner’s syndrome. Rarely, thymic tumors may directly invade the spinal cord and cause neurologic dysfunction. Most thymomas (65% of cases) are locally invasive and rarely metastasize outside of the thorax.60,98 Noninvasive thymomas are characterized by an intact fibrous capsule, movability, and resectability; these clinical attributes are not associated with microscopic invasion through the capsule, although the lesion may be adherent to a neighboring structure. Invasive thymomas either grossly invade mediastinal structures, metastasize, or both. Metastases are most commonly seen on the pleural or pericardial surfaces as “drop” lesions. Metastatic involvement of lymph nodes is rare and when it occurs there is sequential involvement of the anterior mediastinal lymph nodes, then other mediastinal lymph nodes, and finally extrathoracic lymph nodes (supraclavicular, axillary, and para-aortic). This occurs in only 3–7% of patients with thymoma.104 Hematogenous spread is manifest with bony, hepatic, and pulmonary parenchymal lesions and occurs in approximately 5% of patients. Each case is obviously different, but patients with fully resected disease may have recurrent disease within 1–15 years of resection even if the lesion was initially stage I.60 Residual nonresected or metastatic disease has occasionally been observed to behave very indolently and patients may be alive with the disease for up to 10 years.98 In contrast, thymic carcinomas have a more aggressive clinical course. Thymic carcinomas also differ clinically from thymomas in that they are not associated with paraneoplasias such as MG or pure red cell aplasia (PRCA).55 Patients with these malignant lesions usually present with complaints referring to structural displacement, or they are detected incidentally by chest radiography. Patients with thymic carcinomas characteristically are middle-aged or elderly adults; infrequent examples have affected children also. Because some lesions may resemble metastatic tumors from other organs, one must exclude an occult malignancy elsewhere before a final diagnosis is rendered. In contrast to thymoma, thymic carcinoma presents with stage I or II disease in less than 50% of cases; stage III
246
THORACIC TUMORS
in roughly 30% and stage IV – usually with metastasis – in 20%.20,66,100 In spite of a generally poor prognosis, some patients have had prolonged survivals, such as one individual who lived for 10 years with metastatic disease that partially responded to chemotherapy.101 Metastatic lesions are grouped in the system of Masaoka et al.3 into those with pericardial or pleural dissemination (stage IVa) or lymphogenous or hematogenous metastasis (IVb). As expected, high-stage disease is seen more frequently with thymic carcinomas.104 Intrathoracic metastases are characterized by symptoms and signs of malignant effusions. These include chest pain, dyspnea, and in the case of pericardial involvement cardiac tamponade. Common sites of lymphogenous spread include mediastinal, cervical, and axillary lymph nodes. Hematogenous metastases involve bones (especially the spine), liver, lung, brain (rarely), bone marrow, and kidney.20,66 On account of the role of the thymus in the immunologic and hematologic systems, about 50–70% of thymomas are associated with one or more of a variety of systemic diseases (see Table 3).3,60,105,106 The most common and well recognized of these disorders is MG as seen in 30–50% of patients with thymomas. In contrast, only 10–12% of patients with MG have thymoma. The interplay between the two conditions is best illustrated by the fact that thymectomy can lessen MG manifestations in 50–60% of cases and can result in complete remissions (CRs) 8% of the time.107 It has also been noted that the recurrence of MG can herald the progression, either locally or systemically, of a previously controlled tumor. It must be kept in mind that myasthenic crises are associated with significant medical stresses, such as myocardial infarction or sepsis, and may thus occur even though there may be no evidence of active thymomatous growth. Table 3 Disorders associated with thymoma.
Neuromuscular syndromes Myasthenia gravis Myotonic dystrophy Eaton-Lambert syndrome Limbic encephalopathy Sensorimotor radiculopathy Stiff person syndrome Immune deficiency syndromes Hypogammaglobulinemia T-cell deficiency syndrome Collagen diseases and autoimmune disorders Systemic lupus erythematosus Scleroderma Sarcoidosis Rheumatoid arthritis Polymyositis Cardiac disorders Myocarditis Acute pericarditis Gastrointestinal disorders Ulcerative colitis Malignancy Lymphoma, carcinoma, Kaposi’s sarcoma
Hematologic syndromes Red cell aplasia/hypoplasia Pernicious anemia Agranulocytosis Erythrocytosis Multiple myeloma Hemolytic anemia Acute leukemia T-cell lymphocytosis Dermatologic diseases Pemphigus Alopecia areata Chronic mucocutaneous candidiasis Endocrine disorders Cushing’s syndrome Panhypopituitarism Addison’s disease Thyroiditis Hypertrophic osteoarthropathy Renal disease Nephrotic syndrome Minimal change nephropathy
It is unclear if any paraneoplastic association may occur in patients with thymic carcinomas, but most series have reported no association. In one study, two patients were found to have had MG.20 In both cases, however, there was a history of biopsy proven thymoma. One other study has noted a 77% association of MG with “well-differentiated thymic carcinoma” a putative subtype of thymic carcinoma.19 Nevertheless, as discussed below, we subscribe to the view that the latter lesion is a thymoma variant, and in the WHO classification, this variant is now type B3 thymoma. It is significant for the pathogenesis of paraneoplastic MG that there is no age nor gender predilection and almost no HLA association in paraneoplastic MG as opposed to the more frequent MG type that occurs in thymitis.108 Theories attempting to explain the relationship between MG and thymoma include one or more of following: the autosensitization of T lymphocytes to acetylcholine receptors; thymic production of antibodies to acetylcholine receptors; T-cell imbalances caused by thymic production of abnormal proportions of thymocytes; abnormal MHC II expression in the thymic stroma with abnormal thymocyte production; or thymic synthesis of hormones such as thymopoietin, which have been shown to bind to acetylcholine receptors and are found at their highest levels when thymoma or thymic hyperplasia is present.4,109 The most common hematologic condition associated with thymoma is that of PRCA. It is found in 5–10% of patients with such neoplasms. Conversely, 30–50% of patients with PRCA are found to have thymoma and approximately onethird will have improvement in their bone marrow function after thymectomy.110,111 Examination of the bone marrow generally shows a decrease in all three hematopoietic cell lineages, with greatest involvement of the erythroid precursors. The cause of PRCA is thought to be autoimmune. A review of the literature detailed 10 instances of thymoma-associated agranulocytosis since 1967.112 Three of these cases showed a complete absence of myelopoiesis and the remainder demonstrated promyelocyte arrest. Thymectomy was performed in five patients and was thought to be successful in two. Seven patients had associated hypogammaglobulinemia and two had MG. Four of six cases in which the serum was studied showed in vitro inhibition of myeloid colony growth. From this information, it can be speculated that thymomarelated agranulocytosis is also mediated by an autoimmune mechanism. Some authors have shown an association of thymoma and secondary tumors (usually malignancies) occurring in about 20–28% of cases.105,113 However, a more recent study reports subsequent malignancy in only 9% of thymoma patients.1 The single most important cancer association with thymoma was with non-Hodgkin’s lymphoma. The exact cause of this observation is not known but is thought to be due to abnormal regulation of B lymphocyte proliferation by dysfunctional T lymphocytes. Another recognized disease association with thymoma is acquired hypogammaglobulinemia. It arises in 5–10% of patients with thymoma. In counterpoint, 10% of patients with hypogammaglobulinemia develop thymomas.60,105
THYMOMA AND THYMIC CARCINOMA
Diagnostic Considerations Imaging
After a general physical examination (with particular attention to the features listed above including careful examination of the thyroid), plain posteroanterior and lateral chest radiographs should be obtained. These can identify a soft tissue density mass and the lateral chest film may clarify its location within the anterior mediastinum. Computed tomography (CT) using contrast should be performed to further define the nature of the lesion (cystic versus solid, the presence of calcium, its origin, and relationship to the surrounding anatomy).114 Although, magnetic resonance imaging (MRI) is more cumbersome and expensive, it is sometimes helpful preoperatively to evaluate neurovascular structures and the presence of invasion. MRI has been useful in distinguishing fibrosis from recurrent tumor and as an adjunct in resolving other confusing CT findings. It is also preferable to obtain an MRI if iodine contrast is contraindicated. Echocardiography (particularly the transesophageal technique) may detect possible cardiac involvement of the tumor.115 More recently, fluorodeoxyglucose (FDG) positron emission tomography (PET) scans have been utilized and one study showed a high FDG uptake in thymic carcinomas (mean standardized uptake value [SUV] 7.2) with only a moderate uptake in noninvasive and invasive thymomas (mean SUV 3.0 and 3.8, respectively).116 Obtaining Tissue For Histologic Evaluation
Approximately two-thirds of patients with thymic epithelial neoplasms will present with a thymic mass, which on radiographic evaluation appears to be amenable to complete resection.60,98 Some authors express concern that a biopsy will cause tumor seeding99 and disruption of the capsule will result in a decreased chance of cure.117 In the authors’ experience, this has never been observed. In circumstances where the tumor cannot be removed en bloc and the diagnosis is not readily apparent, a biopsy is required to direct therapeutic decisions. Owing to the many types of masses found in the mediastinum, adequate tissue must be obtained to enable the pathologist to make a proper assessment. Cytologic examination of CT guided fine needle aspiration biopsies can identify obvious malignancies. Fine needle aspiration has a reported sensitivity and specificity of over 90% in the evaluation of mediastinal masses.118 Core-cutting needle biopsies are required, however, to further refine the diagnosis of some malignancies, especially lymphoma and thymoma. Immunohistochemical, electron microscopic, and cytogenetic analysis can be employed with the latter procedure to determine the cell of origin. The major risks of these biopsy methods include pneumothorax, hemoptysis, and vascular injury. If the location of the lesion does not permit this approach or adequate tissue cannot be obtained, mediastinoscopy, anterior mediastinotomy, or video-assisted thoracoscopy can be employed.117 Thoracotomy is occasionally required to procure suitable biopsy material. It has been proposed that gallium scanning can direct the evaluation, as thymomas are frequently gallium-avid. For example, if a gallium-avid lesion is found it is thought a mediastinoscopy
247
is the next appropriate step for obtaining tissue in conditions that would not require surgery, such as lymphoma.117 If thymoma is found, surgery will be required. Staging
Several staging systems (see Table 1) have been described, but only three are in common use. Each has its own advantages and disadvantages. Such schemes were created primarily for thymomas but they are used for thymic carcinomas as well. The paradigm devised by Bergh et al.119 is simple, but it fails to account for the tropism that thymomas have for the pleura and pericardium and its ability for extrathoracic dissemination. The staging system of Masaoka and colleagues3 does encompass these patterns of spread as seen with both thymoma and thymic carcinoma. For the most part, it serves as a rational template on which to base therapy. One pitfall is that all lymphatic metastases are grouped as IVb, regardless of whether they are intra- or extrathoracic. The intrathoracic lymph node metastases could generally be treated within a radiation port, but these are still grouped with extrathoracic lymph node metastases that cannot. Yamakawa along with Masaoka and others proposed a tumor-node-metastasis (TNM) classification which takes these issues into consideration.104 Involvement of intrathoracic lymph nodes constitutes N1 or N2 disease, but extrathoracic lymph node metastases are classified as N3. The authors contended that the metastatic sequence affected anterior mediastinal lymph nodes before other intrathoracic nodes and finally extrathoracic nodes. Their study of 207 patients was of insufficient size to determine whether the prognosis of patients with stage IVb, T1–2, N1, M0 disease is the same as that of others with stage IVb, Tx N3, and/or M1 lesions. For now, the Masaoka system is most commonly used because it generally predicts outcome and guides therapy. Moreover, thymoma is only rarely metastatic. However, if combined modality therapy is eventually shown to have a role, the distinction of intrathoracic and extrathoracic disease may be more relevant because of radiotherpeutic implications. The TNM system may be more applicable to thymic carcinomas because of their greater propensity for metastasis. Another area of uncertainty is the definition of “invasion”. Masaoka et al. have described this phenomenon as infiltration of the tumor into its capsule.3 Other investigators observed that recurrence was likely only if a thymoma breached its capsule and involved the mediastinal fat.120 This distinction is clinically relevant with regard to staging and adjuvant therapy and highlights the fact that staging is based on pathology as well as the surgical assessment. Haniuda et al.121 have suggested that further modification of the Masaoka system might be useful for stage II cases; these authors proposed appending a “p” designator to describe the precise status of the mediastinal pleura in the specified context. In such a construction, stage II-p0 tumors show no adhesion to the pleura; p1 tumors demonstrate fibrous adhesions between the tumor and pleura without true invasion of the latter structure; and p2 tumors manifest actual pleural infiltration. In their experience, an adverse breakpoint in behavior was seen at the stage II-p1 level (regardless
248
THORACIC TUMORS
of whether the lesions were stage IIA or IIB), and they therefore recommended adjuvant treatment with irradiation, or chemotherapy, or both at that point. These interventions appear to ameliorate the poorer prognosis that has attended thymomas, which could not be completely resected surgically. The greatest vindication of the Masaoka system is its correlation with therapy and outcome. Specifically, tumors with a low stage that are completely removed have a better prognosis than invasive lesions. As detailed above, histologic grade pertains only to carcinomas. The incidence of recurrence that is associated with each stage is considered in many clinicopathological reviews on thymoma. Caution must be exercised when interpreting such data, however, because the studies are heterogenenous with respect to whether staging was surgical or pathological in nature, and also regarding what adjuvant therapy was given. The other staging system that is worth mentioning is the GETT (Groupe d’Etudes des Tumeurs Thymiques) classification (see Table 2).122 It is based on the Masaoka system, as well as selected surgical findings and the extent of surgery that was performed. Some authors have also divided thymomas into “limited” and “extensive,” based on whether all disease could be encompassed in a single radiotherapy portal. This may become more common parlance if combined modality therapy (radiation and chemotherapy) is proven to have a role.
TREATMENT In this section the therapy commonly employed for each stage of thymoma and the evidence that supports the approach will be presented. It must be kept in mind that thymoma and thymic carcinoma are rare disorders and there is little data from randomized controlled trials. Much of the data is conflicting and is from retrospective reviews with different approaches regarding histologic classification and treatments. Thymic carcinoma is rarer again with just over 300 cases reported in the literature.123
Thymoma Noninvasive Disease (Stage I)
There is little doubt that surgery alone is an effective way of curing a patient with a completely encapsulated tumor without evidence of microscopic invasion. Multiple outcome studies have shown that for stage I thymoma, nearly 90% of patients will be alive at 5 years and approximately 80% at 10 years after surgery, and that there will occasionally be local recurrences.124 – 126 Therefore lifelong surveillance is required, even for stage I. Radiation therapy has not been shown to increase survival as adjuvant therapy. One study of 132 stage I patients had a recurrence rate of 1.5%,124 and smaller studies concurred that radiotherapy following complete resection of a stage I thymoma unlikely adds to overall survival.125,127,128 Therefore, important questions best answered by the experienced pathologist should be addressed: (i) Is invasion defined as “into” or “through” the capsule? (ii) Was the whole capsule
evaluated for integrity? and (iii) Was a portion of the capsule breached during handling?127 A median sternotomy incision is most commonly used and there are reports in the literature of video-assisted thorascopy being employed as a diagnostic and staging instrument and as a therapeutic tool if the disease is found to be well encapsulated.129 The latter at this stage must be considered experimental until long-term results have been published. When the thymoma is laterally displaced a thoracotomy may be required. It must also be emphasized that the whole thymus is removed at the time of surgery. If the patient is deemed unfit for surgery, radiation therapy alone is an alternative as it has been shown to bring about local control in more advanced stages.130,131 Locally Invasive Disease (Stages II and III) Complete Resection Surgery again plays a key role in locally invasive disease. It has been shown that if the whole tumor can be resected the survival is improved, even if invasion is found to be present.122,125,130,132 Complete resection may require frozen section control to ensure that margins are clear and at times may require major resection and vascular reconstruction of mediastinal structures. These include lung, pleura, pericardium, and great vessels. For patients with locally advanced disease at the time of preoperative evaluation, the effectiveness of combination chemotherapy (see below) encourages the use of preoperative chemotherapy to downsize the tumor. In one review of 8 series with a combined total of 115 completely resected stage II patients treated with postoperative radiation, the recurrence rate was decreased from 30 to 5%.133 Adjuvant radiotherapy for completely resected stage II tumors of the higher risk WHO subtypes B2, B3 and C is advocated by many authors.126,134 However, some authors have disputed the role of adjuvant radiotherapy.126 More recently, Strobel et al. showed that thymomas of the WHO subtypes A, AB, and B1 in Masaoka stage I and II with R0 tumor resection may not require adjuvant therapy.135 A large series of more than 1000 patients suggested that adjuvant radiation and/or chemotherapy was not helpful in patients with totally resected thymoma or thymic carcinoma.136 Incomplete Resection If residual disease remains after surgery, radiotherapy has been employed to maximize the chance of local control. In one report, CR with a mean dose of 50 Gy was obtained in approximately 80% of cases, and 80% of these patients have been shown to be disease-free at about 9 years.122 A crude analysis of several retrospective reviews comprising 225 patients with incomplete resection or unresectable tumors revealed a more pessimistic picture.131 Thirty-five percent of cases developed local relapse. There are two controversial issues in the scenario of incomplete resection. The first, does the extent of surgery impact on survival? It has been established, as detailed above, that complete resection if at all possible does improve survival. However, retrospective series of patients with invasive disease report that radiotherapy after biopsy results in similar 5-year survival rates as debulking surgery (with residual disease) followed by radiation therapy.133,137 Other studies have
THYMOMA AND THYMIC CARCINOMA
shown that partial resection and radiation portends a better prognosis than biopsy and radiation therapy.122,124,125,130 The second area of controversy concerns the optimal dose of radiation that is required. Reported radiotherapy doses range from 30 to 60 Gy in 1.8 or 2.0 cGy fractions over 3 to 6 weeks. One retrospective study138 of patients with invasive thymoma correlated doses of radiotherapy after surgery and outcome. Local recurrences at 2 years appeared similar regardless of dose range: 42% (5/12) for those given <48 Gy, 35% (6/17) between 49 and 59 Gy, and 0/3 locally relapsed if given more than 60 Gy. Notably, 21 patients (37.5%) recurred outside of the radiation port. For macroscopic residual thymoma it has been preferred that greater than 50 Gy be given, as less than this dose resulted in a recurrence rate of 34% in 90 patients with residual disease after surgery.122 Although some investigators continue to do so, there is no data for or against treating a clinically negative supraclavicular fossa with radiation therapy. The numbers from these series are small and no firm conclusion can be made. However, if one extrapolates data from subclinical breast carcinoma in which a dose-response curve (with 30–35 Gy controlling 60–70% of cases, 40 Gy 80–90%, and 50 Gy controlling greater than 90%139 ) appears to exist, then it is logical to consider higher-dose therapy for thymoma. Dose fractions and treatment fields should be carefully planned to minimize complications such as fibrosis, pericarditis, and pneumonitis. It must be noted that the retrospective reviews have not demonstrated a consistent survival advantage for patients who received postoperative radiotherapy, despite improvements in local control.126 This may be because of a selection bias and has not been studied in a prospective manner. Furthermore, despite the obvious benefits of surgery and radiotherapy, the role of chemotherapy is yet to be answered. Data exists showing that even patients with completely resected stage II and III disease are at risk for relapse despite adjuvant radiotherapy.140 A retrospective analysis reported that stage III –IV patients treated with local therapy (surgery, radiotherapy) plus cisplatin-based combination chemotherapy were associated with a longer disease-free survival than patients who did not receive chemotherapy.130 Therefore, chemoradiotherapy after debulking surgery may have a role in this setting and a discussion regarding this is detailed below. Local Therapy for Recurrent Thymoma and/or Metastatic Thymoma
After primary local therapy, thymoma may recur with thoracic (pulmonary or pleural metastases) and/or extrathoracic disease. Given the indolent nature of the disease, with relatively long progression-free survival, surgical excision of recurrent disease is a reasonable approach in selected patients. Following surgery alone for recurrent disease, there are reports of patients free of disease up to 13 years.126,141 There is some evidence that surgery provides a better survival than radiation in this setting.126 If surgery is not possible
249
and there has been no prior radiotherapy or the tissue tolerance allows, radiotherapy can be instituted for intrathoracic recurrences with one study showing 6 of 10 patients alive at 7 years.142 In multifocal or extensive local recurrences, chemotherapy alone or followed by surgery or radiation therapy to residual disease is a reasonable therapeutic approach. This approach allows an in vivo evaluation of the sensitivity of the tumor to chemotherapy and may facilitate surgery. It should be noted that distant recurrences (intrathoracic pleural and extrathoracic) carry a worse prognosis than local recurrence.126
Thymic Carcinoma No significant data has been collected regarding the treatment of thymic carcinoma. The series that have been published have employed the same approach used for the treatment of thymoma. Most patients present with locally advanced and/or metastatic disease. There is no reason to think that the principle of a multimodality approach should not apply to thymic carcinoma.143,144 Studies that have looked at this issue found a trend towards increased survival in patients given neoadjuvant chemotherapy, adjuvant radiotherapy, or both for locally advanced disease.100,145 Given its propensity for metastasis, an argument can be made for adjuvant chemotherapy. Metastatic disease has been reported to respond to combination chemotherapy. In one series three out of five patients had responses to bleomycin, etoposide, and cisplatin with a median duration of remission of approximately 12 months.146 Another small series found that two out of four patients responded to cyclophosphamide, doxorubicin, and vincristine.63 A third series reported clinical response to cisplatin, doxorubicin, vincristine and cyclophosphamide (ADOC) with six partial remissions (PRs) in eight treated patients with a median survival time of 19 months.147 In a prospective intergroup trial, three patients with thymic carcinoma were included148 which evaluated cisplatin, doxorubicin and cyclophosphamide (PAC) for thymoma. One out of the 3 patients responded. The only retrospective review which investigated whether chemotherapy was beneficial found no survival advantage.20 However, the conclusions from this study are limited by the fact that the numbers were small and it was arbitrary as to which patients received chemotherapy.
Chemotherapy for the Treatment of Thymoma and Thymic Carcinoma Retrospective reviews show that 20–30% of patients may be candidates for systemic therapy. Only in the last 10 to 20 years have prospective trials been initiated to evaluate the role of chemotherapy in thymic malignancies. Any comments, however, must be prefaced by the knowledge that the optimal regimen and timing of chemotherapy has not yet been determined and enrollment in a clinical trial is still the favored approach. The data that is available can be divided into single-agent chemotherapy, combination chemotherapy, chemoradiotherapy, and corticosteroid and cytokine therapy.
250
THORACIC TUMORS
Single-Agent Chemotherapy:
The data from single-agent chemotherapy evaluations is limited to just a few small prospective trials, multiple case reports and reviews. These limitations are compounded by the fact that most patients had differing regimens and/or had prior therapy. The most active agents appear to be cisplatin,149 corticosteroids,150,151 doxorubicin152 and alkylating agents, but the activity of single agents remains limited. The enthusiasm for cisplatin was based on case reports from 1973, in which objective remissions were observed and one response was maintained for more than 12 months.149 This was followed up 20 years later with a study of cisplatin (50 mg m−2 ) given every 3 weeks in 20 patients with metastatic or recurrent thymoma.149 Fifteen had received prior radiotherapy and three had received prior chemotherapy. Only two (10%) patients had a partial response, eight (44%) had stable disease and the remainder progressed. Ifosfamide is another agent worth discussing in detail. In 1999, Highley et al.153 reported the results of 15 patients with invasive thymoma. Two different regimens of ifosfamide were given and 13 patients were assessable for response. Five patients had a complete response; one patient had a partial response. The median duration of complete response was 66+ months and the estimated survival rate at 5 years was 57%. Results of several agents from smaller series or case reports are worth mentioning. Suramin has been detailed in a case report to bring about a PR for 8 months in a patient who had been treated with two previous courses of combination chemotherapy.154 Doxorubicin as a single agent has been reported to result in two PRs in three patients evaluated150 and maytansine has resulted in five partial responses in seven patients.152,155,156 Other single agents that have been reported in series of two or more patients with no significant responses have included cyclophosphamide,152,157 dacarbazine,157 asparaginase,158 azacitidine,158 chlorambucil,152 chlormethine,152 and vincristine.159 Combination Chemotherapy
The optimal regimen has not been determined and the following is a review of the major studies. A summary of the other studies conducted can be found in Table 4. Cisplatin (50 mg m−2 ), doxorubicin (50 mg m−2 ) and cyclophosphamide (500 mg m−2 ), or “PAC” was evaluated as a combination in 30 patients with metastatic disease, recurrent disease, or disease that could not be encompassed by one radiotherapy portal.160 This group of patients received a mean of six cycles: three patients had complete responses, 12 had partial responses and 10 had stable disease. Median duration of response was 11.8 months and median survival was 37.7 months. Another Intergroup trial involved 23 patients with limited-stage unresectable thymoma (Masaoka stage III) who received two to four cycles of PAC followed by radiotherapy.148 In this study, there were five CRs and 11 PRs to chemotherapy (70% RR) with a median survival of 93 months. Five-year survival was 52.5%. Case reports have also described durable second remissions being obtained
with this combination despite having been treated with these agents in the past.161 In a neoadjuvant setting, investigators treated patients with unresectable thymoma (Masaoka stage III or IVA) with three cycles of induction chemotherapy with PAC plus prednisone, followed by surgery and then radiotherapy.174 These patients then went on to receive three more cycles of chemotherapy. There were 12 evaluable patients, and 11 patients underwent surgical resection. Among the 12 patients, response to induction chemotherapy revealed a 92% RR, with three CRs and eight PRs. Two of the three CRs had 100% tumor necrosis at the time of resection, confirming the efficacy of chemotherapy. Adriamycin (40 mg m−2 ), cisplatin (50 mg m−2 ) and cyclophosphamide (700 mg m−2 ) have been combined with vincristine (0.6 mg m−2 ) (ADOC) in a single institution study of 37 patients with stage III or IV disease.176 Responses were observed in 91.8% of patients with a 43% CR rate. The median duration of response and survival were 12 months and 15 months, respectively. The complete responses lasted for a median of 27 months and the partial responses lasted for 9.5 months. In this study etoposide, ifosfamide, and cisplatin used in combination as second line therapy resulted in stable disease for 2 months and 6 months, in two out of three patients who relapsed. The European Organization for Research and Treatment of Cancer (EORTC) evaluated the regimen of cisplatin (60 mg m−2 ) and etoposide (120 mg m−2 ), i.e. “PE.”180 Sixteen patients with metastatic or recurrent disease received a mean of six cycles and five CRs were noted with four PRs (RR of 56%). A median response duration of 3.4 years was seen with a median survival of 4.3 years. Results of an Eastern Cooperative Group-coordinated intergroup trial evaluating etoposide, ifosfamide and cisplatin (VIP), suggests that ifosfamide does not add to the combination.175 This study produced six PRs out of 14 patients (43%). Chemoradiotherapy
The concept of combined modality therapy is attractive for locally advanced thymoma and thymic carcinoma. Although several investigators have reported the feasibility and outcome of patients treated with chemoradiotherapy there are no prospective randomized trials defining the optimal management of patients with locally advanced disease. Patients in whom a complete resection is accomplished appear to have a prolonged survival over patients with incomplete resection. Chemotherapy used in a neoadjuvant setting may enable a complete resection where it was not previously feasible or facilitate a surgery which is less morbid.181 The approach of preoperative radiotherapy was first reported by Macchiarini and colleagues. Three cycles of cisplatin, epirubicin, and etoposide were given in a neoadjuvant setting to seven patients with clinical stage III disease.181 All patients had a PR following chemotherapy and underwent surgery, with four patients undergoing complete resection (two patients had microscopic disease and one patient had gross residual disease). Similar results were then reported in another study of 16 patients with stage III and IVA, treated with cisplatin, doxorubicin, vincristine, and
THYMOMA AND THYMIC CARCINOMA
251
Table 4 Results of chemotherapy in patients with thymoma.
Number of patients
Responsea
Reference
2 1 4 2 13 21 5 2 3
0 0 0 0 3CR, 8PR 2PR 3CR, 2PR 0 0
Maytansine
2 2 13 1 14 7
0 0 7CR, 1PR 1CR 0 5PR
Suramin Vincristine
1 2
1PR 0
Chahinian et al.158 Chahinian et al.158 Boston152 Boston152 Hu and Levine150 Bonomi et al.149 Harper and Addis188 Donovan et al.157 Boston152 Donovan et al.157 Donovan et al.157 Harper and Addis188 Harper et al.189 Berthaud et al.162 Gordon et al.163 Boston152 Jaffrey156 Chahinian et al.158 LaRocca et al.154 Stolinsky et al.159
2 1 2 5 13 1 5 9
1PR 1CR 2CR 2CR, 3PR 5CR 0 0CR, 4PR 4CR, 1PR
Stolinsky et al.159 Butler et al.164 Loehrer et al.165 Kosmidis et al.166 Goldel et al.167 Hu and Levine150 Evans et al.168 Daugaard et al.169
3 9 30 1 23 1 12 28 32 16 9 1 1 16 7 4
3CR 1CR, 5PR 3CR, 10PR 1CR 5CR, 11PR 1 CR 3CR 8PR 0CR, 9PR 15CR, 14PR 7CR, 5PR 1CR, 5PR 1PR 0 5CR, 1PR 4CR, 3PR 2CR, 1PR
Klipstein et al.170 Mitrou et al.171 Loehrer et al.160 Fornasiero et al.172 Loehrer et al.148 Campbell et al.173 Shin et al.174 Loehrer et al.175 Fornasiero et al.176 Rea et al.177 Chahinian et al.178,179 Harper and Addis188 Hu and Levine150 Giaccone et al.180 Macchiarini et al.181 Dy et al.182
Drug Single agent Aspariginase Azactidine Chlorambucil Chlormethine Corticosteroids Cisplatin
Cyclophosphamide Dacarbazine Doxorubicin Ifosfamide Interleukin-2
Combination chemotherapy Non-cisplatin-containing regimens Chlormethine, vincristine, procarbazine, vinblastine Cyclophosphamide, doxorubicin Cyclophosphamide, doxorubicin, vincristine Cyclophosphamide, doxorubicin, vincristine, prednisone (CHOP) CHOP and cisplatin, etoposide Cyclophosphamide, vincristine, procarbazine, prednisone Lomustine, cyclophosphamide, vincristine, prednisone Cisplatin-containing regimens Cisplatin, doxorubicin Cisplatin, doxorubicin, cyclophosphamide
Cisplatin, doxorubicin, cyclophosphamide ± prednisone Cisplatin, Etoposide, Ifosfamide Cisplatin, doxorubicin, vincristine, cyclophosphamide Cisplatin, doxorubicin, prednisone, bleomycin Cisplatin, etoposide
Cisplatin, epirubicin, etoposide Cisplatin, vinblastine, bleomycin a
CR, complete remission; PR, partial remission.
cyclophosphamide every 3 weeks for three or four cycles.177 Postoperative radiation was given only if there was residual disease. If only fibrosis was present, the surgery was followed by three additional cycles of chemotherapy. Chemotherapy resulted in seven CRs and five partial responses. All patients received postoperative radiotherapy and the authors reported a projected 2-year survival of 80%. A third study investigated three cycles of induction chemotherapy (PAC) combined with prednisone in 13 patients with stage III or IVa disease.174 Twelve patients were evaluable for response:
three patients had complete responses and eight patients had partial responses. Eleven patients had surgical resection with complete resection in nine patients. Postoperative radiotherapy was given and followed by three cycles of consolidation chemotherapy. With a median follow-up of 43 months all patients were alive, two with disease and there was no excess morbidity despite the combination of three modes of therapy. These studies suggest that neoadjuvant chemotherapy with postoperative radiotherapy is a feasible approach. Recently, data has been published from a prospective intergroup trial of combined modality therapy in 23 patients
252
THORACIC TUMORS
with unresectable locally advanced disease. Five complete and 11 partial responses were achieved with two to four cycles of PAC followed by 54 Gy of radiation therapy in 23 patients. The overall response rate was 69.6%.148 The progression-free and overall survival rates at 5 years were 54.3% and 52.5% respectively. It should be noted that only four patients underwent major debulking prior to initiating chemotherapy. There are limitations outlined in the paper, which prevent any definite conclusions. Specifically, the 95% confidence limits overly the results of studies of radiotherapy as monotherapy, there may be as effective, if not more effective, regimens that are less toxic (i.e., radiation without an anthracycline).
Other Agents Responses to corticosteroid therapy have been reported in the literature. These are case reports or small retrospective reviews of patients with recurrent or metastatic disease.150,183 This activity is thought to be due to a lympholytic effect against the lymphocyte component as the responses are usually brief in nature. There is no precedent to suggest that steroids would be active against the true malignant portion of thymomas, the epithelium. Unless the lymphocytes shown to be involved in the tumorigenesis or steroids are proven to be active against the epithelial malignancy, there is no primary role for corticosteroids at this time. However in end stage disease, corticosteroids may be used with palliative intent. Interleukin-2 has been observed to bring about a CR in one patient who had been treated with prior chemotherapy.162 A follow-up study of interleukin-2 in 14 relapsed or refractory patients did not reveal any response.163 Further investigation with this drug is probably not warranted.
One case report184 documented the ability of octreotide combined with prednisone to induce a CR in a patient with thymoma and PRCA who failed to respond to chemotherapy. This spurred a study of octreotide analogs with prednisone in advanced chemotherapy-refractory thymic tumors.185 This study of 16 patients showed an overall response rate of 37%, with one patient with a complete response and five patients with partial responses. Median survival was 15 months. This combination had few side effects and could be considered an effective palliative treatment in previously treated, refractory disease. Because malignant thymic tumors were shown to express EGFR, gefitinib (an EGFR tyrosine kinase inhibitor) was recently studied in patients with advanced thymic malignancies. This study included 26 previously treated patients and although the treatment was tolerable, only one patient had a PR (response duration of 5 months) and 14 patients had stable disease (SD). Five patient samples were analyzed for the EGFR mutation and the mutation did not seem to be present in this small group of patients.186
AUTHORS’ RECOMMENDATIONS Upon the diagnosis of an anterior mediastinal mass, it is our practice to obtain serum β-human chorionic gonadotropin, lactate dehydrogenase, and α-fetoprotein to first rule out a germ cell tumor (see Figure 11). When suspicions of a thymic malignancy are high, an initial assessment of the surgical respectability should be made. If the mass is resectable, surgery is recommended. Histologic evaluation by an experienced pathologist is necessary to obtain the correct
Algorithm Mediastinal mass on CXR History, physical and relevant laboratory evaluation → CT scan Is it resectable?
Yes
No
Final pathology
→ Obtain tissue diagnosis with a core needle biopsy or open biopsy
No invasion Lifelong follow-up
Invasion present
Postoperative radiotherapy (may consider for thymic carcinoma, even if no invasion)
Thymoma or thymic carcinoma in a radiation port Yes
Chemotherapy → then surgery (if sufficient cytoreduction) and/or radiation therapy
No Chemotherapy then consider resection of residual disease
Figure 11 A suggested algorithm to guide the evaluation and management of a patient with a mediastinal mass who is found to have a thymoma or thymic carcinoma.
THYMOMA AND THYMIC CARCINOMA
diagnosis, to determine clear margins, and the presence of local/capsular invasion. If invasion through the capsule is noted, postoperative radiation therapy should be considered, especially for those patients with thymoma and positive surgical margins. The role of postoperative radiation therapy for patients with stage II thymoma and clear surgical margins is controversial. Postoperative radiation therapy should also be considered in all cases of thymic carcinoma. Patients with unresectable thymic tumors should undergo core needle biopsy or open biopsy to confirm tissue diagnosis. For these tumors, neoadjuvant chemotherapy should be given, followed by consideration of surgical resection (or radiation for unresectable disease). Because of the lack of data from adequately sized prospective studies, strong consideration should be given to referral to an academic center for enrollment on a clinical trial.
PROGNOSIS The clinicopathological retrospective reviews are the best resource for determining the prognosis of various stages. It must be realized that some of the older data is flawed by the inclusion of patients with thymic carcinoma and that survival rates have been altered by improvements in perioperative care especially of patients with MG.11 Table 5 is from a series which covers a timespan of nearly four decades. The treatment for most cases has always been surgery, but refinements in radiation therapy and chemotherapy since the 1980s may have improved the outcome of patients with advanced disease that is not apparent from this data. Although there is some contention on whether to treat based on stage and histology type, the data from this is not strong and the one factor that has consistently been shown to be a prognostic factor useful for directing therapy is the stage of the tumor. Histology may have a limited role in prognosis. For example, stage I cortical thymoma is the only scenario for which one study has suggested the need for a change in therapy. It is argued that the prognosis is similar for patients with stage II and III mixed thymoma and thus patients may benefit from postoperative radiation therapy.52 Clinicopathologic observations other than invasion which appear to be associated with a poor prognosis for patients with thymoma include the presence of tumor related symptoms and signs at diagnosis, size greater than 15 cm, age under 30 years at diagnosis, and microscopic predominance of epithelial cells.60,130,187 Table 5 Survival for thymoma and thymic carcinoma.
Disease Thymoma Stage I Stage II Stage III Stage IV Thymic carcinomaa a
Median survival 20 months.
3-year survival (%)
5-year survival (%)
10-year survival (%)
– – – – – 45
– 85 70 70 25 – 50 35
– 75 65 50 10 – 30 –
253
Myasthenia Gravis In earlier reviews, a decreased overall survival was observed in patients with MG and thymoma.3,92 This was likely due to higher perioperative deaths and less effective medical management of these patients. Better control of MG and advances in perioperative management have made the prognosis of a patient with MG equivalent to other patients.60,96,99,124 However, there is also data suggesting that MG portends a better prognosis. This may be due to earlier detection due to the symptoms of MG;125 yet there is evidence that stage-for-stage patients with MG had higher rates of overall survival.132
REFERENCES 1. Engels EA, Pfeiffer RM. Malignant thymoma in the United States: demographic patterns in incidence and associations with subsequent malignancies. Int J Cancer 2003; 105: 546 – 51. 2. Levine GD, Rosai J. Thymic hyperplasia and neoplasia: a review of current concepts. Hum Pathol 1978; 9: 495 – 515. 3. Masaoka A, et al. Follow-up study of thymomas with special reference to their clinical stages. Cancer 1981; 48: 2485 – 92. 4. Muller-Hermelink HK, et al. Characterization of the human thymic microenvironment: lymphoepithelial interaction in normal thymus and thymoma. Arch Histol Cytol 1997; 60: 9 – 28. 5. Skandalakis JEGS, Todd NW. Embryology for Surgeons, 2nd ed. Baltimore, Maryland: Williams and Wilkins, 1994. 6. Marino M, Muller-Hermelink HK. Thymoma and thymic carcinoma. Relation of thymoma epithelial cells to the cortical and medullary differentiation of thymus. Virchows Arch A Pathol Anat Histopathol 1985; 407: 119 – 49. 7. van de Wijngaert FP, et al. Heterogeneity of epithelial cells in the human thymus. An ultrastructural study. Cell Tissue Res 1984; 237: 227 – 37. 8. Sprent J, Kishimoto H. T cell tolerance and the thymus. Ann N Y Acad Sci 1998; 841: 236 – 45. 9. Mizuno T, Hashimoto T, Masaoka A. Distribution of fibronectin and laminin in human thymoma. Cancer 1990; 65: 1367 – 74. 10. Lauriola L, et al. Human thymoma: immunologic characteristics of the lymphocytic component. Cancer 1981; 48: 1992 – 5. 11. Bernatz PE, et al. Thymoma: factors influencing prognosis. Surg Clin North Am 1973; 53: 885 – 92. 12. Rouse RV, Weiss LM. Human thymomas: evidence of immunohistologically defined normal and abnormal microenvironmental differentiation. Cell Immunol 1988; 111: 94 – 106. 13. Chilosi M, et al. Immunohistochemical evidence of active thymocyte proliferation in thymoma. Its possible role in the pathogenesis of autoimmune diseases. Am J Pathol 1987; 128: 464 – 70. 14. Zettl A, et al. Recurrent genetic aberrations in thymoma and thymic carcinoma. Am J Pathol 2000; 157: 257 – 66. 15. Zhou R, et al. Thymic epithelial tumors can develop along two different pathogenetic pathways. Am J Pathol 2001; 159: 1853 – 60. 16. Inoue M, et al. Chromosome 6 suffers frequent and multiple aberrations in thymoma. Am J Pathol 2002; 161: 1507 – 13. 17. Kuzume T, et al. Establishment and characterization of a thymic carcinoma cell line (Ty-82) carrying t(15;19)(q15;p13) chromosome abnormality. Int J Cancer 1992; 50: 259 – 64. 18. Lee AC, et al. Disseminated mediastinal carcinoma with chromosomal translocation (15;19). A distinctive clinicopathologic syndrome. Cancer 1993; 72: 2273 – 6. 19. Kirchner T, et al. Well-differentiated thymic carcinoma. An organotypical low-grade carcinoma with relationship to cortical thymoma. Am J Surg Pathol 1992; 16: 1153 – 69. 20. Suster S, Rosai J. Thymic carcinoma. A clinicopathologic study of 60 cases. Cancer 1991; 67: 1025 – 32. 21. Kuo TT, Chan JK. Thymic carcinoma arising in thymoma is associated with alterations in immunohistochemical profile. Am J Surg Pathol 1998; 22: 1474 – 81.
254
THORACIC TUMORS
22. Suster S, Moran CA. Primary thymic epithelial neoplasms: spectrum of differentiation and histological features. Semin Diagn Pathol 1999; 16: 2 – 17. 23. Hasserjian RP, Klimstra DS, Rosai J. Carcinoma of the thymus with clear-cell features. Report of eight cases and review of the literature. Am J Surg Pathol 1995; 19: 835 – 41. 24. Patton DF, et al. Thymic carcinoma with a defective Epstein-Barr virus encoding the BZLF1 trans-activator. J Infect Dis 1994; 170: 7 – 12. 25. Wu TC, Kuo TT. Study of Epstein-Barr virus early RNA 1 (EBER1) expression by in situ hybridization in thymic epithelial tumors of Chinese patients in Taiwan. Hum Pathol 1993; 24: 235 – 8. 26. Teoh R, et al. Increased incidence of thymoma in Chinese myasthenia gravis: possible relationship with Epstein-Barr virus. Acta Neurol Scand 1989; 80: 221 – 5. 27. Leyvraz S, et al. Association of Epstein-Barr virus with thymic carcinoma. N Engl J Med 1985; 312: 1296 – 9. 28. Imanishi TGT. Diversity in human MHC genes among ethnic groups worldwide. In Tajima KSS (ed) Ethnoepidemiology of Cancer. Tokyo, Japan: Scientific Societies Press, 1996: 89 – 96. 29. Clark S. The intrathymic environment. In Davies AJSCR (ed) Contemporary Topics in Immunobiology. New York: Plenum Press, 1973: 77 – 79. 30. Shier KJ. The morphology of the epithelial thymus. Observations on lymphocyte-depleted and fetal thymus. Lab Invest 1963; 12: 316 – 26. 31. Shier KJ. The thymus according to Schambacher: medullary ducts and reticular epithelium of thymus and thymomas. Cancer 1981; 48: 1183 – 99. 32. Wick M. Mediastinum. In SS Sternberg (ed) Diagnostic Surgical Pathology, 2nd ed. New York: Raven Press, 1994: 1125 – 1182. 33. Moran CA, et al. Thymomas presenting as pleural tumors. Report of eight cases. Am J Surg Pathol 1992; 16: 138 – 44. 34. Rosai J, Limas C, Husband EM. Ectopic hamartomatous thymoma. A distinctive benign lesion of lower neck. Am J Surg Pathol 1984; 8: 501 – 13. 35. Yeoh CB, et al. Intrapulmonary thymoma. J Thorac Cardiovasc Surg 1966; 51: 131 – 6. 36. Asa SL, et al. Primary thyroid thymoma: a distinct clinicopathologic entity. Hum Pathol 1988; 19: 1463 – 7. 37. Chan JK, Rosai J. Tumors of the neck showing thymic or related branchial pouch differentiation: a unifying concept. Hum Pathol 1991; 22: 349 – 67. 38. Suster S, Rosai J. Cystic thymomas. A clinicopathologic study of ten cases. Cancer 1992; 69: 92 – 7. 39. Muller-Hermelink HK, Marino M, Palestro G. Pathology of thymic epithelial tumors. Curr Top Pathol 1986; 75: 207 – 68. 40. Rosai J, Levine GD. Tumors of the thymus pathology. In Hartmann WH (ed): Atlas of Tumor Pathology, Fascicle 13, Series 2. Washington, District of Columbia: Armed Forces Institute of Pathology, 1976: 10 – 150. 41. Shimosato Y, Mukai K. Tumors of the thymus and mediastinum. In Rosai J (ed) Atlas of Tumor Pathology. Washington, District of Columbia: Armed Forces Institute of Pathology, 1997: 33 – 273. 42. Wick MRSR, Niehans GA, Scheithauer BW. Anterior mediastinal tumors: a clinicopathologic study of 100 cases, with emphasis on immunohistochemical analysis. Prog Surg Pathol 1990; 11: 79 – 119. 43. Engel P, Pilsgaard B, Francis D. Thymomas and thymic carcinomas. A retrospective investigation with histological reclassification. Apmis 1995; 103: 671 – 8. 44. Chalabreysse L, et al. Correlation of the WHO schema for the classification of thymic epithelial neoplasms with prognosis: a retrospective study of 90 tumors. Am J Surg Pathol 2002; 26: 1605 – 11. 45. Chen G, et al. New WHO histologic classification predicts prognosis of thymic epithelial tumors: a clinicopathologic study of 200 thymoma cases from China. Cancer 2002; 95: 420 – 9. 46. Nakagawa K, et al. Thymoma: a clinicopathologic study based on the new World Health Organization classification. J Thorac Cardiovasc Surg 2003; 126: 1134 – 40. 47. Okumura M, et al. The World Health Organization histologic classification system reflects the oncologic behavior of thymoma: a clinical study of 273 patients. Cancer 2002; 94: 624 – 32.
48. Rieker RJHJ, et al. Histologic classification of thymic epithelial tumors: comparison of established classification schemes. Int J Cancer 2002; 98: 900 – 6. 49. Bernatz PE, Harrison EG, Clagett OT. Thymoma: a clinicopathologic study. J West Soc Periodontol Periodontal Abstr 1961; 42: 424 – 44. 50. Muller-Hermelink HK, Marx A. Pathological aspects of malignant and benign thymic disorders. Ann Med 1999; 2(Suppl 31): 5 – 14. 51. Dawson A, Ibrahim NB, Gibbs AR. Observer variation in the histopathological classification of thymoma: correlation with prognosis. J Clin Pathol 1994; 47: 519 – 23. 52. Quintanilla-Martinez L, et al. Thymoma. Histologic subclassification is an independent prognostic factor. Cancer 1994; 74: 606 – 17. 53. Ho FCFK, et al. Evaluation of a histogenetic classification for thymic epithelial tumors. Histopathology 1982; 25: 21 – 9. 54. Snover DC, Levine GD, Rosai J. Thymic carcinoma. Five distinctive histological variants. Am J Surg Pathol 1982; 6: 451 – 70. 55. Walker AN, Mills SE, Fechner RE. Thymomas and thymic carcinomas. Semin Diagn Pathol 1990; 7: 250 – 65. 56. Morrison IM. Tumours and cysts of the mediastinum. Thorax 1958; 13: 294 – 307. 57. McCart JA, et al. Predictors of survival following surgical resection of thymoma. J Surg Oncol 1993; 54: 233 – 8. 58. Morgenthaler TI, et al. Thymoma. Mayo Clin Proc 1993; 68: 1110 – 23. 59. Wick MR. Assessing the prognosis of thymomas. Ann Thorac Surg 1990; 50: 521 – 2. 60. Lewis JE, et al. Thymoma. A clinicopathologic review. Cancer 1987; 60: 2727 – 43. 61. Iezzoni JC, Nass LB. Thymic basaloid carcinoma: a case report and review of the literature. Mod Pathol 1996; 9: 21 – 5. 62. Moran CA, Suster S. Mucoepidermoid carcinomas of the thymus. A clinicopathologic study of six cases. Am J Surg Pathol 1995; 19: 826 – 34. 63. Shimizu J, et al. Primary thymic carcinoma: a clinicopathological and immunohistochemical study. J Surg Oncol 1994; 56: 159 – 64. 64. Dadmanesh F, Sekihara T, Rosai J. Histologic typing of thymoma according to the new World Health Organization classification. Chest Surg Clin N Am 2001; 11: 407 – 20. 65. Kuo TT, et al. Thymic carcinomas: histopathological varieties and immunohistochemical study. Am J Surg Pathol 1990; 14: 24 – 34. 66. Wick MR, et al. Primary thymic carcinomas. Am J Surg Pathol 1982; 6: 613 – 30. 67. Weiss LM, Gaffey MJ, Shibata D. Lymphoepithelioma-like carcinoma and its relationship to Epstein-Barr virus. Am J Clin Pathol 1991; 96: 156 – 8. 68. Dimery IW, et al. Association of the Epstein-Barr virus with lymphoepithelioma of the thymus. Cancer 1988; 61: 2475 – 80. 69. Truong LD, et al. Thymic carcinoma. A clinicopathologic study of 13 cases. Am J Surg Pathol 1990; 14: 151 – 66. 70. Wick MR, et al. Clear cell neoplasms of the endocrine system and thymus. Semin Diagn Pathol 1997; 14: 183 – 202. 71. Nishimura M, et al. A case of sarcomatoid carcinoma of the thymus. Pathol Int 1997; 47: 260 – 3. 72. Kuo TT. Carcinoid tumor of the thymus with divergent sarcomatoid differentiation: report of a case with histogenetic consideration. Hum Pathol 1994; 25: 319 – 23. 73. Paties C, et al. Multidirectional carcinoma of the thymus with neuroendocrine and sarcomatoid components and carcinoid syndrome. Pathol Res Pract 1991; 187: 170 – 7. 74. Wick MR, Swanson PE. Carcinosarcomas: current perspectives and an historical review of nosological concepts. Semin Diagn Pathol 1993; 10: 118 – 27. 75. Berezowski K, et al. CD5 immunoreactivity of epithelial cells in thymic carcinoma and CASTLE using paraffin-embedded tissue. Am J Clin Pathol 1996; 106: 483 – 6. 76. Hishima T, et al. CD5 expression in thymic carcinoma. Am J Pathol 1994; 145: 268 – 75. 77. Kornstein MJ, Rosai J. CD5 labeling of thymic carcinomas and other nonlymphoid neoplasms. Am J Clin Pathol 1998; 109: 722 – 6. 78. Tateyama H, et al. p53 protein expression and p53 gene mutation in thymic epithelial tumors. An immunohistochemical and DNA sequencing study. Am J Clin Pathol 1995; 104: 375 – 81.
THYMOMA AND THYMIC CARCINOMA 79. DeLeeuw BSR, et al. Distinct Xp11.2 breakpoint regions in synovial sarcoma revealed by metaphase and interphase FISH: relationship to histologic subtypes. Cancer Genet Cytogenet 1994; 73: 89 – 94. 80. Pollack A, et al. Thymoma. The prognostic significance of flow cytometric DNA analysis. Cancer 1992; 69: 1702 – 9. 81. Asamura H, et al. Degree of malignancy of thymic epithelial tumors in terms of nuclear DNA content and nuclear area. An analysis of 39 cases. Am J Pathol 1988; 133: 615 – 22. 82. Kondo K, et al. Activation of matrix metalloproteinase-2 is correlated with invasiveness in thymic epithelial tumors. J Surg Oncol 2001; 76: 169 – 75. 83. Sogawa K, et al. Increased expression of matrix metalloproteinase2 and tissue inhibitor of metalloproteinase-2 is correlated with poor prognostic variables in patients with thymic epithelial tumors. Cancer 2003; 98: 1822 – 9. 84. Sasaki H, et al. Glycosylphosphatidyl inositol-anchored protein (GPI80) gene expression is correlated with human thymoma stage. Cancer Sci 2003; 94: 809 – 13. 85. Puglisi F. p53 protein expression and p53 mutation in thymic epithelial tumors. Am J Clin Pathol 1996; 105: 657 – 8. 86. Brocheriou I, Carnot F, Briere J. Immunohistochemical detection of bcl-2 protein in thymoma. Histopathology 1995; 27: 251 – 5. 87. Chan JK, et al. The MIC2 antibody 013. Practical application for the study of thymic epithelial tumors. Am J Surg Pathol 1995; 19: 1115 – 23. 88. Henley JD, Koukoulis GK, Loehrer PJ Sr. Epidermal growth factor receptor expression in invasive thymoma. J Cancer Res Clin Oncol 2002; 128: 167 – 70. 89. Ionescu DN, et al. Protein expression and gene amplification of epidermal growth factor receptor in thymomas. Cancer 2005; 103: 630 – 6. 90. Strobel P, et al. Thymic carcinoma with overexpression of mutated KIT and the response to imatinib. N Engl J Med 2004; 350: 2625 – 6. 91. Henley JD, Cummings OW, Loehrer PJ Sr. Tyrosine kinase receptor expression in thymomas. J Cancer Res Clin Oncol 2004; 130: 222 – 4. 92. Batata MA, et al. Thymomas: clinicopathologic features, therapy, and prognosis. Cancer 1974; 34: 389 – 96. 93. Legg MA, Brady WJ. Pathology and clinical behavior of thymomas. A survey of 51 cases. Cancer 1965; 18: 1131 – 44. 94. LeGolvan DP, Abell MR. Thymomas. Cancer 1977; 39: 2142 – 57. 95. Park HS, et al. Thymoma. A retrospective study of 87 cases. Cancer 1994; 73: 2491 – 8. 96. Pescarmona E, et al. Analysis of prognostic factors and clinicopathological staging of thymoma. Ann Thorac Surg 1990; 50: 534 – 8. 97. Salyer WR, Eggleston JC. Thymoma: a clinical and pathological study of 65 cases. Cancer 1976; 37: 229 – 49. 98. Verley JM, Hollmann KH. Thymoma. A comparative study of clinical stages, histologic features, and survival in 200 cases. Cancer 1985; 55: 1074 – 86. 99. Wilkins EW Jr, et al. J. Maxwell Chamberlain memorial paper. Role of staging in prognosis and management of thymoma. Ann Thorac Surg 51: 888 – 92 1991. 100. Hsu CP, et al. Thymic carcinoma. Ten years’ experience in twenty patients. J Thorac Cardiovasc Surg 1994; 107: 615 – 20. 101. Yano T, et al. Treatment and prognosis of primary thymic carcinoma. J Surg Oncol 1993; 52: 255 – 8. 102. Cohen AJ, et al. Primary cysts and tumors of the mediastinum. Ann Thorac Surg 1991; 51: 378 – 84; discussion 385 – 6. 103. Davis RD Jr, Oldham HN Jr, Sabiston DC Jr. Primary cysts and neoplasms of the mediastinum: recent changes in clinical presentation, methods of diagnosis, management, and results. Ann Thorac Surg 1987; 44: 229 – 37. 104. Yamakawa Y, et al. A tentative tumor-node-metastasis classification of thymoma. Cancer 1991; 68: 1984 – 7. 105. Souadjian JV, et al. The spectrum of diseases associated with thymoma. Coincidence or syndrome? Arch Intern Med 1974; 134: 374 – 9. 106. Wilkins EW Jr. Thymoma. In Pearson et al. (ed) Thoracic Surgery. New York: Churchill Livingstone, 1995: 1419 – 27. 107. Blossom GB, et al. Thymectomy for myasthenia gravis. Arch Surg 1993; 128: 855 – 62.
255
108. Marx A, et al. Paraneoplastic autoimmunity in thymus tumors. Dev Immunol 1998; 6: 129 – 40. 109. Berrih-Aknin S, et al. The role of the thymus in myasthenia gravis: immunohistological and immunological studies in 115 cases. Ann N Y Acad Sci 1987; 505: 50 – 70. 110. Masaoka A, et al. Thymomas associated with pure red cell aplasia. Histologic and follow-up studies. Cancer 1989; 64: 1872 – 8. 111. Zeok JV, et al. The role of thymectomy in red cell aplasia. Ann Thorac Surg 1979; 28: 257 – 60. 112. Yip D, et al. Thymoma and agranulocytosis: two case reports and literature review. Br J Haematol 1996; 95: 52 – 6. 113. Welsh JS, et al. Association between thymoma and second neoplasms. JAMA 2000; 283: 1142 – 3. 114. Batra P, et al. Mediastinal masses: magnetic resonance imaging in comparison with computed tomography. J Natl Med Assoc 1991; 83: 969 – 74. 115. Faletra F, et al. Transesophageal echocardiography in the evaluation of mediastinal masses. J Am Soc Echocardiogr 1992; 5: 178 – 86. 116. Sasaki M, et al. Differential diagnosis of thymic tumors using a combination of 11C-methionine PET and FDG PET. J Nucl Med 1999; 40: 1595 – 601. 117. Ferguson MK, et al. Selective operative approach for diagnosis and treatment of anterior mediastinal masses. Ann Thorac Surg 1987; 44: 583 – 6. 118. Grief JSA, et al. Percutaneous core needle biopsy in the diagnosis of mediastinal tumors. Lung Cancer 1999; 25: 169 – 73. 119. Bergh NP, et al. Tumors of the thymus and thymic region: I. Clinicopathological studies on thymomas. Ann Thorac Surg 1978; 25: 91 – 8. 120. Kornstein MJ, et al. Cortical versus medullary thymomas: a useful morphologic distinction? Hum Pathol 1987; 19: 1335 – 9. 121. Haniuda M, et al. Adjuvant radiotherapy after complete resection of thymoma. Ann Thorac Surg 1992; 54: 311 – 5. 122. Mornex F, et al. Radiotherapy and chemotherapy for invasive thymomas: a multicentric retrospective review of 90 cases. The FNCLCC trialists. Federation Nationale des Centres de Lutte Contre le Cancer. Int J Radiat Oncol Biol Phys 1995; 32: 651 – 9. 123. Chung D. Thymic carcinoma. Analysis of nineteen clinicopathological studies. Thorac Cardiovasc Surg 2000; 48: 114 – 9. 124. Maggi G, et al. Thymoma: results of 241 operated cases. Ann Thorac Surg 1991; 51: 152 – 6. 125. Nakahara K, et al. Thymoma: results with complete resection and adjuvant postoperative irradiation in 141 consecutive patients. J Thorac Cardiovasc Surg 1988; 95: 1041 – 7. 126. Ruffini E, et al. Recurrence of thymoma: analysis of clinicopathologic features, treatment, and outcome. J Thorac Cardiovasc Surg 1997; 113: 55 – 63. 127. Pollack A, et al. Thymoma: treatment and prognosis. Int J Radiat Oncol Biol Phys 1992; 23: 1037 – 43. 128. Singhal S, et al. Comparison of stages I-II thymoma treated by complete resection with or without adjuvant radiation. Ann Thorac Surg 2003; 76: 1635 – 41; discussion 1641 – 2. 129. Roviaro GC, et al. Major thoracoscopic operations: pulmonary resection and mediastinal mass excision. Int Surg 1996; 81: 354 – 8. 130. Cowen D, et al. Thymoma: results of a multicentric retrospective series of 149 non-metastatic irradiated patients and review of the literature. FNCLCC trialists. Federation Nationale des Centres de Lutte Contre le Cancer. Radiother Oncol 1995; 34: 9 – 16. 131. Koh WJLP, Thomas C. Thymoma: radiation and chemotherapy. In Wood DTC (ed) Mediastinal Tumors: Update. New York: SpringerVerlag, 1995: 19 – 25. 132. Wilkins KB, et al. Clinical and pathologic predictors of survival in patients with thymoma. Ann Surg 1999; 230: 562 – 72; discussion 572 – 4. 133. Curran WJ Jr, et al. Invasive thymoma: the role of mediastinal irradiation following complete or incomplete surgical resection. J Clin Oncol 1988; 6: 1722 – 7. 134. Schmidt-Wolf IG, et al. Malignant thymoma: current status of classification and multimodality treatment. Ann Hematol 2003; 82: 69 – 76.
256
THORACIC TUMORS
135. Strobel P, et al. Tumor recurrence and survival in patients treated for thymomas and thymic squamous cell carcinomas: a retrospective analysis. J Clin Oncol 2004; 22: 1501 – 9. 136. Kondo K, Monden Y. Therapy for thymic epithelial tumors: a clinical study of 1,320 patients from Japan. Ann Thorac Surg 2003; 76: 878 – 84; discussion 884 – 5. 137. Ciernik IF, Meier U, Lutolf UM. Prognostic factors and outcome of incompletely resected invasive thymoma following radiation therapy. J Clin Oncol 1994; 12: 1484 – 90. 138. Arriagada R, et al. Invasive carcinoma of the thymus. A multicenter retrospective review of 56 cases. Eur J Cancer Clin Oncol 1984; 20: 69 – 74. 139. Fletcher GH. Clinical dose response curves of human malignant epithelial tumours. Br J Radiol 1973; 46: 151. 140. Haniuda M, et al. Recurrence of thymoma: clinicopathological features, re-operation, and outcome. J Surg Oncol 2001; 78: 183 – 8. 141. Kirschner PA. Reoperation for thymoma: report of 23 cases. Ann Thorac Surg 1990; 49: 550 – 4; discussion 555. 142. Urgesi A, et al. Aggressive treatment of intrathoracic recurrences of thymoma. Radiother Oncol 1992; 24: 221 – 5. 143. Eng TY, et al. Thymic carcinoma: state of the art review. Int J Radiat Oncol Biol Phys 2004; 59: 654 – 64. 144. Lucchi M, et al. The multimodality treatment of thymic carcinoma. Eur J Cardiothorac Surg 2001; 19: 566 – 9. 145. Mayer R, et al. Radiotherapy for invasive thymoma and thymic carcinoma. Clinicopathological review. Strahlenther Onkol 1999; 175: 271 – 8. 146. Weide LG, et al. Thymic carcinoma. A distinct clinical entity responsive to chemotherapy. Cancer 1993; 71: 1219 – 23. 147. Koizumi T, et al. Chemotherapy for advanced thymic carcinoma: clinical response to cisplatin, doxorubicin, vincristine, and cyclophosphamide (ADOC chemotherapy). Am J Clin Oncol 2002; 25: 266 – 8. 148. Loehrer PJ Sr, et al. Cisplatin, doxorubicin, and cyclophosphamide plus thoracic radiation therapy for limited-stage unresectable thymoma: an intergroup trial. J Clin Oncol 1997; 15: 3093 – 9. 149. Bonomi PD, et al. EST 2582 phase II trial of cisplatin in metastatic or recurrent thymoma. Am J Clin Oncol 1993; 16: 342 – 5. 150. Hu E, Levine J. Chemotherapy of malignant thymoma. Case report and review of the literature. Cancer 1986; 57: 1101 – 4. 151. Kirkove C, et al. Dramatic response of recurrent invasive thymoma to high doses of corticosteroids. Clin Oncol (R Coll Radiol) 1992; 4: 64 – 6. 152. Boston B. Chemotherapy of invasive thymoma. Cancer 1976; 38: 49 – 52. 153. Highley MS, et al. Treatment of invasive thymoma with single-agent ifosfamide. J Clin Oncol 1999; 17: 2737 – 44. 154. La Rocca RV, et al. Suramin therapy for malignant thymoma: a case report. Eur J Cancer 1994; 30A: 718 – 9. 155. Chahinian AP. Chemotherapy of thymomas and thymic carcinomas. Chest Surg Clin N Am 2001; 11: 447 – 56. 156. Jaffrey IS, Denefrio JM, Chahinian P. Response to maytansine in a patient with malignant thymoma. Cancer Treat Rep 1980; 64: 193 – 4. 157. Donovan PJ, Foley JF. Chemotherapy in invasive thymomas: five case reports. J Surg Oncol 1986; 33: 14 – 7. 158. Chahinian AP, et al. Phase I study of weekly maytansine given by iv bolus or 24-hour infusion. Cancer Treat Rep 1979; 63: 1953 – 60. 159. Stolinsky DC, et al. Clinical trial of weekly doses of vinblastine (NSC-49842) combined with vincristine (NSC-67574) in malignant lymphomas and other neoplasms. Cancer Chemother Rep 1973; 57: 477 – 80. 160. Loehrer PJ Sr et al., The Eastern Cooperative Oncology Group, Southwest Oncology Group and Southeastern Cancer Study Group. Cisplatin plus doxorubicin plus cyclophosphamide in metastatic or recurrent thymoma: final results of an intergroup trial. J Clin Oncol 1994; 12: 1164 – 8. 161. Lara PN Jr, Bonomi PD, Faber LP. Retreatment of recurrent invasive thymoma with platinum, doxorubicin, and cyclophosphamide. Chest 1996; 110: 1115 – 7. 162. Berthaud P, Le Chevalier T, Tursz T. Effectiveness of interleukin-2 in invasive lymphoepithelial thymoma. Lancet 1990; 335: 1590.
163. Gordon MS, et al. A phase II trial of subcutaneously administered recombinant human interleukin-2 in patients with relapsed/refractory thymoma. J Immunother Emphasis Tumor Immunol 1995; 18: 179 – 84. 164. Butler WM, et al. Metastatic thymoma with myasthenia gravis: complete remission with combination chemotherapy. Cancer 1982; 50: 419 – 22. 165. Loehrer PJ, et al. Remission of invasive thymoma due to chemotherapy. Two patients treated with cyclophosphamide, doxorubicin, and vincristine. Chest 1985; 87: 377 – 80. 166. Kosmidis PA, Iliopoulos E, Pentea S. Combination chemotherapy with cyclophosphamide, adriamycin, and vincristine in malignant thymoma and myasthenia gravis. Cancer 1988; 61: 1736 – 40. 167. Goldel N, et al. Chemotherapy of invasive thymoma. A retrospective study of 22 cases. Cancer 1989; 63: 1493 – 500. 168. Evans WKTD, et al. Combination chemotherapy in invasive thymoma: role of COPP. Cancer 1980; 46: 1523. 169. Daugaard G, Hansen HH, Rorth M. Combination chemotherapy for malignant thymoma. Ann Intern Med 1983; 99: 189 – 90. 170. Klippstein TH, et al. High-dose adriamycin (ADM) and cis-platinum (DDP) in advanced soft-tissue sarcomas and invasive thymomas. A pilot study. Cancer Chemother Pharmacol 1984; 13: 78 – 81. 171. Mitrou PS, Bergmann L, Tuengerthal S. Induction of complete remission with adriamycin and cis-platinum in invasive thymoma. Dtsch Med Wochenschr 1982; 107: 1667 – 70. 172. Fornasiero A, et al. Chemotherapy of invasive or metastatic thymoma: report of 11 cases. Cancer Treat Rep 1984; 68: 1205 – 10. 173. Campbell MG, Pollard R, Al-Sarraf M. A complete response in metastatic malignant thymoma to cis-platinum, doxorubicin and cyclophosphamide: a case report. Cancer 1981; 48: 1315 – 7. 174. Shin DM, et al. A multidisciplinary approach to therapy for unresectable malignant thymoma. Ann Intern Med 1998; 129: 100 – 4. 175. Loehrer PJ Sr, et al. Combined etoposide, ifosfamide, and cisplatin in the treatment of patients with advanced thymoma and thymic carcinoma: an intergroup trial. Cancer 2001; 91: 2010 – 5. 176. Fornasiero A, et al. Chemotherapy for invasive thymoma. A 13-year experience. Cancer 1991; 68: 30 – 3. 177. Rea F, et al. Chemotherapy and operation for invasive thymoma. J Thorac Cardiovasc Surg 1993; 106: 543 – 9. 178. Chahinian AP, et al. Treatment of invasive or metastatic thymoma: report of eleven cases. Cancer 1981; 47: 1752 – 61. 179. Chahinian AP, Holland JF, Bhardwaj S. Chemotherapy for malignant thymoma. Ann Intern Med 1983; 99: 736. 180. Giaccone G, et al. Cisplatin and etoposide combination chemotherapy for locally advanced or metastatic thymoma. A phase II study of the European Organization for Research and Treatment of Cancer Lung Cancer Cooperative Group. J Clin Oncol 1996; 14: 814 – 20. 181. Macchiarini P, et al. Neoadjuvant chemotherapy, surgery, and postoperative radiation therapy for invasive thymoma. Cancer 1991; 68: 706 – 13. 182. Dy C, et al. Undifferentiated epithelial-rich invasive malignant thymoma: complete response to cisplatin, vinblastine, and bleomycin therapy. J Clin Oncol 1988; 6: 536 – 42. 183. Tandan R, et al. Metastasizing thymoma and myasthenia gravis. Favorable response to glucocorticoids after failed chemotherapy and radiation therapy. Cancer 1990; 65: 1286 – 90. 184. Palmieri G, et al. Successful treatment of a patient with a thymoma and pure red-cell aplasia with octreotide and prednisone. N Engl J Med 1997; 336: 263 – 5. 185. Palmieri G, et al. Somatostatin analogs and prednisone in advanced refractory thymic tumors. Cancer 2002; 94: 1414 – 20. 186. Kurup ABM, et al. Phase II Study of Gefitinib Treatment in Advanced Thymic Malignancies. Orlando, Florida: American Society of Clinical Oncology, 2005. 187. Regnard JF, et al. Prognostic factors and long-term results after thymoma resection: a series of 307 patients. J Thorac Cardiovasc Surg 1996; 112: 376 – 84. 188. Harper PG, Addis B. Unusual tumors of the mediastinum. In Williams C, Krikorian J, Green M, Ragahvan D (ed 1) Textbook of Uncommon Cancer. Chichester: Wiley, 1988. 189. Harper PG, et al. Ifosfamide monotherapy demonstrates high activity in malignant thymoma. Proc ASCO 1991; 10: 1049.
Section 5 : Thoracic Tumors
21
Primary Lymphomas of the Lung Francis C. Nichols and Stephen D. Cassivi
HISTORICAL BACKGROUND In marked contrast to secondary pulmonary involvement by lymphoma, which has an incidence of 25 to 40%, primary pulmonary lymphoma (PPL) is rare. PPL accounts for only 3.6% of extranodal lymphomas and less than 1% of all primary pulmonary malignancies.1 – 3 The rarity of PPL is demonstrated by several reports. A retrospective review of 10 134 patients treated for primary lung tumors at our institution from 1980 through 1990 found only 33 patients (0.33%) with PPL.4 In a review of the cytology records from five large institutions, Bardales and colleagues found only 20 patients with PPL over a 20-year period.5 A review of the Armed Forces Institute of Pathology records revealed only 146 patients with PPL over a 23-year period.6 The classification of PPL has been confusing. Historically, attempts have been made to classify PPL in the same way as lymphoma arising in lymph nodes. However, it is now recognized that nodal and extranodal lymphomas differ markedly and need to be considered separately.7 If PPLs are divided into Hodgkin’s disease (HD) and non-Hodgkin’s lymphomas (NHL), a notable difference is immediately apparent. Primary pulmonary HD is extremely rare, being reported only in isolated cases.8,9 In our institutional series, all PPLs were NHL.4 Radin reported one case of primary pulmonary HD and reviewed an additional 60 cases found in the literature.10 When lymph node involvement was a prerequisite for the diagnosis of malignant lymphoma, pathologists were reluctant to accept a diagnosis of lymphoma in extranodal sites. The lung was such a site. Pulmonary lymphomas were first described in detail by Saltzstein in 1963, and, for many years, his criteria was the basis for differentiating pulmonary lymphocytic lesions.11 Saltzstein recognized that many patients diagnosed with PPL followed a long and relatively benign course, as distinct from the aggressive behavior some pulmonary lymphocytic lesions can follow. For those pulmonary lymphocytic lesions that lacked aggressive behavior, Saltzstein introduced the term pseudolymphoma. Moreover, Saltzstein proposed that most pulmonary lymphocytic lesions were indeed pseudolymphomas
because of the maturity of the cytologic appearance, presence of reactive germinal centers, rarity of nodal metastasis and excellent long-term prognosis.11 Over time, a number of observers reported cases of pseudolymphoma progressing to lymphoma, thus raising practical questions about the reliability of Saltzstein’s criteria.12 Many of Saltzstein’s pseudolymphoma criteria are now recognized as characteristic of extranodal lymphomas. With the recognition that extranodal lymphomas differ markedly from the more common nodal lymphomas, the viewpoint that pseudolymphomas represent low-grade lymphomas now prevails. This is supported by immunoglobulin gene rearrangement analysis demonstrating that most of these tumors are monoclonal B cell proliferations and thus malignant.13,14 PPL is not a single disease but rather encompasses the histologic spectrum of malignant lymphomas, including HD and NHL. The differential diagnosis of primary pulmonary NHL includes lymphomas of mucosa-associated lymphoid tissue (MALTomas), diffuse large cell NHL, lymphomatoid granulomatosis, post-transplant lymphoproliferative disorders, peripheral T cell NHL, and intravascular lymphomatosis.15
EPIDEMIOLOGY It is estimated that 63 740 people will develop lymphoma in the United States in 2005, representing 4.6% of the new cases of cancer in this country. This incidence includes 7350 patients with HD and 56 390 patients with NHL. Estimated deaths from lymphoma in the United States for 2005 were 20 610 or 3.6% of the total estimated cancer deaths.16 As previously mentioned, PPL is rare. The peak incidence of PPL is in the sixth decade of life and it rarely presents below 30 years of age.3,6,17 There is a slight male predominance, though that varies amongst series. In our institutional review, the median age at diagnosis was 65 years (range: 20–87 years) with 8 women and 15 men.4 In a review by Li and colleagues, patients with low-grade PPL averaged 60 years of age and those with high-grade PPL 70 years.18 In reviewing the French experience, Cordier and colleagues found the mean age at diagnosis to be 58.4 years, and the male-to-female ratio equal.19
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
258
THORACIC TUMORS
In contrast to and separate from PPL, primary NHL of the mediastinum has been reported to occur in 6% of adults and 26% of children with NHL.20,21 Primary mediastinal NHL characteristically presents as an anterior mediastinal mass. There are other unusual presentations for primary intrathoracic NHLs separate from PPL, and these include primary tracheobronchial NHL and intravascular lymphomatosis (angiotropic lymphoma). There are reports of intravascular lymphomatosis, a form of systemic B cell lymphoma typically involving the skin or central nervous system, presenting as PPL.22,23 Congenital and acquired immunodeficiency states including acquired immunodeficiency syndrome (AIDS), solid organ transplantation, and allogenic bone-marrow transplantation are associated with the development of lymphoproliferative disorders.15 In AIDS, the lung may be primarily as well as secondarily involved. In reporting on AIDS-related NHL, Eisner and colleagues found pleuropulmonary involvement in 71% of patients at autopsy.24 Post-transplant lymphoproliferative disorders occur in solid organ and bone-marrow transplant recipients. Primary pulmonary involvement may occur, including in lung transplant recipients where the disease may present in the transplanted lung.
PATHOLOGY AND CLASSIFICATION PPL is defined as clonal lymphoid proliferation affecting one or both lungs (parenchyma and/or bronchi) in patients without detectable extrapulmonary involvement at diagnosis or during the subsequent three months.2,25 Although a variety of morphologic subtypes can be seen, the most common is low-grade (small lymphocytic) B cell lymphoma; highgrade (large-cell) B cell lymphoma and predominantly T cell angioimmunoproliferative lesions (lymphomatoid granulomatosis) account for most of the remaining tumors. It is now widely believed that most primary extranodal lymphomas, including those that arise in the lung, are derived from the MALT. In the lung, the tumors are believed to arise from marginal zone cells (centrocyte-like cells) which are present in normal or hyperplastic bronchusassociated lymphoid tissue (BALT).26 BALT forms part of the wider spectrum of MALT, which was first described by Bienenstock and colleagues in 1973.27,28 MALT is organized aggregates of lymphoid tissue present in the bronchial walls of many species, including humans; 60% of lymphoid cells in these areas are B cells. Most PPL arise from MALT.28 In the review by Cordier and colleagues, 78.3% of the PPL were thought to arise from MALT.19 In our institutional series, 22 of the 33 patients (66.7%) had lymphomas from MALT.4 In a more recent Mayo Clinic series, primary pulmonary MALT lymphoma was found in 73% of the patients, the remaining 27% having the nonMALT type.1 The pathogenesis of MALT lymphoma in the gastrointestinal tract has been shown to occur during Helicobacter pylori infection, and B cells responsive to Helicobacter antigens have been demonstrated. The similarity of MALT lymphomas at all sites suggests that chronic antigen stimulation is the cause; what is unknown about pulmonary MALT lymphomas is what that antigen might be.
It has been speculated that some background stimulus such as cigarette smoke or infection leads to BALT hyperplasia, which is then followed by neoplastic transformation.29 However, many patients with pulmonary MALT lymphomas are nonsmokers and histologic evidence of local or diffuse lymphoid hyperplasia is only occasionally present.30 However, established associations do include human immunodeficiency virus (HIV), human T cell lymphoma virus-I (HTLV-I), pesticide and black hair dye exposure and dietary intake of fat.31 – 33 Primary pulmonary HD is thought to arise in lymphoid follicles or peribronchial lymph nodes that are scattered throughout the lung. While the cause of HD also remains unknown, 20–80% of patients have the Epstein-Barr virus (EBV) present in Reed-Sternberg cells.34 – 36 The histologic types of primary pulmonary HD most commonly seen are the nodular sclerosing and mixed cellularity forms. The multiple classification schemes for HD and NHL have resulted in confusion amongst clinicians and pathologists. The Kiel system refers to these tumors as immunocytomas, the Rappaport system as well-differentiated lymphocytic lymphomas, the Lukes and Collins as lymphocytic, plasmacytoid lymphomas, and according to the Working Formulation for Clinical Usage they should be named small lymphocytic, plasmacytoid lymphomas.17,28,37,38 The Working Formulation for Clinical Usage is a morphologic scheme which groups NHLs by natural history and response to therapy into favorable low-grade lymphomas and unfavorable intermediate and high-grade lymphomas. This was the most commonly used scheme in North America.28 However, in order to recognize new entities and refine recognized disease categories, the International Lymphoma Study Group reported a Revised European-American Lymphoma (REAL) classification using morphologic, immunologic, and genetic techniques.37 Even though this new scheme provides a structural concept for lymphomas, it is not biologically correct in all patients. More recently, Harris and colleagues and Chan have described the World Health Organization (WHO) classification of lymphoma.39 – 41 With no established method of grading pulmonary NHLs, the principles of the Keil classification are used to subdivide tumors as low- and high-grade, mirroring the scheme used for gastric lymphomas. In the low-grade form, three cell types have been described – centrocyte-like, small lymphocyte, and monocytoid forms – which probably represent histological variants of the same neoplastic cell. Immunohistochemical analysis shows positive reactions for CD20 and CD3 in the atypical lymphoid cells in virtually all tumors, confirming B cell differentiations.29 Because the degree of cytologic atypia of the lymphoid cells is usually slight, the diagnosis may be difficult on the basis of histology alone. In these difficult cases, ancillary immunohistochemical and molecular biologic studies are often helpful. The histologic diagnosis of primary high-grade lymphoma is usually easily made by means of traditional pathologic features. The high-grade form of NHL consists of confluent blast cells. Other characteristics of high-grade NHL include alveolar space involvement, focal or massive necrosis, and pleural
PRIMARY LYMPHOMAS OF THE LUNG
infiltration. Polymerase chain reaction (PCR) revealed monoclonal bands in 60% of low-grade tumors, but only 25% in high-grade tumors.42
CLINICAL FEATURES While the signs and symptoms of primary pulmonary NHL may be quite varied, most patients are asymptomatic at presentation. The disease is often discovered on a screening chest radiograph.3,6,19 (Figure 1). When symptomatic, pulmonary symptoms are more common than systemic symptoms and may include cough, chest pain, dyspnea, and hemoptysis.6,19 In Li and colleagues’ review, 21 of 62 patients (33.9%) had symptoms which included dyspnea and cough.18 In the Armed Forces Institute of Pathology review 52 of 123 patients (42.3%) were symptomatic.6 Constitutional symptoms, when present, may include fatigue, fever, night sweats, or weight loss.6,19 Symptoms may be present for several weeks to months prior to presentation.3,6 The physical exam is normal in most patients. In a recent Mayo Clinic series involving 48 patients operated on for primary pulmonary NHL, 18 patients (37.5%) were asymptomatic and 60 (62.5%) presented with pulmonary or systemic symptoms, or both. Cough and dyspnea were the most common pulmonary symptoms occurring in 22 and 11 patients respectively. Fatigue, weight loss, and fever occurred in 15, 13, and 11 patients respectively.1 In contrast to patients with primary pulmonary NHL, patients with pulmonary HD tend to be symptomatic with B symptoms (fever, night sweats and unexplained weight loss).
Figure 1 Solitary pulmonary nodule. Posteroanterior chest radiograph showing a solitary indeterminate pulmonary nodule in the right upper lobe. Right upper lobectomy was performed; pathology revealed non-Hodgkin’s lymphoma.
259
Cough is present in almost 50% of the patients with chest pain, dyspnea, and hemoptysis being less frequent findings.10 Endobronchial involvement in HD, NHL, and posttransplant lymphoproliferative disorders is rare. Cough, dyspnea, respiratory distress, obstructive pneumonia, and abscess formation secondary to obstructive lesions in an airway may occur.9,19,43 Patients with lymphomatoid granulomatosis (or T cell–rich B cell lymphoma) usually present with lower respiratory tract symptoms including chest pain, cough, and dyspnea. Fever and skin rash are common. Neurologic symptoms are present in approximately 33% of patients.44,45
IMAGING Patients who are asymptomatic are usually found to have primary pulmonary NHL by routine chest X ray. A solid nodule (Figure 1), up to several centimeters in diameter,6 is seen in approximately 50% of cases. Solitary nodules are most commonly located in the lower lobes.6 However, multiple pulmonary nodules may also be seen (Figure 2), pulmonary infiltrates (Figure 3), or a pulmonary effusion may be present in 25% of patients.6 Diffuse hilar lymphadenopathy and cavitation of the nodules is rare. In the series reported by Cordier et al., 79% of patients had localized opacities discovered by routine chest X ray, tomograms, or chest computerized tomography (CT).19 Tumor size varied from 1.0 cm to a mass filling the entire hemithorax. Air bronchograms were present in about half the patients and were more commonly seen on the chest CT (Figure 4). The number of masses
Figure 2 Multiple pulmonary nodules. Posteroanterior chest radiograph showing bilateral indeterminate pulmonary nodules. Wedge excision via right video-assisted thoracic surgery demonstrated high-grade nonHodgkin’s lymphoma.
260
THORACIC TUMORS
Figure 3 Bilateral pulmonary infiltrates. Posteroanterior chest radiograph showing diffuse bilateral pulmonary infiltrates. Multiple wedge excisions via left video-assisted thoracic surgery demonstrated high-grade non-Hodgkin’s lymphoma.
was one in 54% of patients, two in 14%, three or more in 7%, and bilateral in 16%. Pleural effusion was seen in only 7%, and no patient had pathologically enlarged mediastinal nodes.19 Primary pulmonary HD presents with multiple nodular lesions in the majority of patients. The nodular changes are most common in the upper lobes.10 Less common radiographic findings include cavitation of nodules, diffuse reticulonodular infiltrates, pneumonic consolidation and pleural effusions.9 The chest radiographic findings in patients with lymphomatoid granulomatosis are indistinguishable from those found in Wegener’s granulomatosis. Other radiographic findings include diffuse alveolar or interstitial infiltrates and, less commonly, a localized infiltrate or mass.44,46
DIAGNOSIS Laboratory evaluation is usually unrevealing, and mean lymphocyte count is usually normal. However, a monoclonal immunoglobulin spike, typically IgA or IgM and rarely IgG, can be found in approximately 20–30% of PPL patients.28 The diagnosis of PPLs requires an adequate biopsy specimen. While bronchoscopy has been reported to yield adequate diagnostic tissue in some patients, it is not recognized as the procedure of choice in establishing the initial diagnosis of NHL or HD. In the Mayo Clinic series, fiberoptic bronchoscopy was performed in 39 of the 48
Figure 4 Primary pulmonary lymphoma seen as a chronic consolidation. Chest computerized tomography (CT) shows a homogeneous opacity containing air bronchograms in the left lower lobe.
patients and provided a tentative diagnosis in 7 (18%).1 Fiberoptic bronchoscopy was performed in 60 of the 69 patients in the Cordier and colleagues series.19 While macroscopic abnormalities were seen in 43%, transbronchial pulmonary biopsy revealed lymphomatous lesions in only three patients. Bronchoalveolar lavage (BAL) was performed in 13 patients. Immunocytochemical studies of two of these patients showed monotypic immunoglobulin. Nonetheless, transbronchial biopsy as well as BAL have been used to diagnose PPL.19,46 – 50 Fine-needle aspiration biopsy has been used in the initial diagnosis, staging and evaluation of residual masses after therapy. Recent publications have supported the role of radiologically-guided core-needle biopsies in the diagnosis of malignant lymphoma.51 – 53 Nonetheless, establishing a diagnosis of lymphoma on limited tissue presumes that the biopsy is representative of the entire neoplastic process. The priority in evaluating any lymphoid lesion is to obtain adequate tissue for high-quality histologic sections and enough material for supportive ancillary studies as required to establish the diagnosis. In our experience, these needle biopsy techniques are of limited value. Except in critically ill patients, most can withstand either an open or minimally invasive biopsy procedure. While the results of these techniques are less than optimal, the recent development of genetic analysis may lead
PRIMARY LYMPHOMAS OF THE LUNG
to more successful bronchoscopic or fine-needle aspiration diagnosis in patients with PPL. By using the PCR, small fragments of tissue obtained bronchoscopically can provide a diagnosis. Subramanian and colleagues reported that PCR is superior to the cytochemical methods because it requires less tissue, does not have to be frozen, and can report results in one day, whereas some of the blotting techniques can take days to weeks.28,48 Nonetheless, at present one has significant reservations regarding the routine use of needle biopsy to establish a definitive diagnosis of PPL because of the broad range of possible lymphoproliferative lesions. Thus, in order to obtain adequate tissue for diagnosis, a lung biopsy, either through thoracoscopic techniques or, when a curative resection is anticipated, via either thoracoscopic or open thoracotomy, is strongly recommended. An excisional or incisional biopsy is performed to obtain an adequate tissue for histologic analysis. A portion of the tissue should be fresh frozen to allow for immunoglobulin gene rearrangement analysis. However, if the tissue is not fresh frozen, it is still possible to analyze gene rearrangements by PCR.28,48
STAGING Once the diagnosis is established, defining the extent of disease is necessary for both the determination of appropriate treatment options and establishing the prognosis for the patient. In both NHL and HD, the modified Ann Arbor system of classification is used for staging these patients (Table 1).54,55 This system is based on the number of sites involved. It describes the number and location of lymph node stations involved, whether or not the chest wall or diaphragm is involved, and if there are other organs involved. The staging evaluation should include a history, physical examination, complete blood count, chemistry panel (liver function tests, serum calcium, creatinine, and albumin levels), serum lactate dehydrogenase (LDH), erythrocyte sedimentation rate, chest radiograph, bone-marrow aspiration, and CT of the chest, abdomen, and pelvis with intravenous contrast.15
TREATMENT The treatment of PPL is based on the histologic subtype, the extent of disease, and any co-morbid medical conditions that may exist. Most important is the histology. Table 1 Surgical (pathological) staging classification.a
IE II1E II2E II2EW III IV a
Modified Ann Arbor staging classification.
Involvement of the lung only (may be bilateral) Lung and hilar lymph nodes Lung and mediastinal lymph nodes Lung and adjacent chest wall or diaphragm Lung and lymph nodes below diaphragm Diffuse involvement of one or more extralymphatic organs/tissue
261
Surgical intervention, whether by traditional thoracotomy or video-assisted thoracic surgery (VATS) is often necessary to establish a definitive diagnosis of PPL. However, surgery also may play a therapeutic role in the management of PPL. For example, 74–88% of patients with primary pulmonary NHL are reported to have low-grade B cell neoplasms.18,19,30 MALTomas accounted for 69 to 78% of the cases. Cordier and colleagues reported an overall survival of 100 and 94% at 2 and 5 years respectively.19 Forty-two (69%) of the patients underwent surgical resection. Twenty-one of the 42 resected patients received no additional treatment. Sixteen patients were treated with chemotherapy alone, 3 with radiation, and 3 with no treatment at all. Five patients died, 3 of extrathoracic recurrent lymphoma. In our series, in patients with NHL arising from MALT, a wedge excision was performed in 25, lobectomy in 11, pneumonectomy in 4, and segmentectomy in 2. Complete excision was achieved in 19 (40%) of the 48 patients.1 Median follow-up was 4.2 years (range: 1 month to 16 years). At last follow-up, 27 patients (56%) were alive without evidence of disease, 3 (6%) were alive with disease, and the disease status was unknown in 1 patient. For patients with MALTomas, the survival rate at 1, 5, and 10 years was 91, 68, and 53% respectively.1 Prognostic factors for 5-year survival were studied by univariate analysis. Of the seven factors studied (presentation, smoking history, bilateral disease, histology, postoperative stage, complete resection, and adjuvant therapy) none was found to significantly influence overall survival.1 Li and colleagues reported a 5year survival of 84%18 while Koss and colleagues reported 10-year survival of 85%.6 A remarkable clinical feature of MALTomas is their tendency to remain localized for extended periods of time. It is probably true that patients with a low-grade lymphoma have a survival that is not statistically different from that of the normal population. However, there have been reports of transformations of up to 5% of these lymphomas into a high-grade lymphoma up to 78 months after the initial diagnosis.18 The survival prospects of patients with low-grade lymphoma are quite good, regardless of the type of resection. If the patient can tolerate a complete resection via a lobectomy, then that would appear to be a reasonable approach.56 Recognizing the retrospective nature of our study, our data do not support the use of extended resections such as pneumonectomy in the treatment of PPL.1 If it is not possible to completely remove all disease, then incisional biopsy to obtain diagnostic tissue is appropriate. Thus, an overall approach to MALTomas is as follows: In patients who have undergone surgical resection of localized disease and in those without bulky disease, initial observation is prudent. In patients with bulky disease, cytotoxic chemotherapy may be a reasonable option. However, while patients treated with chemotherapy tend to achieve an earlier remission, there does not appear to be an improvement in disease-free or overall survival. An alternative approach that might be considered is the use the of anti CD20-antibody rituximab (Rituxan).57 – 59 Longterm disease-free and overall survival has been reported with rituximab even after use of classic cytotoxic chemotherapy.59 In addition, thalidomide has been reported to effect a good
262
THORACIC TUMORS
partial response in a single patient with a pulmonary MALT lymphoma.60 The prognosis and treatment of high-grade PPL is much more pessimistic. However, in patients with high-grade lymphoma, radiation therapy and/or cytotoxic chemotherapy is indicated. The five-year survival of patients with a highgrade lymphoma is only 44–60%, much less than that of those with low-grade lymphomas.4 Combined modality therapy including possible surgical resection, radiation therapy, and possibly cytotoxic chemotherapy appears to be beneficial in patients with unfavorable non-MALT type PPL. For primary pulmonary HD, therapy is generally chemotherapy (e.g. doxorubicin, bleomycin, vincristine, and doxorubicin or ABVD). It must be noted that at the dose required for the primary treatment of pulmonary HD, radiation-induced lung toxicity to the lung parenchyma, is not reversible. The prognosis is dependent on the patient’s age and the extent of pulmonary disease.5 In the series by Yousem and colleagues, 9 of 15 patients experienced a relapse between three and 48 months after therapy (average 14.5 months).9 After an average follow-up of 39.5 months, 6 patients have died, 1 has persistent disease, and 8 remain disease-free. In summary, PPL is a rare disease which is usually a lowgrade NHL related to MALT. The priority in evaluating any lymphoid lesion is to obtain high-quality histologic sections and enough material for any other ancillary studies. Therefore, diagnosis is still best accomplished by obtaining an adequate sample of tissue for pathological examination via either VATS or open thoracotomy. Low-grade primary pulmonary NHL has an excellent prognosis and can usually be treated by surgical resection alone. High-grade lymphomas will require additional chemotherapy or radiation therapy. Treatment of primary pulmonary HD is usually with chemotherapy and/or radiation therapy; long-term survival, however, is good.
REFERENCES 1. Ferraro P, et al. Primary non-Hodgkin’s lymphoma of the lung. Ann Thorac Surg 2000; 69: 993 – 7. 2. Freeman C, Berg JW, Cutler SJ. Occurrence and prognosis of extranodal lymphomas. Cancer 1972; 29: 1397 – 406. 3. L’Hoste RJ, et al. Primary pulmonary lymphomas: a clinicopathologic analysis of 36 cases. Cancer 1984; 54: 1397 – 406. 4. Miller DL, Allen MS. Rare pulmonary neoplasms. Mayo Clin Proc 1993; 68: 492 – 8. 5. Bardales RH, et al. Exfoliative respiratory cytology in the diagnosis of leukemia and lymphomas in the lung. Diagn Cytopathol 1996; 14: 108 – 13. 6. Koss MN, et al. Primary non-Hodgkin’s lymphoma and pseudolymphoma of lung: a study of 161 patients. Hum Pathol 1983; 14: 1024 – 38. 7. Salhany KE, Pietra GC. Extranodal lymphoid disorders. Am J Clin Pathol 1993; 99: 472 – 85. 8. Wood NL, Coltman CA Jr. Localized primary extranodal Hodgkin’s disease. Ann Intern Med 1973; 78: 113 – 8. 9. Yousem SA, Weiss LW, Colby TV. Primary pulmonary Hodgkin’s disease: a clinicopathologic study of 15 cases. Cancer 1986; 57: 1217 – 24. 10. Radin AI. Primary pulmonary Hodgkin’s disease. Cancer 1990; 65: 550 – 63. 11. Saltzstein SL. Primary malignant lymphomas and pseudolymphomas: classification, therapy, and prognosis. Cancer 1963; 16: 928 – 55. 12. Colby TV, Carrington CB. Lymphoreticular tumors and infiltrates of the lung. Pathol Annu 1983; 18: 27 – 70.
13. Arnold A, et al. Immunoglobulin-gene rearrangements as unique clonal markers in human lymphoid neoplasms. N Engl J Med 1983; 309: 1593 – 9. 14. Weiss LM, Yousem SA, Warnke RA. Non-Hodgkin’s lymphomas of the lung. A study of 19 cases emphasizing the utility of frozen section immunologic studies in differential diagnosis. Am J Surg Pathol 1985; 9: 480 – 90. 15. Ryu JH, Habermann TM. Pulmonary lymphoma: primary and systemic disease. Semin Respir Crit Care Med 1997; 18: 341 – 52. 16. Jemal A, et al. Cancer statistics 2005. CA Cancer J Clin 2005; 55: 10 – 30. 17. Herbert A, et al. Primary malignant lymphoma of the lung: histopathologic and immunologic evaluation of nine cases. Hum Pathol 1984; 15: 415 – 22. 18. Li G, et al. Primary lymphomas of the lung: morphological, immunohistochemical, and clinical features. Histopathology 1990; 16: 519 – 31. 19. Cordier JF, et al. Primary pulmonary lymphomas. A clinical study of 70 cases in nonimmunocompromised patients. Chest 1993; 103: 201 – 8. 20. Lichtenstein AK, et al. Primary mediastinal lymphoma in adults. Am J Med 1980; 68: 509 – 14. 21. Levitt LJ, et al. Primary non-Hodgkin’s lymphoma of the mediastinum. Cancer 1982; 50: 2486 – 92. 22. Stroup RM, et al. Angiotropic (intravascular) large cell lymphoma. A clinicopathologic study of seven cases with unique clinical presentations. Cancer 1990; 66: 1781 – 8. 23. Demirer T, Dail DH, Aboulafia DM. Four varied cases of intravascular lymphomatosis and a literature review. Cancer 1994; 73: 1738 – 45. 24. Eisner MD, et al. The pulmonary manifestations of AIDS-related nonHodgkin’s lymphoma. Chest 1996; 110: 729 – 36. 25. Isaacson PG, Norton AJ. Extranodal Lymphomas. New York: Churchill Livingston, 1994. 26. Harris NL. Low-grade B cell lymphoma of mucosa-associated lymphoid tissue and monocytoid B cell lymphoma: related entities that are distinct from other low-grade B cell lymphomas. Arch Pathol Lab Med 1993; 117: 771 – 5. 27. Bienenstock J, Johnston N, Perey DY. Bronchial lymphoid tissue. Lab Invest 1973; 28: 686 – 98. 28. Allen MS, Miller DL. Primary lymphomas of the lung. In Raghavan D, et al. (eds) Textbook of Uncommon Cancer 2nd ed. Chichester, England: John Wiley and Sons, 1999. 29. Nicholson AG, et al. Pulmonary B cell non-Hodgkin’s lymphomas: the value of immunohistochemistry and gene analysis in diagnosis. Histopathology 1995; 26: 395 – 403. 30. Fiche M, et al. Primary pulmonary non-Hodgkin’s lymphomas. Histopathology 1995; 26: 529 – 37. 31. Kwak LW, Longo DL. Lymphomas. Cancer Chemother Biol Response Modif 1996; 16: 376 – 440. 32. Biggar RJ, Rabkin CS. The epidemiology of acquired immunodeficiency syndrome-related lymphomas. Curr Opin Oncol 1992; 4: 883 – 93. 33. Tukudome S, et al. Incidence of adult T cell leukaemia/lymphoma among human T lymphotropic virus type 1 carriers in Saga Japan. Cancer Res 1989; 49: 226 – 32. 34. Anagnostopoulos I, et al. Demonstration of monoclonal EBV genomes in Hodgkin’s disease and Ki-1 positive anaplastic large cell lymphoma by combined Southern blot and in situ hybridization. Blood 1989; 74: 810 – 6. 35. Brousset P, et al. Detection of Epstein-Barr virus messenger RNA in Reed-Sternberg cells of Hodgkin’s disease by in situ hybridization with biotinylated probes on specially processed modified acetone methyl benzoate xylene (ModAMeX) sections. Blood 1991; 77: 1781 – 6. 36. Herbst H, et al. Epstein-Barr virus latent membrane protein expression in Hodgkin and Reed-Sternberg cells. Proc Natl Acad Sci USA 1991; 88: 4766 – 70. 37. Harris NL, et al. A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood 1994; 84: 1361 – 92. 38. Chan JK, et al. A proposal for classification of lymphoid neoplasms (by the International Lymphoma Study Group). Histopathology 1994; 25: 517 – 36.
PRIMARY LYMPHOMAS OF THE LUNG 39. Harris NL, et al. The World Health Organization classification of haematological malignancies report of the Clinical advisory Committee, Airlie House, Virginia, November 1997. Mod Pathol 2000; 13: 193 – 207. 40. Harris NL, et al. The World Health Organization classification of haematological malignancies report of the Clinical advisory Committee, Airlie House, Virginia, November 1997. Hematol J 2000; 1: 53 – 66. 41. Chan JK. The new World Health Organization classification of lymphomas: the past, the present, and the future. Hematol Oncol 2001; 19: 129 – 50. 42. Diss TC, et al. Detection of monoclonality in low-grade B cell lymphoma using the polymerase chain reaction is dependent on primer selection and lymphoma type. J Pathol 1993; 169: 291 – 5. 43. Rose RM, et al. Endobronchial involvement with non-Hodgkin’s lymphoma: a clinical-radiologic analysis. Cancer 1986; 57: 1750 – 5. 44. Katzenstein AA, Carrington CB, Liebow AA. Lymphomatoid granulomatosis: a clinicopathologic study of 152 cases. Cancer 1979; 43: 360 – 73. 45. Fauci AS, et al. Lymphomatoid granulomatosis: prospective clinical and therapeutic experience over 10 years. N Engl J Med 1982; 306: 68 – 74. 46. Pisani RJ, DeRemee RA. Clinical implications of the histopathologic diagnosis of pulmonary lymphomatoid granulomatosis. Mayo Clin Proc 1990; 65: 151 – 63. 47. Bolton-Maggs PH, et al. Mucosa associated lymphoma of the lung. Thorax 1993; 48: 670 – 2. 48. Subramanian D, et al. Primary pulmonary lymphoma: diagnosis by immunoglobulin gene rearrangement study using a novel polymerase chain reaction technique. Am Rev Respir Dis 1993; 148: 222 – 6.
263
49. Morales FM, Matthews JI. Diagnosis of parenchymal Hodgkin’s disease using bronchoalveolar lavage. Chest 1987; 91: 785 – 7. 50. Oka M, et al. Bronchoalveolar lavage in primary lymphoma with monoclonal gammopathy. Am Rev Respir Dis 1988; 134: 957 – 9. 51. North L, et al. What is the role of fine-needle biopsy in the diagnosis of lymphoma. Am J Roentgenol 1995; 165: 1299. 52. Hanson CA. Fine-needle aspirate and immunotyping: a role in diagnostic hematopathology? Am J Clin Pathol 1994; 101: 555 – 6. 53. Pan JF, et al. Needle aspiration biopsy of malignant lung masses with necrotic centers. Improved sensitivity with ultrasonic guidance. Chest 1993; 103: 1452 – 6. 54. Carbone PP, et al. Report of the committee on Hodgkin’s disease staging classification. Cancer Res 1971; 31: 1860 – 1. 55. Lister TA, Crowther D. Staging for Hodgkin’s disease. Semin Oncol 1990; 17: 696 – 703. 56. Uppal R, Goldstraw P. Primary pulmonary lymphoma. Lung Cancer 1992; 8: 95 – 100. 57. Raderer M, et al. Rituximab for treatment of advanced extranodal marginal zone B cell lymphoma of the mucosa-associated lymphoid tissue lymphoma. Oncology 2003; 65(4): 306 – 10. 58. Ahmed S, et al. Bronchial-associated lymphoid tissue lymphoma: a clinical study of a rare disease. Eur J Cancer 2004; 40(9): 1320 – 6. 59. Chong EA, et al. Regression of pulmonary MALT lymphoma after treatment with rituximab. Leuk Lymphoma 2005; 46(9): 1383 – 6. 60. Kees M, et al. Very good partial response in a patient with MALTlymphoma of the lung after treatment with low-dose thalidomide. Leuk Lymphoma 2005; 46(9): 1379 – 82.
Section 5 : Thoracic Tumors
22
Primary Sarcomas of the Lung
Rachel E. Sanborn, Adriana L. Gonzalez, Thomas M. Ulbright, Guru Sonpavde and Alan B. Sandler
INTRODUCTION Soft tissue sarcomas are uncommon tumors, with an estimated incidence of approximately 9500 cases and 3500 deaths in the United States in 2005.1 Sarcomas derive from mesenchymal tissues, with half arising in the extremities. Sarcomas may arise from the head and neck, retroperitoneum, and trunk, however, origin from a visceral structure is infrequent.2 Although sarcomas may frequently metastasize to the lungs, primary pulmonary sarcomas (PPSs) are very rare, accounting for less than 0.2% of all malignant pulmonary tumors.2,3 The clinical course and treatment approaches for PPSs differs from other primary pulmonary malignancies, and thus the clinician needs to be aware to include these entities in the differential diagnosis during the evaluation of a pulmonary lesion. This chapter reviews the available information on the different types of pulmonary sarcomas, focusing on differentiating pathological characteristics, clinical presentation, differential diagnosis, prognosis, and approaches to treatment.
EPIDEMIOLOGY AND ETIOLOGY PPSs may occur at any age, being described in neonates and in the elderly. There is no evidence of a gender predilection, and no racial predisposition, although the rarity of the tumors precludes the analysis of a large population database for reliable statistics. Most patients are middle aged at the time of diagnosis. In leiomyosarcoma, the most commonly diagnosed type of PPS, there are only 16 cases described in children under 16 years in the literature.4 The second most common type of PPS is fibrosarcoma, although more recently, the reported incidence of this diagnosis is decreasing because of better characterization of this tumor through immunohistochemical and molecular testing. Only 55 cases have been reported and twenty-nine of these occurred in children under 19 years of age.5 – 7 Rhabdomyosarcoma has a bimodal age distribution, with the median age of diagnosis in children being 2 years old (usually embryonal and alveolar subtypes), and the median age of diagnosis in adults
of 57 years (usually the pleomorphic subtype).8 Primary pulmonary angiosarcoma, as an extremely rare tumor, has not yet been described in children.9 Given the heterogeneity as well as the rarity of the subtypes of PPS, no large cohorts have been examined to suggest causative factors. Scattered associations with radiation exposure, Thorotrast exposure, and chemical or insecticide exposures have been made, but no convincing evidence has been presented.
PATHOLOGY Most histologic subtypes of soft tissue sarcoma are represented in the literature on PPS (see Table 1). Pulmonary sarcomas, like soft tissue sarcomas, are generally categorized according to the normal mesenchymal tissues they mimic.10 They arise from the stromal elements of bronchial or vascular walls or from lung parenchymal interstices. While some morphologic features are distinctive, the different histologic types of PPS have many similarities that render definitive identification on the basis of microscopic examination alone unreliable. Immunohistochemistry and electron microscopy, however, can often reveal features that aid in diagnosis. Despite this, tumor grade, size, and location have been found to correlate more strongly with outcome than individual histologic subtypes.11 Criteria for malignancy have been proposed for some tumor types on the basis of mitotic count, atypia, and necrosis, but specific criteria vary with tumor histology.12 At the time of diagnosis, PPS frequently demonstrates local invasion.13 As with the more common soft tissue sarcomas, tumor invasion of the lymphatic system is unusual, with the exception of a few subtypes such as synovial sarcoma. Metastatic spread occurs hematogenously but is not usually found at the time of initial diagnosis.13 The most commonly reported PPSs are leiomyosarcoma, fibrosarcoma, and hemangiopericytoma. Most are grossly round to oval, well circumscribed, and pseudoencapsulated masses (see Figures 1–3). Peripheral lesions may invade the adjacent pleura and thoracic wall. On cut section, they are
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
PRIMARY SARCOMAS OF THE LUNG
265
Table 1 Histological subtypes of primary pulmonary sarcomas.
1. Parenchymal and endobronchial sarcomas
2. Sarcoma of large vessel origin 3. Sarcomas of small vessel origin
Fibrosarcoma Leiomyosarcoma Rhabdomyosarcoma Malignant fibrous histiocytoma Neurogenic sarcoma Chondrosarcoma Osteosarcoma Synovial sarcoma Liposarcoma Triton tumor Pulmonary artery sarcoma Angiosarcoma Hemangiopericytoma Epithelioid hemangioendothelioma Kaposi’s sarcoma
white to yellowish-tan and firm to soft and friable. Areas of necrosis and hemorrhage occur frequently, but cavitation is rare.2,13 Most PPSs are parenchymal, while a smaller proportion is endobronchial. Although hemoptysis may be a common presenting feature prompting initial diagnostic evaluation, PPSs do not have the exfoliative tendency that is present in primary pulmonary carcinomas. Sputum collection for cytology generally is of lower yield than even for primary pulmonary carcinomas, and a biopsy specimen is generally required for diagnosis.2,14 Soft tissue sarcomas frequently have a distinctive spindle cell morphology separating them morphologically from most
Figure 2 The lateral chest radiograph of the same patient as in Figure 1 confirms the presence of a mass in the left upper lobe of the lung.
Figure 3 A CT scan of the chest of the same patient as in Figure 1 show the peripheral location of the mass.
Figure 1 A routine posteroanterior chest radiograph in an asymptomatic patient detected this small mass in the left mid-lung field that was discovered to be a synovial sarcoma after biopsy.
carcinomas. Certain pulmonary carcinomas, such as spindle cell carcinomas and sarcomatoid mesotheliomas, may necessitate immunohistochemical evaluation for diagnostic distinction.15 PPSs are typically vimentin positive and keratin and epithelial membrane antigen (EMA) negative, in addition to showing features of particular lineage differentiation (smooth muscle, vascular, etc.). Spindle cell carcinomas, carcinosarcomas, and sarcomatoid mesotheliomas are generally cytokeratin and EMA positive in addition to showing vimentin positivity. Given that sarcomas most commonly originate elsewhere in the body and subsequently metastasize to the lung, once a diagnosis of pulmonary sarcoma is made, careful exclusion of a primary lesion outside of the lung is
266
THORACIC TUMORS
necessary before designating the sarcoma as PPS. This would include careful examination throughout the follow-up of a patient for subsequent detection of a previously inapparent primary tumor. As PPSs do not typically involve regional lymphatic structures, the staging of PPS does not follow the common tumornode-metastasis (TNM) staging system. Staging distinctions depend on the presence or absence of metastases as significant for prognosis. Histologic grade, including mitotic count, is the next most important prognostic factor (see section “Prognostic Factors”).3 A standardized system of assessing grade for these rare and varied tumors does not exist, rendering evaluation of tumor grade highly variable among individual institutions. Gal et al. proposed that a high-grade pulmonary sarcoma is typified by a mitotic count greater than 5 per 50 high-powered fields (HPF), the presence of giant cells, tumor necrosis, high cellularity, and poor tumor circumscription.16
Figure 4 Cytologic features of leiomyosarcoma.
Leiomyosarcoma Leiomyosarcoma is one of the more common PPSs17 and may occur anywhere within the respiratory tract, although they tend to be peripheral and subpleural. Some are more centrally located. The larger tumors are prone to extension into the chest wall, mediastinum, and diaphragm. They are thought to arise from smooth muscle in either the blood vessels or the bronchi, and are commonly adherent to these structures. Leiomyosarcomas may also arise in large vessels such as the pulmonary veins and arteries (see section “Sarcomas of the Pulmonary Arterial Tree”). In any location, the morphology is the same. The spindleshaped cells of leiomyosarcoma are arranged in prominent interlacing bundles and have long, blunt-ended darkly staining nuclei, with relatively abundant cytoplasm and indistinct borders (see Figure 4). Rarely, epithelioid polygonal cells are seen, with distinct cell borders and more abundant cytoplasm. Such tumors may be mistaken for carcinomas.18 – 21 Immunohistochemical studies show the presence of vimentin and desmin (see Table 2). Actin is expressed in at least 60% of cases. In one study of 18 PPSs, 75% of cases stained with smooth-muscle actin and only 5 of the 18 (31%) stained with desmin.12 Leu-7 (CD57) can be positive,22 but S-100 is observed only rarely.2,23 Electron microscopy reveals the following features: cytoplasmic thin actin filaments, dense bodies, subplasmalemmal
Figure 5 Electron microscopy of leiomyosarcoma.
dense plaques, pericellular basal lamina, and membraneassociated micropinocytosis (see Figure 5). The mitochondria tend to assume polarized positions. The epithelioid variant demonstrates villiform cytoplasmic projections and lacks true desmosomes.24
Fibrosarcoma Fibrosarcoma is another commonly reported PPS, arising from fibroblasts in the lung parenchyma. Many tumors previously diagnosed as fibrosarcomas may now be classified as
Table 2 Immunohistochemical staining of PPS.
FS LMS SS RMS NS AS KS MFH
VIM
EMA
Desmin
Actin
S-100
FVIIIR-Ag
CD31
Myo
Leu7
A1AT
CD68
CD34
+ + + + + + + +
− − + − − − − −
− + − + − − − −
− + − + − − − −
− +rare − +rare + − − −
− − − − − + + −
− − − − − + + −
− − − + − − − −
− + + − + − − −
+ − − − − − − +
− − − − − − − +
− −(rare) − − −(rare) + + −
VIM, vimentin; EMA, epithelial membrane antigen; FVIIIR-Ag, factor VIII-related antigen; Myo, myosin; A1AT, α-1-antitrypsin; FS, fibrosarcoma; LMS, leiomyosarcoma; SS, synovial sarcoma; RMS, rhabdomyosarcoma; NS, neurogenic sarcoma; AS, angiosarcoma; KS, Kaposi’s sarcoma; MFH, malignant fibrous histiocytoma.
PRIMARY SARCOMAS OF THE LUNG
malignant fibrous histiocytoma (MFH) or synovial sarcomas based on immunohistochemical or molecular features.12,25 Like leiomyosarcoma, the tumor is composed of malignant spindled cells, although they tend to have shorter, oval and vesicular nuclei, prominent nucleoli, scanty cytoplasm, and ill-defined cell outlines. Fibrosarcoma classically shows a pattern of intersecting fascicles, or a “herringbone” pattern. Collagen is abundant in well-differentiated fibrosarcomas (see Figure 6). Reticulin fibers typically encircle tumor cells.26,27 Immunohistochemistry evaluation reveals vimentin-reactive cells. Other immunostains are used to rule out tumors with a similar histologic appearance. Pure fibrosarcomas do not display actin, desmin, S-100, or Leu-7 positivity (see Table 2). Fibrohistiocytic markers such as α-1-antitrypsin and α-1-antichymotrypsin are not observed.2,28,29 Ultrastructurally, fibrosarcomas show fibroblastic or myofibroblastic differentiation. The nuclear outlines are irregular with prominent nucleoli and dilated rough endoplasmic reticulum. Intra- and extracellular collagen fibers are commonly seen (see Figure 7). When the tumor shows myofibroblastic differentiation, there are some
267
overlapping ultrastructural features with leiomyosarcoma, although membrane-associated pinocytosis and subplasmalemmal dense plaques are absent.30,31
Hemangiopericytoma The origin of hemangiopericytoma has been described in the literature as the capillary pericytic cell. However, ultrastructural analysis has not been definitive in showing pericytic differentiation in this tumor, and the tumor cells often show features of an undifferentiated or fibroblastic cell type.32,33 The tumor is composed of blunt spindle cells arranged in solid sheets or around vascular spaces resembling staghorns (see Figure 8). The cytoplasm is scanty, with the oval nucleus filling the cell. Both benign and malignant hemangiopericytomas have been described in the lung.25,34,35 Criteria for malignancy are based on degree of atypia, presence of necrosis, tumor size, and increased mitotic activity. In soft tissue, criteria include a size >5 cm and mitotic count of >4 per 10 hpf,33 while in one study of primary lung hemangiopericytomas a size >8 cm and mitotic count >3 per 10 hpf suggested a more aggressive clinical course.36,37 Immunohistochemical analysis reveals vimentin positivity in most and actin positivity in 50% of hemangiopericytomas. Hemangiopericytomas are characteristically CD34 positive, a helpful finding.23 Electron microscopy is not very helpful, although sparse cytoplasmic filaments, membraneassociated micropinocytosis, and pericellular basal lamina may be apparent.36,38 – 41
Solitary Fibrous Tumor
Figure 6 Histology of fibrosarcoma.
Figure 7 Electron micrograph of fibrosarcoma.
Solitary fibrous tumor (SFT) was first described as a pleural or subpleural tumor, but is now known to occur in any location.23,42 While many SFTs behave in a benign fashion, malignant SFTs are also described. Proposed criteria for malignancy are similar to those proposed for hemangiopericytoma,43 – 45 and interestingly, some histologic features of SFT overlap with hemangiopericytoma.23 SFT classically consists of spindled rather than round to fusiform cells seen in hemangiopericytoma, which are arranged in a “patternless pattern” with areas of hyalinization (see Figures 9,10). However, hemangiopericytoma-like areas may
Figure 8 Histologic section of hemangiopericytoma.
268
THORACIC TUMORS
Figure 9 Histologic section of solitary fibrous tumor, 200× magnification.
Figure 11 Histologic section of synovial sarcoma.
Figure 10 Histologic section of solitary fibrous tumor, 400× magnification. Figure 12 Histologic section of synovial sarcoma, 600× magnification.
be present, and immunohistochemistry for CD34 is usually positive, such that some have suggested that this lesion may be related to hemangiopericytoma.33 Electron microscopy reveals fibroblastic differentiation.23
Synovial Sarcoma Synovial sarcoma arising in the lung is a relatively newly recognized entity.46 – 48 The morphology of this tumor (which does not actually arise from the synovium, but from immature mesenchymal cells) is divided into four general subtypes: biphasic, monophasic epithelial, monophasic fibrous, and poorly differentiated.48 The monophasic subtype is most commonly diagnosed primary pulmonary synovial sarcoma. The gross features resemble those of other PPSs. In the monophasic fibrous type, the histology is characterized by atypical spindle cell proliferation composed of plump fusiform cells with oval nuclei and indistinct cytoplasm (see Figures 11, 12). Mast cells are frequently seen in the surrounding stroma.48,49 There is a moderate deposition
of collagen and calcification in some cases. Some tumors display epithelioid, hemangiopericytic, myxoid, or neural growth patterns.46,47 The histologic differential diagnosis can include carcinosarcomas, fibrosarcoma, malignant peripheral nerve sheath tumors (MPNST), and round blue cell tumors such as Ewing’s sarcoma and primitive neuroectodermal tumors (PNET). Immunohistochemical staining shows vimentin-positive cells, as in all soft tissue sarcomas. However, unlike most other sarcomas, EMA and keratin reactivity are common in synovial sarcoma, both in the epithelial component of biphasic synovial sarcomas, and in the spindle cell component in monophasic fibrous synovial sarcoma (see Table 2). In addition, up to 30% of synovial sarcomas are positive for S-100.50 These features can bring carcinomas, and tumors with neural differentiation into the differential diagnosis. BCL-2 and CD99 positivity is commonly seen in synovial sarcoma, a finding that can be suggestive of the diagnosis.25
PRIMARY SARCOMAS OF THE LUNG
Cytogenetic analysis is considered specific for the diagnosis of synovial sarcoma. A unique chromosomal translocation t(X;18) has been identified with its fusion gene SYTSSX, which has been used to confirm the diagnosis in a number of primary pulmonary synovial sarcomas.47,48,51,52 The SYT-SSX fusion gene is thought to function in transcription regulation, although the precise function has not yet been identified. Two different fusion genes (SYT-SSX1 and SYT-SSX2 ) have been recognized in pulmonary synovial sarcoma, with indications that the SSX1 fusion gene may indicate a more aggressive tumor type. Over 90% of synovial sarcomas show this translocation50 and the diagnosis can be made not only using conventional cytogenetic analysis but now can also be performed on paraffinembedded tissue by either fluorescence in situ hybridization (FISH) or reverse transcriptase-polymerase chain reaction (RT-PCR).53 – 55 Identification of this chromosomal translocation and fusion gene products have led to increasing diagnoses of pulmonary synovial sarcoma in recent years. Ultrastructural examination reveals a cytoplasm containing abundant scattered ribosomes, occasional dilated mitochondria, and rough endoplasmic reticulum. Desmosome-type cell junctions suggestive of epithelial differentiation are frequent.46 Eccentric nucleoli may be noted.49
Malignant Fibrous Histiocytoma MFH is the most common soft tissue sarcoma, but is rare in the lung. The first lung MFH was described in 1979.56,57 MFH is typified by a mixture of spindled and large pleomorphic fibroblastic and histiocytic cells arranged in a matlike or storiform pattern of growth (see Figure 13). Relatively abundant cytoplasm is present and may display a glassy homogeneous quality. The stroma is variably fibrous and may contain areas of myxoid change, acute inflammation, and osteoid-like material. In nonpulmonary sarcomas, the myxoid type portends a better prognosis.23 The diagnosis of MFH is usually a morphologic one. Immunohistochemistry and electron microscopy are used to rule out histologically similar tumors, such as sarcomatoid
Figure 13 Histologic section of malignant fibrous histiocytoma.
269
carcinomas, leiomyosarcomas, pleomorphic rhabdomyosarcoma. Vimentin positivity is universal (see Table 2). Fibrohistiocytic markers including α-1-antichymotrypsin, lysozyme, α-1-antitrypsin, CD68, and ferritin are also commonly positive.56,58 – 61 Ultrastructural features are not specific for MFH, but include evidence of fibroblastic, histiocytic, or primitive mesenchymal differentiation.62 Electron microscopy displays eccentric nuclei with prominent nucleoli, abundant cytoplasm, and villiform cell borders. Numerous lysosomes and prominent Golgi complexes are seen, in addition to sparse intermediate microfilaments. The tumoral matrix is composed of collagen fibers. True desmosomes are absent.2,60 In pulmonary MFH, 10% are endobronchial lesions, with half of these demonstrating vascular invasion.63
Rhabdomyosarcoma Rhabdomyosarcoma recapitulates the morphology of the embryonic myoblast. The etiology of this tumor, commonly thought to arise from striated muscle in an organ without striated muscle, is not known, although postulations regarding early mesenchymal or stem cell origins have been suggested.8 Rhabdomyosarcomas are separated into embryonal, alveolar, and pleomorphic subtypes. The embryonal and alveolar rhabdomyosarcomas usually occur in the pediatric population (<15 and 10–25 years of age, respectively23 ) and resemble other round blue cell tumors, while the pleomorphic subtype usually occurs in adults and resembles other pleomorphic sarcomas such as MFH. In the lung, all three subtypes have been described. The majority of lung rhabdomyosarcomas are of the pleomorphic type.8 Given the bimodal age distribution, it is possible that different etiologies exist for children compared with adults.64 Rhabdomyosarcoma can be endobronchial or intraparenchymal and tends to fill an entire lobe of the lung and invade pulmonary veins and bronchi. It may be associated with a congenital adenomatoid malformation.8 The histologic features vary with the subtype. Embryonal rhabdomyosarcoma consists of poorly differentiated round to spindled cells with variable numbers of more differentiated straplike cells that may have discernible cytoplasmic cross-striations, resembling embryonic myoblasts. Alveolar rhabdomyosarcoma shows similar cytologic features, but tends to show loosely cohesive groups or nests of round to oval cells. The centers of these nests show degenerative changes where the cells become detached, forming “alveolar” spaces. Pleomorphic rhabdomyosarcoma, on the other hand, is a high-grade sarcoma composed of large bizarre malignant cells that may be arranged in a fascicular or storiform patter, reminiscent of MFH or leiomyosarcoma.65 – 69 Evidence of muscle differentiation by immunohistochemistry and by electron microscopy is seen in all subtypes. Vimentin and desmin reactivity is usual, and actin positivity is found in at least 50% of cases.8 Myoglobin is observed in well-differentiated rhabdomyosarcomas (see Table 2). More recently, myoD1 and myogenin have been found to be specific and sensitive markers. Ultrastructural features include myofilaments displaying the Z disk of normal sarcomeres. In less differentiated cases,
270
THORACIC TUMORS
intermixtures of thin (actin) and thick (myosin) filaments may be identified, but there is a lack of distinct Z bands. Other characteristics include presence of glycogen and pericellular basal lamina.2,65
Angiosarcoma Only a handful of reports of primary pulmonary angiosarcoma exist.9 Angiosarcomas of the lung can assume a multifocal, bilateral, nodular appearance, and hence are most likely to be confused with metastases. They can present as a fulminant hemorrhagic syndrome70 and are characterized by extensive local invasion and hematogenous metastasis.9 Some authorities consider primary pulmonary angiosarcoma to be metastatic from an inapparent primary lesion.71 Two cases of angiosarcoma occurring along the pleural surface and mimicking mesothelioma have also been noted.72 On microscopic examination, anastomosing vascular channels are lined by malignant endothelial cells and form sievelike patterns (see Figure 14). The cells have scanty cytoplasm, and hobnail nucleoli often project into the vascular spaces. Foci of necrosis, hemorrhage, and hemosiderin deposits are common. Peripheral spread along vascular channels may simulate the appearance of lymphangitic carcinomatosis.64,71,73 The most helpful immunohistochemical stains are vascular markers, CD34, and CD31. CD31 is more sensitive and specific.23 Vimentin is usually present in all angiosarcomas, while factor VIII –related antigen is seen in 50% of cases (see Table 2). ulex europeus agglutinin-I (UEA) (lectin) and blood group isoagglutinins (BGI) are positive in 75% of cases, although carcinomas can sometimes exhibit reactivity; these immunostains are seldomly used. More recently, Fli-1, a nuclear transcription factor, has shown positivity in almost all vascular tumors.33,74 Ultrastructurally, angiosarcomas exhibit abundant microfilaments, Golgi complexes, and secondary phagolysosomes. Membrane-associated micropinocytosis and pericellular basal lamina are prominent. Intercellular junctions are well developed. Peculiar rodlike cytoplasmic inclusions (Weibel–Palade bodies) may be seen in a third of cases.2,71,73
Figure 14 Microscopic appearance of angiosarcoma.
Sarcomas of the Pulmonary Arterial Tree Sarcomas of the pulmonary arteries are distinct clinicopathologically from parenchymal or endobronchial sarcomas. They clinically resemble congestive heart failure or pulmonary embolism, but display filling defects of the pulmonary artery tree with or without a hilar mass. They can extend outside the arterial lumen into the surrounding lung or mediastinum. Emboli are commonly present in the lung parenchyma owing to distal embolization of tumor along the arterial tree. Proximal extension can cause right heart failure.2,13,71,75 – 77 Thoracic vascular lesions are divided into intimal and mural sarcomas. Intimal sarcomas are most common in the large arteries (aorta, pulmonary artery), probably arising from pleuripotential mesenchymal cells in the intima.78 Histologically, these are diverse high-grade sarcomas and include “differentiated” sarcomas such as fibrosarcomas, leiomyosarcomas, rhabdomyosarcomas, angiosarcomas, malignant fibrous histiocytomas, chondrosarcomas, and osteosarcomas, and “undifferentiated” sarcomas, which may represent very poorly differentiated vascular sarcomas and stain positively for vascular markers, CD31 and Fli-1.74 In one series of seven patients, leiomyosarcoma was the most frequent diagnosis in four of the patients examined.79 The majority of cases diagnosed, however, fall within an “undifferentiated” category. It is thought that subclassification beyond “primary pulmonary arterial sarcoma” is not worthwhile, as clinical behavior and outcomes for all subtypes are the same.56 Mural sarcomas occur most commonly in the pulmonary veins and inferior vena cava, and are usually leiomyosarcomas arising in the vascular smooth-muscle wall. Immunohistochemistry and ultrastructural morphology resemble those for each of the histological subtypes.71 Pulmonary arterial sarcomas tend to metastasize into the distal pulmonary vessels via tumor embolism, but may also metastasize systemically.80
Kaposi’s Sarcoma Kaposi’s sarcoma (KS) is a tumor of vascular origin that usually affects the lung in patients with disseminated tumor beginning in the skin. However, AIDS patients can present with pulmonary KS in the absence of cutaneous or other visceral involvement. Diffuse interstitial infiltrates can be seen without a discrete mass. However, abnormalities may be extremely focal.2,71,81 – 83 The gross appearance of pulmonary KS demonstrates flat or slightly raised plaques of red, violaceous, or red-blue coloration. These plaques may coalesce to form nodules, and may additionally be seen in the bronchial airways on transbronchial examination.84 Spindle tumor cells are found in concert with telangiectatic vessels, spreading along lymphatics and infiltrating vessel walls, airways, and pleural surfaces.71 Hemosiderin deposition may be prominent. The spindle cells trap extravasated erythrocytes. Occasional spindle cells contain cytoplasmic vacuoles or eosinophilic, hyaline globules, a characteristic finding. Immunohistochemistry demonstrates positivity for vascular markers CD31, CD34, and FVIII-related antigen, and BGI or receptors for UEA (lectins) in the angiomatoid form, while the spindle cell
PRIMARY SARCOMAS OF THE LUNG
271
component is reactive for vimentin, CD31, and CD34, but not FVIII (see Table 2). All tumors are human herpesvirus8 positive. Ultrastructurally, varying degrees of endothelial differentiation are seen.2,71
Epithelioid Hemangioendothelioma This tumor was initially termed IVBAT (intravascular bronchioalveolar tumor) as it was thought to demonstrate an aggressive variant of bronchioloalveolar cell carcinoma.71,85 – 88 It is more frequent in young women. It is considered a low-grade endothelial malignancy with clinical and radiologic features overlapping those of angiosarcoma of the lung. Peripheral multifocal masses are usually seen. The tumor is unique in that it forms polypoid masses of tumor cells. The cytoplasm of tumor cells is eosinophilic and may contain vacuoles, an attempt to form vascular channels. At the periphery of the tumor, the cells appear embedded in a collagenous to myxoid stroma, sometimes having chondroid quality, while the center of the tumor shows sclerosis and necrosis. The central portion may ossify or calcify (see Figure 15). Immunohistochemical and ultrastructural analysis resembles that of angiosarcoma, including positive staining for CD31, CD34, and FVIII-related antigen.50,85 – 88
Figure 16 Histologic section of malignant peripheral nerve sheath tumor, 200× magnification.
Malignant Peripheral Nerve Sheath Tumor (Neural Sarcoma) The term “malignant peripheral nerve sheath tumor” has replaced several previously used terms for neural malignancies, including neurogenic sarcoma and malignant schwannoma and denotes a malignant tumor arising from a peripheral nerve, or showing differentiation along the same path as nerve sheath tumors. Pulmonary MPNST is very rare, is encountered usually in the setting of neurofibromatosis, and arises in the posterior mediastinum.71,89 – 91 Benign lesions are more common than malignant tumors. MPNST is characterized by malignant spindle cells with indented nuclei and tapered cytoplasm displaying a fascicular or whorled pattern, with alternating hypocellular and hypercellular areas (“marblelike” pattern) (see Figures 16, 17). Histologic and immunohistochemical studies are not entirely specific and
Figure 17 Malignant peripheral nerve sheath tumor, s98.
electron microscopy may be necessary to confirm the diagnosis. S-100 immune staining is reactive in only half of the cases (see Table 2). Ultrastructural analysis demonstrates features of peripheral nerve sheath differentiation including mesoaxon formation and cytoplasmic microtubules.89 – 91
Other Primary Pulmonary Sarcomas
Figure 15 Histologic section of epithelioid hemangioendothelioma.
Very few cases of glomus tumors of pulmonary origin have been described. Glomus tumors are categorized into three basic subtypes: locally infiltrative glomus tumors, glomangiosarcoma arising within a glomus tumor, and de novo glomangiosarcoma.92 Glomus tumors may arise in the bronchus, mediastinum, or pulmonary parenchyma. They tend to be well circumscribed. Intratumoral hemorrhage is variably present. As opposed to the uniform epithelioid cells with distinct borders, scant clear to eosinophilic cytoplasm, and central, round nuclei seen in the glomus tumors, the described case of glomangiosarcoma demonstrated focal
272
THORACIC TUMORS
cystic degeneration and prominent zones of necrosis, a high mitotic count, and diffuse cytologic atypia.92 Tumors stained positively for vimentin, smooth-muscle actin, musclespecific actin (cytoplasmic staining), and collagen type IV (pericellular staining).92 Rare cases of chondrosarcoma, including myxoid chondrosarcoma and mesenchymal chondrosarcoma, osteosarcoma, liposarcoma (including myxoid and pleomorphic types), malignant mesenchymoma, and “triton” tumor (neurogenic tumor with rhabdomyoblastic differentiation) have been described in the literature.93 – 101 Histologic, immunohistochemical, and ultrastructural features resemble those of primary osseous or soft tissue tumors. In triton tumors, mature rhabdomyoblasts are scattered throughout sheets of schwannoma cells.102
In a patient with evidence of unilateral thromboembolism not responding to conventional therapy, the possibility of vascular involvement of malignancy must be considered. In angiosarcomas and primary pulmonary arterial sarcomas, the tumor origin may be intralumenal, mimicking thromboembolism on imaging studies.104 Pulmonary KS may be mistaken on imaging for pulmonary infections, particularly opportunistic infections with unusual imaging findings, in immunocompromised patients. The primary malignancy may indeed be intermingled with opportunistic infections in these patients, and the partial response of a treated infection must bring to mind the suspicion of a concurrent malignancy.84
CLINICAL PRESENTATION
Lesions Originating in the Lung
PPSs may affect any age group, although they are most commonly diagnosed in older adults. In children, PPSs are usually found on workup for unresolving pneumonia or dyspnea. This prompts imaging evaluation, which may reveal a large lung mass, or fiber-optic bronchoscopy, which may demonstrate an endobronchial lesion. Common presenting features in adults with PPS include dyspnea, cough, chest pain, or hemoptysis. Patients may experience fever or weight loss from postobstructive pneumonia, or less frequently from the primary malignancy. Symptomatic pulmonary thromboembolism may be evident, particularly with angiosarcomas, primary pulmonary artery sarcomas, or MFH. Alternately, patients may manifest with intrathoracic or pulmonary hemorrhage secondary to angiosarcoma or KS. More dramatic presentations with superior vena cava (SVC) syndrome, massive hemoptysis, cyanosis, and impending or acute respiratory failure have been documented. A case of hemangiopericytoma mimicking a Pancoast tumor has been published.39 Primary pulmonary artery sarcomas may produce symptomatic pulmonary hypertension or right heart failure as the initial presenting symptoms. Many patients may be asymptomatic, with abnormalities detected on routine imaging for other reasons. Clinical examination may demonstrate evidence of consolidation over the involved lung fields. This may be secondary to the mass effect of the tumor itself or due to a postobstructive pneumonia. Paraneoplastic syndromes may occur in rare patients with hemangiopericytoma; with hypoglycemia, hypertension, coagulopathy (both thrombotic and hemorrhagic diatheses), and pulmonary osteoarthropathy documented in the literature.103 Hypoglycemia may also occur secondary to significant glucose utilization of the primary tumor.27 Pulmonary osteoarthropathy has additionally been linked to cases of pulmonary angiosarcoma and pulmonary artery sarcoma.2,13,71,75 Sites of tumor metastasis may include the contralateral lung, liver, brain, bone, or soft tissues. Tumor detection in soft tissues, however, prompts the question as to whether this might represent a previously inapparent primary tumor, with metastasis to the lung.
Bronchogenic Carcinoma
DIFFERENTIAL DIAGNOSIS
Bronchogenic carcinomas are vastly more common as a pulmonary malignancy than sarcomas and manifest a great variety of histologic differences between different subtypes. Bronchogenic carcinomas of the spindle cell variant (sarcomatoid carcinoma) can resemble PPS. These are carcinomas composed entirely of spindled tumor cells. However, immunohistochemical staining of sarcomatoid carcinomas demonstrates the differences, as these malignancies are cytokeratin and EMA positive not usually demonstrated in PPS. To exclude synovial sarcoma, electron microscopy or cytogenetics may be necessary. Ultrastructurally, characteristics of carcinomas as opposed to sarcomas are demonstrated. See Tables 3 and 4. Table 3 Differential diagnosis of primary pulmonary sarcomas for the pathologist.
Malignant lesions 1. Bronchogenic carcinoma 2. Spindle cell (sarcomatoid) carcinoma 3. Small cell and atypical neuroendocrine tumors 4. Carcinosarcoma 5. Pulmonary blastoma 6. Melanoma 7. Lymphoma 8. Thymoma 9. Solitary fibrous tumor (SFT) of the lung 10. Metastatic sarcomatoid renal cell carcinoma 11. Sarcomatoid mesothelioma
Benign lesions 1. Inflammatory pseudotumor 2. Lymphangiomyomatosis 3. Leiomyoma 4. Chondroma
Table 4 Differential diagnosis of primary pulmonary sarcomas for the clinician.
Malignant lesions
Benign lesions
1. 2. 3. 4.
1. Thromboembolism 2. Opportunistic infection 3. Interstitial lung disease
Bronchogenic carcinoma Sarcoma metastatic to lung Melanoma Lymphoma
PRIMARY SARCOMAS OF THE LUNG
Small Cell Carcinoma Small cell carcinomas and atypical neuroendocrine tumors may also be confused with PPS on morphologic examination. Small cell carcinoma is composed of malignant smaller cells with high nuclear to cytoplasmic ratio. Thus, the differential diagnosis includes PPSs such as Ewing’s sarcoma, PNET, synovial sarcoma, and nonsarcomatous malignancies such as lymphoma. Immunohistochemical staining, however, demonstrates at least focal cytokeratin, although less than most nonsmall cell carcinomas, and markers of neuroendocrine differentiation. These include neuron-specific enolase, chromogranin, synaptophysin, and neural cell adhesion molecule (NCAM)/CD56. Small cell carcinoma of the lung is also positive for thyroid transcription factor (TTF-1) in up to 90% of cases.50 Carcinomas are vimentin negative, as opposed to sarcomas.2,13,71,75 Carcinosarcoma Carcinosarcoma comprises both epithelial and sarcomatous elements.1,13,35,71,75,105 – 107 Pulmonary carcinosarcomas are considered a variant of sarcomatoid carcinomas.108 Carcinosarcomas may contain elements of adenocarcinoma or squamous cell carcinoma in addition to leiomyosarcoma, fibrosarcoma, rhabdomyosarcoma, chondrosarcoma, or osteosarcomas. Undifferentiated sarcomas may also be present.108 Carcinosarcomas more commonly occur in men (a 7.25–9 : 1 ratio compared with women), versus the equal occurrence of PPS in women and men.108 Lymph nodes are the most common sites of metastasis (whereas this is unusual for PPS).108 Carcinosarcomas tend to carry a poor prognosis, with very aggressive behavior of both tumor cell populations, and 5-year survival reaching only 21.3%.104 – 108 Pulmonary Blastoma Pulmonary blastomas are rare tumors with bimodal distribution. A childhood form, pleuropulmonary blastoma, is considered possibly the same tumor as rhabdomyosarcoma of embryonal type, given the presence of embryonic and sarcomatous features in multidirectional differentiation.109 There is a proclivity for metastasis to the central nervous system. Pulmonary blastoma in adults is thought to be a different entity, although it resembles fetal lung tissue. It is composed of malignant epithelial and mesenchymal elements, similar to a carcinosarcoma.110 – 114 The fetal-type, well-differentiated adenocarcinoma is considered a monophasic purely epithelial variant with a better prognosis than the biphasic form. Overall survival is only 25% at 1 year.13,75 Solitary Fibrous Tumor In addition to the previously described overlapping features with hemangiopericytoma, SFT is often confused with fibrosarcoma. Fibroblasts constituting the SFT are unique in that they express CD34, a hematopoietic progenitor cell antigen.71,115,116 Lymphoma and Thymoma Lymphoma may histologically resemble sarcoma, especially embryonal or alveolar cell rhabdomyosarcoma, Ewing’s sarcoma, and PNET. However, staining for leukocyte common
273
antigen (LCA) and T and B cell–associated antigens, and electron microscopy displaying absence of sarcomeric differentiation assist in recognizing a lymphoma.2,13,71,75 Unusual cases of pulmonary thymomas resembling sarcoma have been reported.117 Benign Lesions
Benign lesions may occasionally imitate PPS. Inflammatory pseudotumors can resemble sarcomas, although the presence of abundant inflammatory cells, lack of cellular atypia, and, more recently, lack of p53 immunostaining have been employed in making the distinction.2,71,118 Lymphangioleiomyomatosis is a disorder typified by the proliferation of bland-appearing smooth muscle in the bronchial, vascular, and lymphatic structures. Reticulonodular shadows are apparent radiologically, and women of reproductive age are commonly affected.71 Other benign lesions of the lung that can be confused with sarcomas include leiomyomas and chondromas.2,71,119
Metastatic Disease Originating in Other Sites Metastatic Sarcoma
PPS is a diagnosis of exclusion. The most common etiology of pulmonary sarcoma is secondary to metastatic disease from another primary. Therefore, in any patient diagnosed with pulmonary sarcoma, a thorough clinical and radiographic evaluation to exclude any other source of primary malignancy must be carried out. This includes careful evaluation on follow-up exams, as a different primary site may manifest at a later date. Melanoma
Malignant pulmonary melanoma is almost always a metastatic lesion, and can mimic sarcoma histologically.2,13,71,75 Once again, immunohistochemistry (HMB-45 and S-100 positivity) and ultrastructural findings (premelanosomes) usually assist in distinguishing it from sarcoma. Of note, some poorly differentiated melanomas, particularly the spindled melanomas can lose immunoreactivity for melanocytic markers, and ultrastructural features may become less well defined.120 In those cases in which S-100 is the only positive marker, MPNST may be a diagnostic consideration. Sarcomatoid Renal Cell Carcinoma
The sarcomatoid variant of renal cell carcinoma when metastatic to the lung can simulate a PPS. However, a careful clinical evaluation and positive staining for epithelial antigens help to identify this entity.71
DIAGNOSTIC EVALUATION Imaging studies may demonstrate evidence of pulmonary infarction secondary to tumor invasion of vascular structures, or from thromboembolism. On imaging, tumors arising from the pulmonary vascular system tended to be located more centrally when compared with the more peripheral locations of pulmonary emboli, and more often caused complete
274
THORACIC TUMORS
occlusion of the vessel (as opposed to partial occlusion with pulmonary emboli) in one series.79 Sarcomas arising from the vascular system were more often unilateral as opposed to the more frequently bilateral pulmonary emboli, and tended to cause an expansion of the involved pulmonary artery, while most emboli did not.79 Of course, any extension beyond the lumen of the vessel into the parenchymal space would indicate malignancy instead of benign emboli. MRI or echocardiography may assist in the differentiation between a tumor of vascular origin and thromboembolism, but this requires that the diagnostician have a suspicion of a vascular tumor at the time of imaging workup.80 Other radiographic findings may include nodules of varying sizes suspicious for carcinomas, or, in the case of the tumors of vascular origin, may show patchy, diffuse infiltrates, some resembling interstitial lung disease. Postobstructive pneumonia may be seen. Calcification of the lesion is not typical. As tumors are commonly very advanced at the time of diagnosis, invasion into the surrounding structures (mediastinum, chest wall, etc.) is frequent.
Utility of FNA in Primary Pulmonary Sarcomas Adequate tumor biopsy frequently requires surgical intervention. However, in cases in which fine needle aspiration (FNA) might be clinically preferred, a substantial amount of information can be gleaned from aspiration biopsies, in some cases, yielding a definitive diagnosis.6,17,121 In general, fine needle aspirates of sarcomas yield cellular smears with atypia ranging from bland to obviously malignant. Certain cytologic patterns can suggest a diagnosis. For example, tumors composed of atypical spindled cells include synovial sarcoma, fibrosarcoma, leiomyosarcoma, and MPNST. A round cell morphology can be seen in synovial sarcoma, epithelioid leiomyosarcoma, Ewing’s sarcoma, PNET, embryonal or alveolar rhabdomyosarcoma, and round cell liposarcoma. Pleomorphic spindle tumors are classically MFH and pleomorphic rhabdomysoarcoma.29,121 From these differential diagnoses, immunohistochemistry can be performed in a judicious manner on either the cytologic smears or a cell block. Cases in which a definitive diagnosis is feasible usually consist of sufficient material to produce a cytologic cell block preparation, upon which immunohistochemical stains, FISH, or sometimes electron microscopy can be performed, just as would be performed with larger histologic samples. As tumors may contain areas of necrosis, adequate or larger tissue samples containing viable tissue for morphologic, immunohistochemical, electron microscopic, and possibly cytogenetic, evaluation is necessary. Mediastinoscopy is generally not helpful in diagnosis or staging, as metastasis to the mediastinal lymph nodes is rare. In pulmonary KS, diagnosis is typically based on the clinical appearance of lesions with airway inspection, as biopsy carries a high risk of hemorrhagic complications.84
of distant metastases. Among the majority of reported cases, surviving patients have undergone complete surgical resection of their tumors. Histologic grade is the next most important prognostic factor for PPS. Despite the lack of standardization for grading among institutions, attempts are made to classify tumors into categories of low-, intermediate-, or high-grade malignancies. Moran et al. proposed a classification system for primary pulmonary leiomyosarcomas, with low-grade tumors demonstrating low mitotic rate (<3 mitoses per 10 hpf), and absence of pleomorphism, hyperchromism, hemorrhage, or necrosis. Intermediate-grade tumors demonstrate increased mitotic rate (3–8 mitoses per 10 hpf), increased cellularity, and mild to moderate nuclear pleomorphism. High-grade tumors were those with increased cellularity, high mitotic rate (>8 mitoses per 10 hpf), and marked nuclear atypia and pleomorphism.20 Median survival time in this series of 18 patients differed, with the median survival time for eight patients with high-grade tumors of 5 months, while six patients with lower-grade tumors were alive and free of disease between 2 and 12 years after diagnosis.20 The nonspecific determinations of the degree of tumor cellularity and nuclear atypia and polymorphism obviously leave a large amount of the assignment of tumor grade to the individual pathologist. An endobronchial location of the primary lesion has been associated with better prognosis.4,5,122 This is likely secondary to a relatively earlier stage at the time of symptomatic presentation, although an association with lowergrade tumors in this location has been suggested.123 Size of the primary tumor has been identified as a prognostic factor, as well, with tumors greater than 5 centimeters (cm) demonstrating poorer prognosis when compared to smaller lesions.27 Different histologic types of PPS and age at diagnosis may carry varying prognostic implications. In leiomyosarcomas, younger age at presentation portends a poorer prognosis, whereas for fibrosarcomas, survival rates of 78% have been reported for childhood cases.7,122 Adults with leiomyosarcomas tend to have a better outcome than adults with other types of PPS.11 Pulmonary synovial sarcoma is thought to represent a more aggressive subtype, with overall 5-year survival of 50%.124 Primary pulmonary angiosarcomas and pulmonary artery sarcomas have been demonstrated to be highly aggressive tumors, with patients dying within a few months of diagnosis.9,125 In a series of nine patients with pulmonary arterial sarcomas seen at the Mayo Clinic, the longest survival time in seven evaluable patients was 3.5 years. The majority of patients died within 5 months of diagnosis.80 Epithelioid hemangioendotheliomas may exhibit a range of aggressiveness. Malignant triton tumors, in the few reported cases, seem to represent highly aggressive malignancies.
TREATMENT PROGNOSTIC FACTORS The most important prognostic factor for any patient with PPS is based on whether the tumor is amenable to complete surgical resection. This includes evaluation for the presence
Definitive therapy for PPSs involves complete surgical excision of tumor. This is most commonly accomplished with lobectomy or pneumonectomy, although reports of endobronchial resection for isolated endobronchial tumors exist
PRIMARY SARCOMAS OF THE LUNG
(with varying degrees of success).4,5 Frequently, PPSs are very locally advanced at the time of presentation and diagnosis, invading vital structures, and surgery is precluded. Experience with neoadjuvant chemotherapy for PPS is limited to a few case reports in the literature. Neoadjuvant combination chemotherapy was attempted in a patient with advanced pulmonary fibrosarcoma as well as in a patient with advanced pulmonary hemangiopericytoma in order to render resection easier; however, no response to therapy was documented on imaging or at the time of surgical resections.7,35 A report has been published of radical resection involving cardiopulmonary bypass and resection of the left atrium or pulmonary trunk, demonstrating that such approaches may be possible; however, follow up is short and overall outcome is not disclosed.126 Combination chemoradiotherapy was administered preoperatively to a patient with primary pulmonary artery leiomyosarcoma with secondary SVC syndrome. This patient demonstrated significant tumor response, with only microscopic residual disease demonstrated among tumor necrosis at pneumonectomy.127 Even for those patients who are able to undergo initial complete surgical resection, local relapse with subsequent unresectability remains a significant problem. Chemotherapy and radiation therapy have been administered alone or in combination in the adjuvant setting, however, no definitive evidence of benefit has been shown, given the short followup of all case reports. For patients with completely resected PPS, adjuvant therapy could be considered, in an approach similar to patients with completely resected extremity soft tissue sarcomas. A benefit from adjuvant chemotherapy could be postulated to be similar to patients with resected sarcomas elsewhere. No randomized trial exists to support the use of adjuvant chemotherapy for PPS, and given the rarity of the tumors, a randomized trial will not be possible. Follow-up of resected patients should include regular imaging in addition to history and physical examination. PPSs tend to recur locally more commonly than distantly, and early recognition of local recurrence may allow repeat resection for possible survival benefit. A proven standard of care for imaging follow-up does not exist, although the authors recommend chest radiograph with each clinic visit, as well as the consideration of an annual chest CT. Chemotherapy, either as single-agent or in combination regimens, may have palliative benefit in the setting of unresectable disease. Experience with chemotherapy for PPS is predominantly gathered from patients with traditional soft tissue sarcomas and is extrapolated to pulmonary sarcomas. The most commonly used agents include doxorubicin (20% response rate), ifosfamide, and dacarbazine.128,129 A case report of a child with recurrent unresectable bronchopulmonary leiomyosarcoma administered the combination of vincristine, dactinomycin, ifosfamide, and doxorubicin noted complete response with three cycles of therapy, documented on subsequent surgical exploration.130 The patient subsequently received a total of nine cycles of chemotherapy followed by involved-field radiation, with no evidence of relapse 16 months after disease recurrence.130 Immunotherapy with high-dose interleukin-2 was reported to induce tumor regression when used in combination with
275
radiotherapy for a patient with unresectable pulmonary angiosarcoma.9 It is unclear whether the combination, or one treatment modality over another, might be the cause of the tumor shrinkage. Interferon α-2B was administered to a patient with epithelioid hemangioendothelioma without response.88 In patients with AIDS and pulmonary KS, highly active antiretroviral therapy (HAART) has changed the outlook for this previously highly lethal disease (survival time less than 12 months) dramatically. In a retrospective series of patients with pulmonary KS in the pre- and post-HAART era, median survival was 8.9 weeks for patients before HAART, and was not reached in the post-HAART era –findings that were highly statistically significant.131 Patients previously receiving chemotherapy for pulmonary KS were able to obtain remission on HAART and discontinue chemotherapy.131 This represents the first major breakthrough in the treatment of a type of pulmonary sarcoma, dramatically improving survival. While many PPSs tend to be very aggressive if not completely surgically resected, reports of long-term survival even with recurrent or metastatic disease exist, including a patient surviving more than 7 years with local and metastatic disease who did not receive therapy beyond the initial diagnostic surgery.4 Partial spontaneous regression after long-term presence of disease has been described in a few patients with epithelioid hemangioendothelioma.88 Given the extreme rarity of all types of PPS, dedicated therapeutic trials for this group of diseases are not feasible. This is unfortunate, as (with the exception of HAART for KS) no effective therapy has been demonstrated for advanced disease. If possible, enrollment in clinical trials with newer agents (phase I or phase II studies) should be recommended for patients with advanced unresectable disease.
SUMMARY PPS is a rare cause of pulmonary malignancy. It consists of a diverse collection of tumor subtypes, with different origins and ranges of clinical behaviors. PPS tends to be very advanced at the time of diagnosis, and the prognosis as a group is poor. Complete surgical resection remains the only potentially curative therapeutic modality. The addition of chemotherapy and radiation therapy have not been demonstrated to be of benefit in the adjuvant setting, although both modalities are frequently employed in the case of incomplete resection margins. It is hoped that the dramatic increase in the understanding of malignancies on a molecular level that has occurred in recent years might lead to further progress in the treatment of PPS. Clinical trials evaluating newer agents and treatment approaches should be considered for patients with PPS.
REFERENCES 1. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55: 10 – 30. 2. Sonpavde G, Ulbright TM, Sandler A. Primary sarcomas of the lung. In Raghavan D, et al. (eds) Textbook of Uncommon Cancer, 2nd ed. Chichester, England: John Wiley and Sons Ltd, 1999: 511 – 521.
276
THORACIC TUMORS
3. Attanoos RL, Appelton MAC, Gibbs AR. Primary sarcomas of the lung: a clinicopathological and immunohistochemical study of 14 cases. Histopathology 1996; 29: 29 – 36. 4. Takeda F, et al. Leiomyosarcoma of the main bronchus in a girl: a long-time survivor with multiple lung metastases. Pediatr Pulmonol 2004; 37: 368 – 74. 5. Savas C, Candir O, Ozguner F. Acute respiratory distress due to fibrosarcoma of the carina in a child. Pediatr Pulmonol 2004; 38: 355 – 7. 6. Logrono R, et al. Diagnosis of primary fibrosarcoma of the lung by fine-needle aspiration and core biopsy: a case report and review of the literature. Arch Pathol Lab Med 1999; 123: 731 – 5. 7. Picard E, et al. Pulmonary fibrosarcoma in childhood: fiber-optic bronchoscopic diagnosis and review of the literature. Pediatr Pulmonol 1999; 27: 347 – 50. 8. Comin CE, et al. Primary pulmonary rhabdomyosarcoma: report of a case in an adult and review of the literature. Ultrastruct Pathol 2001; 25: 269 – 73. 9. Kojima K, et al. Successful treatment of primary pulmonary angiosarcoma. Chest 2003; 124: 2397 – 400. 10. Dail DH. Uncommon tumors. In Dail DH, Hammar SP (eds) Pulmonary Pathology. New York: Springer-Verlag, 1988: 847. 11. McCormack PM, Martini N. Primary sarcomas and lymphomas of the lung. In Martini N, Vogt-Moykopf I (eds) Thoracic Surgery: Frontiers and Uncommon Neoplasms, St. Louis, Missouri: CV Mosby, 1989, Vol. 5: 269. 12. Ono N, et al. Primary bronchopulmonary fibrosarcoma: report of a case. Surg Today 1998; 28: 1313 – 5. 13. Robinson PG, Shields TW. Uncommon primary malignant tumors of the lungs. Textbook Thorac Surg, 1994: 1320 – 1333. 14. Martini N, Hadju DI, Beattie EJ. Primary sarcoma of the lung. J Thorac Cardiovasc Surg 1971; 61: 33 – 8. 15. Lucas DR, et al. Sarcomatoid mesothelioma and its histological mimics: a comparative immunohistochemical study. Histopathology 2003; 42: 270 – 9. 16. Gal AA, et al. Prognostic factors in pulmonary fibrohistiocytic lesions. Cancer 1994; 73: 1817 – 24. 17. Ali SZ, et al. Solitary fibrous tumor. A cytologic-histologic study with clinical, radiologic, and immunohistochemical correlations. Cancer 1997; 81: 116 – 21. 18. Lillo-Gil R, Albrechtsson U, Jakobsson B. Pulmonary leiomyosarcoma appearing as a cyst: report of one case and review of the literature. Thorac Cardiovasc Surg 1985; 33: 250 – 2. 19. Ramanathan T. Primary leiomyosarcoma of the lung. Thorax 1974; 29: 482 – 9. 20. Moran CA. Primary leiomyosarcoma of the lung: a clinicopathologic and immunohistochemical study of 18 cases. Mod Pathol 1997; 10: 121 – 8. 21. Shaw RR, et al. Primary pulmonary leiomyosarcomas. J Thoracic Cardiovasc Surg 1961; 41: 430. 22. Ockner DM, et al. Genital angiomyofibroblastoma. Comparison with aggressive angiomyxoma and other myxoid neoplasms of skin and soft tissue. Am J Clin Pathol 1997; 107: 36 – 44. 23. Weiss SW, Goldblum JR. Enzinger and Weiss’s Soft Tissue Tumors, 4th ed. St. Louis, Missouri: CV Mosby, 2001. 24. Wick MR, et al. Primary pulmonary leiomyosarcomas: a light and electron microscopic study. Arch Pathol Lab Med 1982; 106: 510 – 4. 25. Suster S, Moran CA. Primary synovial sarcomas of the mediastinum: a clinicopathologic, immunohistochemical, and ultrastructural study of 15 cases. Am J Surg Pathol 2005; 29: 569 – 78. 26. Guccion JG, Rosen SH. Bronchopulmonary leiomyosarcoma and fibrosarcoma: a study of 32 cases and review of the literature. Cancer 1972; 30: 836 – 46. 27. Nascimento AG, Unni KK, Bernatz PE. Sarcomas of the lung. Mayo Clin Proc 1982; 57: 355 – 9. 28. Kindblom LG, Jacobsen GK, Jacobsen M. Immunohistochemical investigations of tumors of supposed fibroblastic-histiocytic origin. Hum Pathol 1982; 13: 834 – 40. 29. DeMay RM. The art & science of cytopathology. Aspiration Cytology, Vol. II. Chicago, Illinois: ASCP Press, 1996.
30. Stembridge VA, Luibel FJ, Ashworth CT. Soft tissue sarcomas: electron microscopic approach to histogenic classification. South Med J 1964; 57: 772 – 9. 31. Van Haelst UJGM. General considerations on electron microscopy of soft tissues. In Fenoglio CM, Wolff M (eds) Progress in Surgical Pathology. New York: Masson, 1980, Vol. 2: 225 – 257. 32. Erlandson RA, Woodruff JM. Role of electron microscopy in the evaluation of soft tissue neoplasms, with emphasis on spindle cell and pleomorphic tumors. Hum Pathol 1998; 29: 1372 – 81. 33. World Health Organization Classification of Tumours. Pathology & Genetics of Tumours of Soft Tissue and Bone. In Fletcher CDM, Unni KK, Mertens F (eds) Lyon, France: IARC Press, 2002. 34. Brega Massone PP, et al. A particular case with long-term follow-up of rare malignant hemangiopericytoma of the lung with metachronous diaphragmatic metastasis. Thorac Cardiovasc Surg 2002; 50: 178 – 80. 35. Wu Y.-C, et al. Primary pulmonary malignant hemangiopericytoma associated with coagulopathy. Ann Thorac Surg 1997; 64: 841 – 3. 36. Yousem SA, Hochholzer L. Primary pulmonary hemangiopericytoma. Cancer 1987; 59: 549 – 55. 37. Katz DS, et al. Primary malignant pulmonary hemangiopericytoma. Clin Imaging 1998; 22: 192 – 5. 38. Rusch VW, et al. Massive pulmonary hemangiopericytoma: an innovative approach to evaluation and treatment. Cancer 1989; 64: 1928 – 36. 39. Murad TM, von Haam E, Murthy MS. Ultrastructure of a hemangiopericytoma and a glomus tumor. Cancer 1968; 22: 1239 – 49. 40. Chong KM, Hennox SC, Sheppard MN. Primary hemangiopericytoma presenting as a pancoast tumor. Ann Thorac Surg 1993; 55: 9. 41. Mead JB, et al. Primary hemangiopericytoma of the lung. Thorax 1974; 29: 1 – 15. 42. Klemperer P, Rabin C. Primary neoplasm of the pleura: a report of five cases. Arch Pathol. 1931; 11: 385. 43. England DM, Hochholzer L, McCarthy MJ. Localized benign and malignant fibrous tumors of the pleura. A clinicopathologic review of 223 cases. Am J Surg Pathol 1989; 13: 640 – 58. 44. Vallat-Decouvelaere AV, Dry SM, Fletcher CD. Atypical and malignant solitary fibrous tumors in extrathoracic locations: evidence of their comparability to intra-thoracic tumors. Am J Surg Pathol 1998; 22: 1501 – 11. 45. Magdeleinat P, et al. Solitary fibrous tumors of the pleura: clinical characteristics, surgical treatment and outcome. Eur J Cardiothorac Surg 2002; 21: 1087 – 93. 46. Zeren H, et al. Primary pulmonary sarcomas with features of monophasic synovial sarcoma: a clinicopathological, immunohistochemical and ultrastructural study of 25 cases. Hum Pathol 1995; 26: 474 – 80. 47. Kaplan MA, et al. Primary pulmonary sarcoma with morphologic features of monophasic synovial sarcoma and chromosome translocation t(X;18). Am J Clin Pathol 1996; 105: 195 – 9. 48. Okamoto S, et al. Primary pulmonary synovial sarcoma: a clinicopathologic, immunohistochemical, and molecular study of 11 cases. Hum Pathol 2004; 35: 850 – 6. 49. Essary LR, Vargas SO, Fletcher CDM. Primary pleuropulmonary synovial sarcoma: reappraisal of a recently described anatomic subset. Cancer 2002; 94: 459 – 69. 50. World Health Organization Classification of Tumours. Pathology & Genetics of Tumours of the Lung, Pleura, Thymus and Heart. In Travis WD, et al. (eds) Lyon, France: IARC Press, 2004. 51. de Leeuw B, Geurts VKA. Molecular cloning of the synovial sarcoma specific translocation breakpoint. Hum Mol Genet 1994; 3: 745 – 9. 52. Duran-Mendicuti A, Costello P, Vargas SO. Primary synovial sarcoma of the chest: radiographic and clinicopathologic correlation. J Thorac Imaging 2003; 18: 87 – 93. 53. Guillou L, et al. Detection of the synovial sarcoma translocation t(X;18) (SYT;SSX) in paraffin-embedded tissues using reverse transcriptase-polymerase chain reaction: a reliable and powerful diagnostic tool for pathologists. A molecular analysis of 221 mesenchymal tumors fixed in different fixatives. Hum Pathol 2001; 32: 105 – 12. 54. Coindre JM, et al. Should molecular testing be required for diagnosing synovial sarcoma? A prospective study of 204 cases. Cancer 2003; 98: 2700 – 7.
PRIMARY SARCOMAS OF THE LUNG 55. Hill DA, et al. Real-time polymerase chain reaction as an aid for the detection of SYT-SSX1 and SYT-SSX2 transcripts in fresh and archival pediatric synovial sarcoma specimens: report of 25 cases from St Jude Children’s Research Hospital. Pediatr Dev Pathol 2003; 6: 24 – 34. 56. Bedrossian CWM, et al. Pulmonary malignant fibrous histiocytoma. Chester 1979; 75: 186 – 9. 57. Kimizuka G, Okuzawa K, Yarita T. Primary giant cell malignant fibrous histocytoma of the lung: a case report. Pathol Int 1999; 49: 342 – 6. 58. Yousem SA, Hochholzer L. Malignant fibrous histiocytoma of the lung. Cancer 1987; 60: 2532 – 41. 59. Kern WH, et al. Malignant fibrous histiocytoma of the lung. Cancer 1979; 44: 1793 – 801. 60. Lee JT, Shelburne JD, Linder J. Primary malignant fibrous histiocytoma of the lung: a clinicopathologic and ultrastructural study of 5 cases. Cancer 1984; 53: 1124 – 30. 61. McDonnel T, et al. Malignant fibrous histiocytoma of the lung. Cancer 1988; 61: 137 – 45. 62. Jabi M, Jeans D, Dardick I. Ultrastructural heterogeneity in malignant fibrous histiocytoma of soft tissue. Ultrastruct Pathol 1987; 11: 583 – 92. 63. Pui MH, Yu S-P, Chen J-D. Primary intrathoracic malignant fibrous histiocytoma and angiosarcoma. Australas Radiol 1999; 43: 3 – 6. 64. Spragg RG, et al. Angiosarcoma of the lung with fatal pulmonary hemorrhage. Am J Med 1983; 74: 1072 – 6. 65. Avignina A, et al. Pulmonary rhabdomyosarcoma with isolated small bowel metastases: a report of a case with immunohistological and ultrastructural studies. Cancer 1984; 53: 1948 – 51. 66. Lee SH, Reganchary SS, Paramesh J. Primary pulmonary rhabdomyosarcoma: a case report and review of the literature. Hum Pathol 1981; 12: 92 – 6. 67. Shariff S, et al. Primary pulmonary rhabdomyosarcoma in a child, with review of the literature. J Surg Oncol 1988; 38: 261 – 4. 68. Ueda K, et al. Rhabdomyosarcoma of lung arising in congenital cystic malformation. Cancer 1977; 40: 383 – 8. 69. Luck SR, Reynolds M, Raffensperger JG. Congenital bronchopulmonary malformations. Curr Probl Surg 1986; 23: 245 – 314. 70. Colby TV, Koss MN, Travis WD. Atlas of Tumor Pathology: Tumors of the Lower Respiratory Tract, 3rd ed series. Washington, District of Columbia: Armed Forces Institute of Pathology, Vol. Fascicle 13: 1995. 71. Suster S. Primary sarcomas of the lung. Semin Diagn Pathol 1995; 12: 140 – 57. 72. Falconieri G, et al. Pseudomesotheliomatous angiosarcoma: a pleuropulmonary lesion simulating malignant pleural mesothelioma. Histopathology 1997; 30: 419 – 24. 73. Ott RA, et al. Primary pulmonary angiosarcoma associated with multiple synchronous neoplasms. J Surg Oncol 1987; 35: 269 – 76. 74. Sebenik M, et al. Undifferentiated intimal sarcoma of large systemic blood vessels: report of 14 cases with immunohistochemical profile and review of the literature. Am J Surg Pathol 2005; 29: 1184 – 93. 75. Miller DL, Allen MS. Rare pulmonary neoplasms. Mayo Clin Proc 1993; 68: 492 – 8. 76. Nonomura A, et al. Primary pulmonary artery sarcoma: report of two autopsy cases studied by immunohistochemistry and electron microscopy and review of 110 cases reported in the literature. Acta Pathol Jpn 1988; 38: 883 – 96. 77. Gebauer C. The postoperative prognosis of primary pulmonary sarcomas: a review with a comparison between the histological forms and the other primary endothoracal sarcomas based on 474 cases. Scand J Thorac Cardiovasc Surg 1982; 16: 91 – 7. 78. Yi ES. Tumors of the pulmonary vasculature. Cardiol Clin 2004; 22: 431 – 40, vi – vii. 79. Yi CA, et al. Computed tomography in pulmonary artery sarcoma: distinguishing features from pulmonary embolic disease. J Comput Assist Tomogr 2004; 28: 34 – 9. 80. Parish JM, et al. Pulmonary artery sarcoma: clinical features. Chest 1996; 110: 1480 – 8. 81. Nash G, Fligel S. Kaposi’s sarcoma presenting as pulmonary disease in the acquired immunodeficiency syndrome: diagnosis by lung biopsy. Hum Pathol 1984; 15: 999 – 1001.
277
82. Misra DP, Sunderrajan EV, Hurst DJ. Kaposi’s sarcoma of the lung: radiography and pathology. Thorax 1982; 37: 155 – 6. 83. Epstein DM, Gefter WB, Conrad K. Lung disease in homosexual men. Radiology 1982; 143: 7 – 10. 84. Aboulafia DM. The epidemiologic, pathologic, and clinical features of AIDS-associated pulmonary Kaposi’s sarcoma. Chest 2000; 117: 1128 – 45. 85. Weiss SW, Enzinger FM. Epithelioid hemangioendothelioma: a vascular tumor often mistaken for a carcinoma. Cancer 1982; 50: 970 – 81. 86. Weiss SW, et al. Epithelioid hemangioendothelioma and related lesions. Semin Diagn Pathol 1986; 3: 259 – 87. 87. Dail DH, et al. Intravascular, Bronchiolar and Alveolar Tumor of the lung (IVBAT): an analysis of twenty cases of a peculiar sclerosing endothelial tumor. Cancer 1983; 51: 452 – 64. 88. Cronin P, Arenberg D. Pulmonary epithelioid hemangioendothelioma: an unusual case and a review of the literature. Chest 2004; 125: 789 – 92. 89. Bartley TD, Arean VM. Intrapulmonary neurogenic tumors. J Thorac Cardiovasc Surg 1965; 50: 114 – 23. 90. McCluggage WG, Bharucha H. Primary pulmonary tumors of nerve sheath origin. Histopathology 1995; 26: 247 – 54. 91. Roviaro G, et al. Primary pulmonary tumors of neurogenic origin. Thorax 1983; 38: 842 – 5. 92. Gaertner EM, et al. Pulmonary and mediastinal glomus tumors: report of five cases including a pulmonary glomangiosarcoma: a clinicopathologic study with literature review. Am J Surg Pathol 2000; 24: 1105 – 14. 93. Smith EAC, Cohen RV, Peale ARE. Primary chondrosarcoma of the lung. Ann Intern Med 1960; 53: 838. 94. Morgenroth A, et al. Primary chondrosarcoma of the left inferior lobar bronchus. Respiration 1989; 56: 241 – 4. 95. Morgan AD, Salama FD. Primary chondrosarcoma of the lung: case report and review of the literature. J Thorac Cardiovasc Surg 1972; 64: 460 – 6. 96. Loose JH, et al. Primary osteosarcoma of the lung: report of two cases and review of the literature. J Thorac Cardiovasc Surg 1990; 100: 867 – 73. 97. Sawamura K, et al. Primary liposarcoma of the lung: report of a case. J Surg Oncol 1982; 19: 243 – 6. 98. Kalus M, et al. Malignant mesenchymoma of the lung. Arch Pathol 1973; 95: 199 – 202. 99. Moran CA, Suster S, Koss MN. Primary malignant “triton” tumor of the lung. Histopathology 1997; 30: 140 – 4. 100. Huang HY, et al. Primary mesenchymal chondrosarcoma of the lung. Ann Thorac Surg 2002; 73: 1960 – 2. 101. Ichimura H, et al. Primary chondrosarcoma of the lung recognized as a long-standing solitary nodule prior to resection. Jpn J Thorac Cardiovasc Surg 2005; 53: 106 – 8. 102. Bose AK, Deodhar AP, Duncan AJ. Malignant triton tumor of the right vagus. Ann Thorac Surg 2002; 74: 1227 – 8. 103. Engelke C, et al. Pulmonary haemangiosarcoma with main pulmonary artery thrombosis imitating subacute pulmonary embolism with infarction. Br J Radiology 2004; 77: 623 – 5. 104. Ishida T, et al. Carcinosarcoma and spindle cell carcinoma of the lung: clinicopathologic and immunohistochemical studies. J Thorac Cardiovasc Surg 1990; 100: 844 – 52. 105. Cabarcos A, Gomez DM, Lobo BJL. Pulmonary carcinosarcoma: a case study and review of the literature. Br J Dis Chest 1985; 79: 83 – 94. 106. Meade P, et al. Carcinosarcoma of the lung with hypertrophic pulmonary osteoarthropathy. Ann Thorac Surg 1991; 51: 488 – 90. 107. Davis MP, et al. Carcinosarcoma of the lung: May Clinic experience and response to chemotherapy. Mayo Clin Proc 1984; 59: 598 – 603. 108. Koss MN, Hochholzer L, Frommelt RA. Carcinosarcomas of the lung: a clinicopathologic study of 66 patients. Am J Surg Pathol 1999; 23: 1514 – 26. 109. Perkikogianni Ch, Delides G, Kalmanti M. Pleuropulmonary blastoma: an aggressive intrathoracic neoplasm of childhood. Pediatr Hematol Oncol 2001; 18: 259 – 66. 110. Manivel JC, et al. Pleuropulmonary blastoma: the so-called blastoma of childhood. Cancer 1988; 62: 1516 – 26.
278
THORACIC TUMORS
111. Francis D, Jacobsen M. Pulmonary blastoma. Curr Top Pathol 1983; 73: 265 – 94. 112. Kodama T, et al. Six cases of well differentiated adenocarcinoma simulating fetal lung tubules in pseudoglandular stage: comparison with pulmonary blastoma. Am J Surg Pathol 1984; 8: 735 – 44. 113. Koss MN, Hochholzer L, O’Leary T. Pulmonary blastomas. Cancer 1991; 67: 2368 – 81. 114. Spencer H. Pulmonary blastoma. J Pathol 1961; 82: 161. 115. Yousem SA, Glynn SD. Intrapulmonary localized fibrous tumor. Am J Clin Pathol 1988; 89: 365 – 9. 116. van de Rijn M, Lombard CM, Rouse RV. Expression of CD34 by solitary fibrous tumor of the pleura, mediastinum and lung. Am J Surg Pathol 1994; 18: 814 – 20. 117. Fukayama M, et al. Pulmonary and pleural thymoma: diagnostic application of lymphocytic markers to the thymoma of unusual site. Am J Clin Pathol 1988; 89: 617 – 21. 118. Ledet SC, Brown RW, Cagle PT. P53 immunostaining in the differentiation of inflammatory pseudotumor from sarcoma involving the lung. Mod Pathol 1995; 8: 282 – 6. 119. Aakhus T, Mylius EA. Leiomyoma of the lung. Acta Chir Scand 1962; 124: 372 – 6. 120. Winnepenninckx V, et al. New phenotypical and ultrastructural findings in spindle cell (desmoplastic/neurotropic) melanoma. Appl Immunohistochem Mol Morphol 2003; 11: 319 – 25. 121. McGee RS Jr, Ward WG, Kilpatrick SE. Malignant peripheral nerve sheath tumor: a fine-needle aspiration biopsy study. Diagn Cytopathol 1997; 17: 298 – 305. 122. Lai D-S, et al. Primary bronchopulmonary leiomyosarcoma of the left main bronchus in a child presenting with wheezing and atelectasis of the left lung. Pediatr Pulmonol 2002; 33: 318 – 21.
123. Pettinato G, et al. Primary bronchopulmonary fibrosarcoma of childhood and adolescence: reassessment of a low-grade malignancy. Hum Pathol 1989; 20: 463 – 71. 124. Dennison S, Weppler E, Giacoppe G. Primary pulmonary synovial sarcoma: a case report and review of current diagnostic and therapeutic standards. Oncologist 2004; 9: 339 – 42. 125. Choong CK, et al. Failure of medical therapy for pulmonary “thromboembolic” disease: beware the unsuspected primary sarcoma of the pulmonary artery. J Thor Cardiovasc Surg 2004; 128: 763 – 5. 126. Shimono T, et al. Pulmonary leiomyosarcoma extending into left atrium or pulmonary trunk: complete resection with cardiopulmonary bypass. J Thorac Cardiovasc Surg 1998; 115: 460 – 1. 127. Faul JL, et al. Superior vena cava syndrome caused by pulmonary artery sarcoma. J Thorac Cardiovasc Surg 1999; 118: 749 – 50. 128. Edmonson JH, et al. Randomized comparison of doxorubicin alone versus ifosfamide plus doxorubicin or mitomycin, doxorubicin and cisplatin against advanced soft tissue sarcomas. J Clin Oncol 1993; 11: 1269 – 75. 129. Santoro A, et al. Doxorubicin versus CYVADIC versus doxorubicin plus ifosfamide in first line treatment of advanced soft tissue sarcomas: a randomized study of the EORTC and Bone Sarcoma Group. J Clin Oncol 1995; 13: 1537 – 45. 130. Ferrari A, et al. Response to chemotherapy in a child with primary bronchopulmonary leiomyosarcoma. Med Pediatr Oncol 2002; 39: 55 – 7. 131. Holkova B, et al. Effect of highly active antiretroviral therapy on survival in patients with AIDS-associated pulmonary Kaposi’s sarcoma treated with chemotherapy. J Clin Oncol 2001; 19: 3848 – 51.
Section 5 : Thoracic Tumors
23
Mesotheliomas Giuseppe Giaccone and Paul Baas
ETIOLOGY AND EPIDEMIOLOGY Asbestos is the major cause of malignant mesothelioma (MM) localized to the pleura, peritoneum, and tunica vaginalis testis. After the identification of asbestos as a major cause of pleural mesothelioma before 1960, an increased interest in MM led to a better registration.1 The incidence of MM in the United States and western Europe has shown a steep increase in the past decades and a modification of the ratio of incidence between the genders.2,3 In western Europe, asbestos use remained high until 1980, and men born in the period 1945–1950 suffered the highest risk, and of these 1 in 150 will die of mesothelioma.3 In Europe, about 9000 deaths due to MM are expected by the year 2018, because of the very long latency period, and then a decline.3 In a recent update of the European epidemic, however, it would appear that the leveling off is already taking place in the present decade.4 As shown in Figure 1, the incidence in the United States has leveled off after a peak in 2000–2004, as a result of an earlier ban on asbestos, and it would seem to have already started decreasing, according to another report.5 Over the past 40 years, the male : female ratio has changed from 1 : 1 to 3 : 1, as would be expected, considering the occupational relationship of exposure to asbestos.2,3 According to Hillerdal, the incidence of MM in autopsy series varies from 0.02 to 0.7% and is closely related to the level of asbestos exposure.6 However, there is no evidence of a threshold level below which there is no risk of developing mesothelioma.7 The highest incidences are observed in areas with shipyards, and involve construction workers, asbestos miners, manufacturers of heating equipment, and insulation workers.6 A major concern is the widespread use of asbestos in developing countries. Now that the western world has stringent regulations for the handling and manufacturing of asbestos products, companies have found their way to countries like India and Pakistan. Unrestricted and unprotected use of asbestos is unfortunately often common, and people are not informed about the risks of asbestos exposure.8 The male population born after 1953, however, shows a dramatic decrease in incidence in the western industrialized countries because of government restrictions on the use and handling of asbestos materials in the mid-1960s. In general,
two-thirds of patients with MM are between 50 and 70 years of age. The partners of men employed in the asbestos industry also have an increased risk of developing MM. Most case reports consider the cleaning of clothes contaminated with asbestos fibers as a major source of asbestos exposure.9 The tumor occurs rarely in children and in those cases it is mostly localized to the peritoneal cavity. Finally, it must be emphasized that because of the occurrence at an older age and the difficulty of diagnosis, the real incidence of this disease is probably being underestimated. Approximately 70% of patients presenting with MM have a documented history of asbestos exposure. The mechanism of induction of MM and the risk of developing this disease after exposure to different types of asbestos fibers have been the subject of a number of studies. Different asbestos fibers are known, with different physical properties and carcinogenicity. The length : diameter ratio is considered to be of importance in evading the neutralizing action of the immune system and leading to penetration of the fibers in the pleura. The needlelike configuration combined with the longevity and the presence of iron in the fiber leads to chronic irritation of phagocytic cells. Fibers with the highest ratio are considered to be the most carcinogenic. Active oxygen species and reactive metabolites of oxygen produced by phagocytic cells are probably also involved in the chronic irritation and cytotoxicity of the asbestos fibers.10 Serpentines, also known as white asbestos (chrysotile), and anthophyllite are curlier and can be broken down to some extent by the immune system and removed by the lymphatic system. The carcinogenicity is considered to be relatively low for the serpentines, of which chrysotile is the best-known fiber. The amphiboles have the highest malignant potential and have a high length : diameter ratio.11 Crocidolite (blue asbestos), amosite (brown asbestos), and tremolite also contain more iron than the serpentines, and this probably increases their malignancy potential. In Table 1, a summary of the risk and some of the features of the fibers are presented.
Nonasbestos-Related Mesothelioma Although in 70–87% of cases of MM direct or indirect exposure to asbestos can be found, only 10% of those
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
280
THORACIC TUMORS 2500
900 000
800 000 2000
700 000 Males
1500 500 000
400 000 1000
Number of cases
Consumption (metric tons)
Consumption 600 000
300 000
200 000
500
Females 100 000
0 1916
0 1926
1936
1946
1956
Consumption (metric tons)
1966
1976
Male 2000
1986
1996
2006
Females 2000
2016
2026
Males 1992
2036
2046
Females 1992
Figure 1 Asbestos use (consumption) in the United States and projected numbers of male and female mesothelioma cases based on a birth cohort and age model estimated from Surveillance, Epidemiology, and End Results (SEER) Program data for two periods, 1973 – 1992 and 1973 – 2000. (Reproduced from reference 2 by permission of Oxford University Press).
Table 1 Types of asbestos fibers and estimated risk of carcinogenicity.
Fiber type
Color
Exposure site
Specifications ratio length/diameter
Carcinogenicity
Crocidolite
Blue asbestos
Amosite
Brown asbestos
Erionite Tremolite Chrysotile
– – White asbestos
Mining (South Africa, Australia) Factory workers Shipyards Mining and milling Insulators Factory workers As contaminant in chrysotile Environmental (Greece) Mining and milling Insulators Factory workers Miners (Finland)
High – – High – – High – Low – – Thick and coarse
High – – High – – High Possibly high Low – – Remote
Anthophyllite
–
with asbestos exposure develop mesothelioma, and therefore other etiologies have been suggested. Volcanic minerals like zeolite have been associated with an increased incidence of MM. Erionite, a respirable form of zeolite, was found in the soil and rocks of Anatolian villages and was identified as the causative agent, since asbestos fibers were absent, and erionite was proved to be carcinogenic in mice and rats.12 Of special interest are the man-made mineral fibers, of which fiberglass is the best known. These fibers have been developed to replace asbestos fibers and are used for insulation and construction.
Familial aspects have been the subject of some reports: in a study of 196 MM patients and 511 deceased controls, there was a twofold elevation in the risk of mesothelioma among men exposed to asbestos when reporting cancer in two or more first-degree relatives. However, no correlation among women and nonasbestos-exposed men was observed. This may suggest that a family history of cancer may enhance susceptibility to MM given asbestos exposure.13 A more recent study reported 40 cases of familial pleural mesothelioma, all of whom had been exposed to asbestos.14 This report raises the question whether environmental factors
MESOTHELIOMAS
that family members share may act as cofactors in asbestosrelated mesothelioma. The viral nature of MM has been the object of several controversial reports. SV40 induces mesotheliomas in hamsters, and 60% of human mesotheliomas contain and express SV40 sequences, as published by Carbone et al.15 Furthermore, SV40 large T antigen (Tag) retains its ability to bind and inactivate p53, which would then be an important way to inactivate p53 in the absence of mutations in MM.16 SV40 Tag targets and inactivates growth suppressive proteins, such as those of the Rb family and p53. SV40 is able to initiate the transformation of mesothelial cells into malignant cells by blocking tumor suppressor proteins such as p53 and the products of the retinoblastoma-susceptibility gene. These transformations can occur easily in rodent cells, but the conversion of human cells is more complex. SV40 is not specific to the development of MM. It has been shown that different tumor types such as lymphoblastomas, sarcomas, ependymomas, and osteosarcomas can develop after infection with the SV40 virus. Some studies in humans have identified the presence of SV40 large T antigen in mesothelioma samples but others have observed a total absence. Contamination by plasmids during the sequencing or the selection of patients might explain some of these differences. Large-scale prospective studies will be required to answer this issue.17 – 20
HISTOPATHOLOGY Histologic material should be obtained through thoracoscopic biopsies, pleurectomy, or pneumonectomy in the case of pleural mesothelioma, and by peritoneoscopy or preferably open directed biopsy in the case of peritoneal mesothelioma.12 Possibly a part of the sample should be fixed in glutaraldehyde for electron microscopy. Diffuse MM is a biphasic tumor of the mesothelium, and three major variants can be distinguished by optic microscopy: epithelial (approximately 50% of cases), sarcomatoid (16%), and mixed (34%).6 The frequency of the diagnosis of the mixed type is dependent on the number of samples available, as biphasic features are often absent in a single sample. The major difficulty in the diagnosis of MM is the differential diagnosis with a pleural diffusion of a tumor originating elsewhere, usually an adenocarcinoma of various origin. However, there may be difficulties in also determining whether a pleural proliferation is benign or malignant. Several criteria have been developed to separate these entities,21 and recently, the determination of telomerase expression has been shown to be only present in MM and not in benign proliferations.22 The presence of a biphasic aspect of predominant sarcomatoid features may make the diagnosis rather straightforward; however, immunohistochemistry is often necessary, and sometimes also electron microscopy, to differentiate epithelial mesothelioma from a metastasis of an adenocarcinoma. Periodic acid-Shiff-diastase (PAS)-diastase positive intracellular vacuoles are usually absent in mesotheliomas but present in adenocarcinomas; mucicarmine staining is also usually negative in mesotheliomas and positive in adenocarcinomas. On the contrary, Alcian blue disappearance after digestion with hyaluronidase is characteristic of mesothelioma.
281
Immunohistochemistry may help in rendering evident the biphasic nature of diffuse mesothelioma, and can be used to differentiate it not only from adenocarcinoma but also from sarcomas. Carcinoembryonic antigen (CEA) is usually weak or absent in mesothelioma, but strongly expressed in several carcinomas. The calcium-binding protein calretinin was shown to be expressed in MM but not in metastatic adenocarcinomas;23 however, this marker was exquisitely expressed in the epithelial type but was not expressed in the sarcomatoid type and the mixed type.24 The pattern of calretinin staining in pleural effusions is also differential between adenocarcinoma and mesothelioma.25 The immunostaining with surfactant protein B and thyroid transcription factor 1 antibodies has been shown to be able to differentiate adenocarcinoma of the lung, often positive, from mesothelioma, always negative.26,27 Other epithelial markers and cytokeratins can also distinguish adenocarcinomas from epithelial mesothelioma.12,28,29 Of possible help in differentiating MM from nonsmall cell lung cancer is the expression of Wilms’ tumor 1 susceptibility (WT1 ) gene products: detection of WT1 expression (mRNA or protein) is frequent in MM but absent in nonsmall cell lung cancer.29,30 Furthermore, immunoreactivity has been shown to be particularly strong in the sarcomatous type.31 N-cadherin staining has also been shown in most pleural MM.32 Electron microscopy can be considered the reference method in doubtful cases;33 epithelial mesothelioma is characterized by numerous long, slender, branching surface microvilli, desmosomes, abundant tonofilaments, and intracellular lumen formation on polygonal cells; in the sarcomatous type, elongated nuclei and abundant rough endoplasmic reticulum are observed.
BIOLOGY A number of recent studies have focused on the multiple genetic changes that take place in MM. A detailed and thorough review of the cytogenetic abnormalities described in mesothelioma has been published recently.34 Frequent genetic losses involving specific regions of chromosome arms 1p, 3p, 6q, and 9p, and numerical losses of chromosome 22 have been described in MM. These chromosomal changes suggest the presence of multiple important genes involved in the molecular development of this disease. Deletion of chromosome bands 9p13-p22 has been described cytogenetically in 50% of MMs;35 in this region, two putative tumor suppressor genes, p15 and p16, are located. Alterations of p16 were found to be common in MM cell lines (homozygous deletions in 85% of 40 cell lines), but less in primary tumors (22% of 23 specimens).36 Deletions of 9p21-p22 outside of the p16 locus may reflect the involvement of other putative tumor suppressor genes. Codeletion of p15 and p16 was observed in 72% of 50 cases by FISH analysis, including all the sarcomatoid cases (21 of 21).37 This finding strongly suggests that p15 and p16 or other neighboring genes on chromosome 9p are the targets for the development of this tumor. Frequent losses of 6q have been observed by cytogenetic analysis (about 40%). By using 32 microsatellite markers,
282
THORACIC TUMORS
loss of heterozygosity (LOH) was observed in 61% of 46 MMs. These deletions fall into four distinct locations, possibly loci of putative tumor suppressor genes.38 Of the 28 cases with LOH of 6q, 23 (82%) also had LOH of 1p; therefore deletions of regions on both chromosomes occur frequently. The neurofibromatosis type II (NF2 ) gene, which is located on chromosome 22, often lost in MM, was found to be mutated in 53% of 15 MM cell lines.39 Inactivation of NF2 in MM probably occurs via a two-hit mechanism.40 LOH in 3p14 to 3p25 was also found in a total of 42% of MMs (cell lines and tumors).41 The Fhit gene, located at 3p14.2, was found to be absent or reduced in 54% of 13 mesothelioma cases, which suggests a potential significance of Fhit inactivation in the pathogenesis of this tumor.42 Recent data using comparative genomic hybridization identified a region on 15q, which is deleted in 54% of mesotheliomas; this finding suggests that the region probably harbors a putative tumor suppressor gene that may contribute to the pathogenesis of MM.43 Data suggest the great importance of angiogenesis in mesothelioma. Interleukin 8 (IL-8), a potent kemokine with angiogenesis function, has been shown to be an autocrine growth factor for mesothelioma cell lines.44 A higher expression of vascular endothelial growth factor (VEGF) and higher microvessel density (MVD) were found in epithelial type mesothelioma,45 supporting an important role in angiogenesis and lymphangiogenesis in mesothelioma.45 – 47 Angiogenesis has also been shown to be a poor prognostic factor in this disease.48 With the advent of new technologies, also in MM attempts have been made to discover novel candidate oncogenes and tumor suppressor genes using high-density oligo nucleotide microarrays. These genes should be further validated in functional assays.49 Differential patterns of gene expression may also aid in diagnosis, as different patterns have been observed between sarcomatoid and epithelial tumors.50
MALIGNANT PLEURAL MESOTHELIOMA Clinical Presentation Malignant pleural mesothelioma tends to remain confined to one hemithorax, and is mainly characterized by locoregional growth and spread. However, at autopsy as many as 70% of cases have detectable tumor invasion in thoracic lymph nodes, and distant metastases are observed in liver, lungs, kidney, adrenals, and bones in approximately 50% of cases.51 It is generally believed that the initial growth rate of MM is slow and that therefore symptoms appear at a late stage. Sometimes, a pleuritic effusion is noted, which initially subsides but, after a few years, results in the development of mesothelioma. The tendency of the mesothelioma to first grow along the pleural lining and finally invade adjacent structures like muscles, ribs, and diaphragm makes early diagnosis difficult. The most frequently reported complaints are pain, shortness of breath, and cough; unexplained fever is also encountered in a number of patients. In the majority of cases, there is a pleural effusion that can vary in volume, and a number of patients present with a large exudate, which causes compression of the lung and displacement of the
mediastinum. The fluid is often a bloody exudate in which a provisional diagnosis of MM can be made.12,52 A more typical feature of the tumor in advanced cases is the retraction of the involved chest side, which occurs after obliteration of the pleural space and encasement of the lung. Growth into the ribs and muscles will lead to localized swelling and pain. The invasion of the mediastinum can result in dysphagia, superior vena cava syndrome, and pericardial effusion. When cardiac involvement is present, arrhythmias, nonspecific ST-T changes, conducting abnormalities, or atrial fibrillation can ensue.53 Paraneoplastic symptoms are not frequently reported except for thrombocytosis, which has been observed in 40–90% of patients, depending on the stage of the disease.54 Other paraneoplastic symptoms, such as diffuse intravascular coagulation, autoimmune hemolytic anemia, syndrome of inappropriate antidiuretic hormone (ADH) secretion, hypoglycemia, and hypercalcemia, have also been observed.
Evaluation and Staging A wide variety of radiographic methods are used in MM to determine the extent of the disease. In some cases, the combination of the history and chest X rays can lead to a presumptive diagnosis of MM. The diagnosis, however, can only be made with certainty on a histologic specimen. The chest X ray usually shows a unilateral effusion and a pleural thickening that is often diffuse, circumferential in most cases, and presenting with nodules or masses. In the course of the disease, the tumor grows into the fissures of the lung and encases the lung, leading to the typical retracted hemithorax.51,52 An additional clue to the diagnosis of MM is the finding of calcified pleural plaques, which are often seen in patients exposed to asbestos. It is, however, not believed that the tumor arises from these plaques. Invasion of the mediastinum and diaphragm can be identified by computed tomography (CT) scan or magnetic resonance imaging (MRI).55 – 57 MRI has the advantage of generating sagittal images that are informative of the growth of the tumor in the sinuses. Also, invasion of the heart, pericardium, and diaphragm can be discriminated better by MRI than by CT scan. Furthermore, the diaphragm can also be visualized by laparoscopic examination.58 The use of ultrasound is of help in defining pleural effusions and possible involvement of the heart. Relatively new among the diagnostic imaging modalities is the fluorodeoxyglucose (FDG) positron-emission tomography (PET) scan, which can be used to differentiate between malignant and benign (fibrotic) tissue, and which can be a valuable addition to CT and MRI in detecting disease spreading outside the chest, especially in patients to be submitted to surgery.59 PET scans can also help assess whether the tumor responds to chemotherapeutic regimens.55 Furthermore, recent data indicate that, as for other malignant diseases, in pleural mesothelioma also the standardized uptake value (SUV) can be a strong prognostic factor. Patients with a high SUV have a threefold increase in death risk.60,61 In order to get tissue for histologic diagnosis, the investigation of choice is a thoracoscopic examination, by which excessive fluid can be drained, followed by pleurodesis;
MESOTHELIOMAS
hence, staging is facilitated. The staging of malignant pleural mesothelioma is difficult. The first staging system was developed by Butchart et al.62 Six other staging systems have been proposed, of which the latest one, by the International Mesothelioma Interest Group (IMIG) (Table 2)63 based on a TNM modification, is considered the best framework for analyzing prospective clinical trials. Optimal staging is very important for selection of patients who have localized disease and could be candidates for surgery or combined modality treatments. Also, a more reliable comparison of data obtained from international studies can be made.
Management Median survival of patients with MM treated with supportive care alone is approximately 7 months.65 Attitudes toward the treatment of this disease vary greatly, and range from supportive treatment only to aggressive surgery and combined modality treatment. Surgery
Local control in malignant pleural mesothelioma is important as the tumor tends to remain localized to the hemithorax for a large part of the course of the disease. The role of surgery is debated, in the presence of different staging systems, which did not provide very accurate estimates of prognosis. However, in general, patients with stage I disease, according to the Butchart classification should be considered candidates for radical surgery; these patients have a tumor confined within the capsule of the parietal pleura, involving only the ipsilateral lung, pericardium, and diaphragm. Unfortunately, the preoperative staging procedures are not precise and can only stage the disease rather poorly. There are two major types of operation that have been employed in patients with malignant pleural mesothelioma: pleurectomy and extrapleural pneumonectomy. Pleurectomy consists of stripping the pleura from the apex of the lung to the diaphragm, along with the pericardium, if necessary. This operation generally requires a thoracotomy. The results of the most representative series published are summarized in Table 3. Operative mortality is only 1–2%, but complications include bronchopleural fistulas, hemorrhage, and subcutaneous emphysema. The value of pleural decortication as a palliative measure in the case of recurrent effusion has not been well established and might be taken into consideration in case pleurodesis repeatedly fails. Extrapleural pneumonectomy is the en bloc removal of the parietal pleura, lung, pericardium, and hemidiaphragm. Diaphragmatic resection is followed by reconstruction to prevent herniation. In the hands of experienced thoracic surgeons, the operative mortality of this complex procedure is nowadays 5–9%, but serious complications are seen in 25% of patients or more, and include bronchopleural fistulas and empyema, vocal cord paralysis, chylothorax, arrhythmia, and respiratory insufficiency. Extrapleural pneumonectomy is a rather complex operation, which should be performed by skilled surgeons and in select centers, with interest in the treatment of this disease. The results of the most representative series are summarized in Table 4. The majority
283
Table 2 New international staging developed by the International Mesothelioma Interest Group64 .
T = Tumor T1 T1a Tumor limited to the ipsilateral parietal including mediastinal and diaphragmatic pleura No involvement of the visceral pleura T1b Tumor involving the ipsilateral parietal including mediastinal and diaphragmatic pleura Scattered foci of tumors also involving the visceral pleura T2 Tumor involving each of the ipsilateral pleural surfaces (parietal, mediastinal, diaphragmatic, and visceral pleura) with at least one of the following features: • Involvement of diaphragmatic muscle • Confluent visceral pleural tumor (including the fissures), or extension of tumor from visceral pleura into the underlying pulmonary parenchyma T3 Describes locally advanced but potentially resectable tumor Tumor involving all of the ipsilateral pleural surfaces (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 tumor extending into the soft tissue of the chest wall • Nontransmural involvement of the pericardium T4 Describes locally advanced technically unresectable tumor Tumor involving all of the ipsilateral pleural surfaces (parietal, mediastinal, diaphragmatic, and visceral pleura) with at least one of the following features: • Diffuse extension or multifocal masses of tumor in the chest wall, with or without associated rib destruction • Direct transdiaphragmatic extension of tumor to the peritoneum • Direct extension of tumor to the contralateral pleura • Direct extension of the tumor to one or more mediastinal organs • Direct extension of tumor into the spine • Tumor extending through the internal surface of the pericardium with or without a pericardial effusion; or tumor involving the myocardium N = Lymph node NX Regional lymph nodes cannot be assessed N1 No regional lymph node metastases N2 Metastases in the ipsilateral bronchopulmonary or hilar lymph nodes N3 Metastases in the contralateral mediastinal, contralateral internal thoracic, ipsilateral, or contralateral supraclavicular lymph nodes M = Metastases MX Presence of distant metastases cannot be assessed M0 No distant metastases M1 Distant metastases present Stage I Ia T1a N0 M0 Ib T1b N0 M0 Stage II T2 N0 M0 Stage III Any T3 N0 Any N1 M0 Any N2 M0 Stage IV Any T4 Any N3 Any M1
of the most recent series report a median survival above a year. Of course, in the absence of randomized trials one should view these results with caution, as they could simply be the effect of patient selection. This operation alters the natural history of malignant pleural mesothelioma, as in
284
THORACIC TUMORS
Table 3 Results of pleurectomy.
References 66
McCormack et al. Law et al.67 Faber et al.68 DaValle et al.69 Wanebo et al.70
Achatzy et al.71 Ruffie et al.65 Brancatisano et al.72 Rusch et al.73 Branscheid et al.74 Lee et al.75
Patients
2-year survival (%)
Median survival (months)
95a 28 35 23 17 epithelial 16 sarcomatous 46 63 45 51 82 32
35 32 12 NA NA NA 11 NA 21 40 25 32
12.6 20 10 11.2 21 11 10 9.8 16 18.3 315 days 18
NA = not available. a With implant or external irradiation.
Table 4 Results of extrapleural pneumonectomy.
References 77
Worn et al. Bamler and Maassen78 Butchart et al.62 DeLaria et al.79 Branscheid et al.74 Faber68 Geroulanos et al.80 Rush and Venkatraman76 Sugarbaker et al.81 Ruffie et al.65
Patients
2-year survival (%)
Median survival (months)
Operative mortality (%)
62 17 29 11 76 33 18 50 183 23
37 35 10 NA 10 27 NA 20 38 17
19 NA 4 15 284 days 13.5 20 9.9 19 9.3
20 – 25 23 31 0 11.8 9 7 6 3.8 14
NA, not available.
fact distant metastases are seen more frequently after this operation than after decortication or no surgical treatment.76 Results of surgery are difficult to interpret because of different patient selection, the relatively small number of patients, the lack of randomized trials, and often the addition of another treatment modality to surgery. Radiotherapy
The role of radiotherapy in the management of malignant pleural mesothelioma is still rather unsettled; however, radiotherapy alone has probably no major role in disease control and survival. The most recent series do not indicate that irradiation improves survival in comparison to best supportive care. However, radiotherapy, given after extrapleural pneumonectomy, has been used extensively as adjuvant treatment in several series in order to reduce the local recurrence rate.81 The treatment volume is a crucial aspect of radiation of malignant pleural mesothelioma, and treatment of the entire pleura is indicated. This is extremely difficult to achieve because tumoricidal doses cannot be delivered without causing serious side effects in normal surrounding tissues, such as the lung, heart, and liver. Of course, these issues are less complex after a pneumonectomy, where higher doses can be given.82,83 Several different techniques have been developed in order to spare normal tissue from the irradiated fields,
including the use of intraoperative radiotherapy.75,84 Radical irradiation has delivered 40–50 Gy to the entire pleural space and the mediastinum, followed by boost irradiation up to 55–71 Gy to areas of gross disease; however, until now no satisfactory technique has been developed that allows highdose radiation without major risks for the adjacent normal tissues. Radiotherapy is often used for palliation of pain, and it has often been added to surgery in an attempt to improve local control and reduce local failures. The results of the published literature are, however, rather complex to interpret, because radiotherapy was used as part of a multimodality treatment in locally advanced cases of pleural mesothelioma, because of the small number of patients reported in single studies and because of the lack of randomized trials. A randomized trial of radiotherapy proved that radiotherapy to the thoracoscopy entry tract significantly reduced the incidence of local relapse: in this study, 0 of 20 patients who received 21 Gy delivered in three fractions 10–15 days after thoracoscopy had local recurrence, whereas local recurrence developed in 8 of 20 (40%) patients who did not receive radiation.85 Interestingly, a more recent randomized study, utilizing only a single 10 Gy dose, was ineffective in preventing local recurrences.86 Chemotherapy and Novel Therapeutics
A large number of chemotherapeutic agents have been investigated in MM. The level of activity of several of these agents is rather low, and variable from study to study; given the rarity of the disease, phase II studies have usually been performed in a small number of patients, especially in the past. Extensive reviews of chemotherapy in MM have been published recently.87 – 89 Because phase II trials have as a major objective the assessment of response rate, measurability of the disease becomes an important issue; in malignant pleural mesothelioma, even by CT scan it is sometimes difficult to assess exactly the disease extension and therefore identify changes during treatment. Disease assessment by chest X ray is unreliable, and this is the major reason for overestimation of treatment results in old studies. The introduction of the RECIST criteria in the assessment of response in solid tumors has raised some questions about the applicability of this system to mesothelioma, for which the maximum diameter is usually the pleural base of implant, which does not easily decrease even when a treatment is active.90,91 Among single-agent studies, the anthracyclines doxorubicin and epidoxorubicin, cisplatin, and high-dose methotrexate appear to have some consistent degree of efficacy, which is in the range of 10–20% major response rate (Table 5). Although doxorubicin and cisplatin have been considered in the past to be the most active agents in this disease, conflicting reports on their activity have been published. Most antimetabolites have also been reported to have some degree of activity, and the new multitarget antifolate pemetrexed has shown promising activity in a phase II study. The drug is well tolerated when combined with folate and vitamin B12 supplementation, which significantly reduce the side effects of the drug, without impairing the antitumor activity. The
MESOTHELIOMAS Table 5 Systemic single-agent therapy in series with ≥15 patients in first-line setting.
Patients
Responses
Response rate (%)
Topoisomerase inhibitors Doxorubicin Liposomal doxorubicin Detorubicin Pirarubicin Epirubicin Mitoxantrone Menogaril Etoposide Amsacrine Irinotecan
66 55 35 35 59 62 22 111 19 25
7 2 9 8 8 3 1 6 1 0
11 4 26 22 14 5 3 3 5 0
100,101 102,103 104 105 106,107 108,109 110 111,112 113 114
Alkylating agents Cyclophosphamide Ifosfamide Mitomycin Cisplatin Carboplatin ZD0473a Temozolomide
16 133 19 59 88 43 27
0 9 4 8 10 0 1
0 7 21 14 11 0 4
101 115 – 118 119 120,121 122 – 124 125 126
Antimetabolites High-dose methotrexate Trimetrexate Edatrexate Di-deazofolic acid 5-Fluorouracil Capecitabine 5-DHAC Gemcitabine Pemetrexed
60 51 58 18 20 27 56 44 64
22 6 9 1 1 1 7 2 9
37 12 16 6 5 4 13 5 14
127 128 129 130 131 132 133,134 64,93 92
Tubulin interfering Vincristine Vinblastine Vindesine Vinorelbine Docetaxel Paclitaxel
23 20 38 29 48 50
0 0 1 7 4 3
0 0 3 24 8 6
135 136 137,138 139 140,141 142,143
Others Aziridinylbenzoquinone Acivicin IL-2 Ranpirnaseb Thalidomidec Gefitinib Imatinib
20 19 29 81 40 42 25
0 0 2 4 0 2 0
0 0 8 6 0 4 0
Drug
References
144 145 146 147 148 149 150
DHAC, 5-Dihydroazacytidine. a Second-line treatment, a platinum analog. b 37% had prior chemotherapy. c 50% had prior chemotherapy.
response rate as a single agent was 16%.92 Gemcitabine, an antimetabolite with lots of similarities to pemetrexed, has negligible activity as a single agent,64,93 but reproducible activity in combination with cisplatin.94 – 99 The use of novel targeted therapies has so far not been successful. Both gefitinib, a tyrosine kinase inhibitor (TKI) of epidermal growth factor receptor (EGFR), and imatinib, a TKI of c-KIT, were ineffective in single-agent studies in chemotherapy na¨ıve patients.149,150 These disappointing results are likely due to the lack of patient selection. It is now known that mesotheliomas do not have the EGFR mutations
285
that confer sensitivity in nonsmall cell lung cancer,151 and c-KIT is poorly expressed in MM.152,153 Interestingly, angiogenesis inhibitors may have some promise, as VEGF and VEGFR are frequently overexpressed in mesothelioma and VEGF appears to be a growth factor for this malignancy.46,47 Clinical studies with targeted agents against VEGF or its receptors are ongoing in pleural mesothelioma. Combination chemotherapy has been extensively tested in first-line therapy. It results in higher responses and apparently also survival compared to single agents in nonrandomized comparisons. Most common combinations tested contain doxorubicin and cisplatin.12 A literature review of studies published between 1965 and 2001 was recently reported.154 The combination of cisplatin and doxorubicin had the highest response rate, and cisplatin was the most active single agent. Also based on these data, cisplatin has been recently used as a comparator in two large randomized phase III studies.155,156 In the past, the rarity of the disease and the lack of drugs with substantial activity prevented large studies. It is only recently that large randomized studies have been performed in this disease. These studies have combined cisplatin with the antifolates pemetrexed or raltitrexed compared to cisplatin alone. The results of these studies indicate that a clinically significant survival benefit can be achieved for the combination therapy with better responses and improved quality of life (Table 6). Both studies confirm an increase in median survival of almost 3 months. Pemetrexed in combination with cisplatin has recently been approved for the treatment of advanced pleural mesothelioma. These data have now led to a renewed interest in the chemotherapeutic treatment of patients with advanced mesothelioma and to the implementation of more effective chemotherapy in multimodality regimens and in the neoadjuvant setting. Furthermore, some studies started to investigate chemotherapy given after failure on first-line chemotherapy, either as single agent125 or even combinations,157 though with poor results so far. The percentage of patients who received second-line chemotherapy varied between 37 and 47% in the pemetrexed randomized study.158 Combined Modality Treatment
Given the disappointing results of surgery alone, combined modalities have been attempted in order to reduce local recurrence and systemic spread. Various modalities have been employed in different institutions, and sometimes it is difficult to distil the exact treatment modality employed in the different patient series, in the absence of properly conducted controlled prospective trials and randomized trials. Remarkably, despite the relatively small effect (response rate) of single-agent chemotherapy and combination chemotherapy, often multiple agents have been used in the adjuvant setting. Surgery was employed in the form of pleurectomy or extrapleural pneumonectomy in combination with various forms of radiation treatment and chemotherapy. The experience of the Memorial Sloan–Kettering Cancer Center with pleurectomy, intraoperative brachytherapy, and postoperative irradiation was reported in 1984, with a relatively short follow-up, and has not been recently updated.161 Rusch et al. reported a study of surgical resection, with
286
THORACIC TUMORS
Table 6 Results of randomized trials in malignant pleural mesothelioma.
Regimen
Patients
Doxorubicin + cyclophosphamide Doxorubicin + cyclophosphamide+ dacarbazine (DTIC) Doxorubicin Cyclophosphamide Cisplatin Cisplatin + pemetrexed
36 40 16 16 222 225
Cisplatin Cisplatin + raltitrexed
124 126
Carboplatin c.i. Cisplatin + etoposide
13 12
Response rate (%)
Median survival (weeks)
1-year survival (%)
11 13 0 0 16.7 41.3 P < 0.001 14 24 P = 0.06 58 39
30 25 NR NR 9.3 months 12.1 months P = 0.02 8.8 months 11.4 months P = 0.048 21.7 18.7 p = 0.0135
NR NR NR NR 38 50.3
References 159 101 155
39.6 46.2
156
33 0
160
NR, not reported.
intrapleural and systemic chemotherapy.162 The authors concluded that this cannot be considered routine treatment. Three other smaller studies with a similar setup confirmed the limited indication for this approach and a high morbidity rate.163 – 165 In a recent update of the Memorial experience,166 it was reported that a total of 231 thoracotomies were performed between 1983 and 1998 for the treatment of pleural mesothelioma. In 115 cases, extrapleural pneumonectomy was performed and in 59 only pleurectomy/decortication. A total of 142 patients received adjuvant therapy. By multivariate analysis stage, histology, gender, and adjuvant therapy had impact on survival, but not type of surgical resection. The Dana Farber Cancer Center developed a large experience with extrapleural pneumonectomy combined with multiagent chemotherapy and postoperative radiation.167 The operative mortality after 328 consecutive extrapleural pneumonectomies (performed from 1980 to 2003) was 4% with 60.4% minor and major morbidity rate.167 The most common complications were atrial fibrillation (44.2%), prolonged intubation (7.9%), vocal cord paralysis (6.7%), deep vein thrombosis (6.4%), and technical complications (6.1%).167 A total of 183 patients were treated between 1980 and 1997 by en bloc extrapleural pneumonectomy, followed 4–6 weeks later by four to six cycles of doxorubicin-cyclophosphamide (+ cisplatin after 1985), followed by ipsilateral hemithorax and mediastinum radiotherapy.81 The median survival was 19 months with two- and five-year survival rates of 38 and 15%, respectively.81 Epithelial cell type had better survival and nodal involvement was a negative prognostic factor. The conclusion of this group is that trimodality treatment is safe and effective for selected patients with malignant pleural mesothelioma without nodal involvement. From the same group, Baldini et al. also analyzed the pattern of failure of 46 evaluable patients treated with trimodality therapy:168 25 (54%) had recurrences; sites of recurrence were 35% local, abdominal in 26%, contralateral in 17%, and distant in 8%. Median time to first recurrence was 19 months. This and other studies clearly indicate that more effective strategies should be sought to increase local control. It is also well recognized that preoperative CT and MRI do not accurately stage patients preoperatively,169 and that now that these extensive
operations are being performed in more centers, better staging procedures should be developed. In a study performed at NCI, 36 patients received 25 mg m−2 cisplatin per week four times, interferon-α, and tamoxifen, for one to five cycles; 10 additional patients had debulking surgery followed by two cycles. Only 19% partial responses were observed, and median survival was 8.7 months.170 Nineteen potentially resectable pleural mesothelioma patients received cisplatin-gemcitabine as neoadjuvant treatment.171 Extrapleural pneumonectomy was successfully performed in 16 patients, and 13 received postoperative radiotherapy. Median survival was 23 months. Larger studies of neoadjuvant chemotherapy are ongoing at several institutions. Intracavitary Therapy
Because mesothelioma of the pleural cavity tends to remain localized for a long time, local treatment with cytotoxic agents or other new approaches may be interesting. Different approaches have been tested, such as local instillation of interferon-γ ,172,173 IL-2,174 tumor necrosis factor (TNF)α,175 and the use of radiocolloids, such as 198 Au or 32 P with limited success. In a more recent study using IL-2 in 31 patients with pleural mesothelioma, the response rate was 22% and median survival 15 months, with only mild toxicity.176 Another approach to the problem of killing residual disease after surgery is the use of perioperative photodynamic therapy. The photosensitizer is administered a few days before the operation and is retained to some extent in the tumor tissue and vasculature. This treatment modality has been used in a few small studies.177 A study of intraoperative photodynamic therapy after pleuropneumonectomy has been performed in 28 patients.178 Considerable toxicity was encountered, including three perioperative deaths, associated with local control in 50% of cases. The same group of investigators attempted intraoperative hyperthermic intrathoracic chemotherapy with cisplatin and doxorubicin heated at 40–41 ◦ C after tumor debulking.179 Although no postoperative deaths occurred, significant morbidity was observed in 65% of patients, and median survival was only 11 months.
MESOTHELIOMAS
A small feasibility study in 10 patients demonstrated that pleural space perfusion with cisplatin was feasible and safe after operation.180 A liposome-entrapped platinum compound was given to 33 patients with pleural mesothelioma and freeflowing pleural effusions.181 After at least two cycles, a pathological complete response, ascertained by repeat biopsies, was obtained in 42% of cases, and median survival was 11 months. Although feasible, this approach demonstrated substantial morbidity and the instillation of the drug was unable to penetrate all tumor deposits. The use of suicidal genes in the pleural cavity has been the subject of phase I clinical trials.182,183 Given the lack of activity and the substantial problems with gene therapy studies, no further studies have been performed in recent years in MM. Prognostic Factors
A number of prognostic factors have been identified in large series of mainly advanced MM. Staging (in particular, nodal status) appears to be important in early onset of the disease.76,81 Thrombocytosis has also been mentioned in surgical series as an important prognostic factor, together with histological subtype, the sarcomatous subtype having the worst prognosis.81 In a database of 204 patients enrolled into sequential phase II studies of the European Organisation for Research and Treatment of Cancer (EORTC), poor prognostic factors by multivariate analysis were a poor performance status, high white blood count, male gender, sarcomatous histology, and having a probable/possible histological diagnosis of MM.184 Similarly, the Cancer and Leukemia Group B (CALGB) have analyzed their database of 337 patients treated in clinical trials in the group and proposed another prognostic categorization.185 A subsequent retrospective analysis of an independent cohort of patients confirmed the prognostic value of both the EORTC and the CALGB scoring systems.186,187 These identified risk groups might help in stratifying patients in future studies. A number of biological factors have been investigated in MM. Among these, angiogenesis48 has been shown to confer poor survival in retrospective series. High expression of p27 and low proliferation were associated with prolonged survival in a small study.188 The advent of microarray technology has been used to investigate prognostic profiles and possibly identify targets for treatment.49,50,189 cDNA expression profiling identified a 27-gene classifier that was highly predictive of survival in 21 patients who underwent cytoreductive surgery.190 Another study was able to confirm the prognostic signature obtained in 39 cases, in another cohort of 52 tumors.191 Of course, these studies will require larger sample sizes and prospective validation.
287
proper staging exists for malignant peritoneal mesothelioma, which tends to be limited to the abdominal cavity at diagnosis, sometimes forming a palpable mass in the omentum. Noninvasive radiological findings are not very helpful; CT scan of the abdomen usually reveals only peritoneal thickenings and ascites. Distant metastases or intra-abdominal metastases to abdominal organs are unusual. Definite diagnosis can only be performed by tissue examination and only very rarely by cytology. Also, in malignant peritoneal mesothelioma, paraneoplastic syndromes are common, such as thrombocytosis, phlebitis, and disseminated intravascular coagulation. Because the tumor remains localized to the peritoneal cavity for most of the course of the disease, local therapy has been investigated. Although surgical resection may provide palliation, especially in the case of bowel obstruction, it can rarely be complete, and probably does not impact on survival. Several groups have reported small series of patients treated with aggressive surgery, followed by intraperitoneal chemotherapy in relatively small series. A study of 18 patients underwent tumor debulking followed by intraperitoneal hyperthermic chemotherapy with cisplatin;194 the treatment was well tolerated, and 80% were alive at 2 years. A similar approach, with different chemotherapy was attempted in 19 patients;195 this study showed feasibility and no excessive toxicity, with a 3-year survival of 69%. The Dana Farber group recently reviewed their experience in treating peritoneal mesothelioma.196 In this retrospective report covering two decades, 69 patients were analyzed and median survival was an impressive 67 months. Women did better than men. More advanced cases have been treated with systemic chemotherapy. The combination of cisplatin-irinotecan yielded a response rate of 24% in 17 patients treated.197 As part of an extended access program for pemetrexed in the United States, 73 assessable patients with peritoneal mesothelioma were treated with cisplatin-pemetrexed or pemetrexed alone.198 The response rate was 19% to pemetrexed alone and 29% to the combination.
MALIGNANT MESOTHELIOMA OF THE TUNICA VAGINALIS TESTIS There are fewer than 100 cases reported in the literature;199 patients present with a hydrocele or hernia, and sometimes diffuse involvement of abdominal lymph nodes is present at diagnosis. Initial aggressive surgery and adjuvant treatment appear to improve survival.
MALIGNANT PERITONEAL MESOTHELIOMA
MALIGNANT MESOTHELIOMA OF THE PERICARDIUM
Malignant peritoneal mesotheliomas represent one-third to one-fifth of all MM.192 Symptoms of malignant peritoneal mesothelioma are usually the signs of advanced disease, and include abdominal pain, ascites, weight loss, and the presence of a palpable abdominal mass. In a large series of MM, of 180 patients, 37 (21%) had peritoneal localization.193 No
There are fewer than 200 cases reported.200 This tumor is usually not diagnosed prior to surgery or autopsy. Patients generally present with pericardial effusion, congestive heart failure, a mass, or tamponade, which often require urgent surgery.
288
THORACIC TUMORS
BENIGN FORMS OF MESOTHELIOMA Benign fibrous tumors of the pleura are also called localized fibrous mesothelioma; they are one-third as common as the diffuse pleural mesotheliomas.201 Histological studies can provide the differential diagnosis with MM (see above). These tumors are usually pedunculated and can often be radically resected; however, local relapses have been described, even after many years.202 Well-differentiated papillary mesotheliomas or cystic mesotheliomas of the peritoneum have been described in young women and are associated with long survival and rare progression to a typical MM.203 Benign fibrous mesotheliomas of the genital tract and of the Atrio-Ventricular (AV) node have also been described; their histogenesis is not clearly related to the mesothelium.
REFERENCES 1. Working Group on Asbestos and Cancer. Report and recommendations of the working group convened under the auspices of the geographical pathology committee of the international union against cancer. Arch Environ Health 1965; 11: 221 – 9. 2. Price B, Ware A. Mesothelioma trends in the United States: an update based on surveillance, epidemiology, and end results program data for 1973 through 2003. Am J Epidemiol 2004; 159: 107 – 12. 3. Peto J, et al. The European mesothelioma epidemic. Br J Cancer 1999; 79: 666 – 72. 4. Pelucchi C, et al. The mesothelioma epidemic in Western Europe: an update. Br J Cancer 2004; 90: 1022 – 4. 5. Weill H, Hughes J, Churg A. Changing trends in US mesothelioma incidence. Occup Environ Med 2005; 62: 270. 6. Hillerdal G. Malignant mesothelioma 1982: review of 4710 published cases. Br J Dis Chest 1983; 77: 321 – 43. 7. Hillerdal G. Mesothelioma: cases associated with non-occupational and low dose exposures. Occup Environ Med 1999; 56: 505 – 13. 8. LaDou J. The asbestos cancer epidemic. Environ Health Perspect 2004; 112: 285 – 90. 9. Smith AH, Wright CC. Chrysotile asbestos is the main cause of pleural mesothelioma. Am J Ind Med 1996; 30: 252 – 66. 10. Mossman BT, Kamp DW, Weitzman SA. Mechanisms of carcinogenesis and clinical features of asbestos-associated cancers. Cancer Invest 1996; 14: 466 – 80. 11. Roggli VL, et al. Malignant mesothelioma and occupational exposure to asbestos: a clinicopathological correlation of 1445 cases. Ultrastruct Pathol 2002; 26: 55 – 65. 12. Pistolesi M, Rusthoven J. Malignant pleural mesothelioma: update, current management, and newer therapeutic strategies. Chest 2004; 126: 1318 – 29. 13. Burdorf A, et al. Future increase of the incidence of mesothelioma due to occupational exposure to asbestos in the past. Ned Tijdschr Geneeskd 1997; 141: 1093 – 8. 14. Bianchi C, et al. Familial mesothelioma of the pleura – a report of 40 cases. Ind Health 2004; 42: 235 – 9. 15. Carbone M, et al. Simian virus 40-like DNA sequences in human pleural mesothelioma. Oncogene 1994; 9: 1781 – 90. 16. Carbone M, et al. New developments about the association of SV40 with human mesothelioma. Oncogene 2003; 22: 5173 – 80. 17. Rollison DE, et al. Case-control study of cancer among US Army veterans exposed to simian virus 40-contaminated adenovirus vaccine. Am J Epidemiol 2004; 160: 317 – 24. 18. Manfredi JJ, et al. Evidence against a role for SV40 in human mesothelioma. Cancer Res 2005; 65: 2602 – 9. 19. Lopez-Rios F, et al. Evidence against a role for SV40 infection in human mesotheliomas and high risk of false-positive PCR results owing to presence of SV40 sequences in common laboratory plasmids. Lancet 2004; 364: 1157 – 66.
20. Cristaudo A, et al. SV40 enhances the risk of malignant mesothelioma among people exposed to asbestos: a molecular epidemiologic casecontrol study. Cancer Res 2005; 65: 3049 – 52. 21. Churg A, et al. The separation of benign and malignant mesothelial proliferations. Am J Surg Pathol 2000; 24: 1183 – 200. 22. Kumaki F, et al. Expression of telomerase reverse transcriptase (TERT) in malignant mesotheliomas. Am J Surg Pathol 2002; 26: 365 – 70. 23. Doglioni C, et al. Calretinin: a novel immunocytochemical marker for mesothelioma. Am J Surg Pathol 1996; 20: 1037 – 46. 24. Gotzos V, Vogt P, Celio MR. The calcium binding protein calretinin is a selective marker for malignant pleural mesotheliomas of the epithelial type. Pathol Res Pract 1996; 192: 137 – 47. 25. Chhieng DC, et al. Calretinin staining pattern aids in the differentiation of mesothelioma from adenocarcinoma in serous effusions. Cancer 2000; 90: 194 – 200. 26. Khoor A, et al. Utility of surfactant protein B precursor and thyroid transcription factor 1 in differentiating adenocarcinoma of the lung from malignant mesothelioma. Hum Pathol 1999; 30: 695 – 700. 27. Bakir K, et al. TTF-1 and surfactant-B as co-adjuvants in the diagnosis of lung adenocarcinoma and pleural mesothelioma. Ann Diagn Pathol 2004; 8: 337 – 41. 28. Brockstedt U, et al. An optimized battery of eight antibodies that can distinguish most cases of epithelial mesothelioma from adenocarcinoma. Am J Clin Pathol 2000; 114: 203 – 9. 29. Ordonez NG. Value of thyroid transcription factor-1, E-cadherin, BG8, WT1, and CD44S immunostaining in distinguishing epithelial pleural mesothelioma from pulmonary and nonpulmonary adenocarcinoma. Am J Surg Pathol 2000; 24: 598 – 606. 30. Amin KM, et al. Wilms’ tumor 1 susceptibility (WT1) gene products are selectively expressed in malignant mesothelioma. Am J Pathol 1995; 146: 344 – 56. 31. Kumar-Singh S, et al. WT1 mutation in malignant mesothelioma and WT1 immunoreactivity in relation to p53 and growth factor receptor expression, cell-type transition, and prognosis. J Pathol 1997; 181: 67 – 74. 32. Thirkettle I, et al. Immunoreactivity for cadherins, HGF/SF, met, and erbB-2 in pleural malignant mesotheliomas. Histopathology 2000; 36: 522 – 8. 33. Oury TD, Hammar SP, Roggli VL. Ultrastructural features of diffuse malignant mesotheliomas. Hum Pathol 1998; 29: 1382 – 92. 34. Sandberg AA, Bridge JA. Updates on the cytogenetics and molecular genetics of bone and soft tissue tumors. Mesothelioma. Cancer Genet Cytogenet 2001; 127: 93 – 110. 35. Taguchi T, et al. Recurrent deletions of specific chromosomal sites in 1p, 3p, 6q, and 9p in human malignant mesothelioma. Cancer Res 1993; 53: 4349 – 55. 36. Cheng JQ, et al. Homozygous deletions within 9p21-p22 identify a small critical region of chromosomal loss in human malignant mesotheliomas. Cancer Res 1993; 53: 4761 – 3. 37. Xio S, et al. Codeletion of p15 and p16 in primary malignant mesothelioma. Oncogene 1995; 11: 511 – 5. 38. Bell DW, Jhanwar SC, Testa JR. Multiple regions of allelic loss from chromosome arm 6q in malignant mesothelioma. Cancer Res 1997; 57: 4057 – 62. 39. Bianchi AB, et al. High frequency of inactivating mutations in the neurofibromatosis type 2 gene (NF2) in primary malignant mesotheliomas. Proc Natl Acad Sci U S A 1995; 92: 10854 – 8. 40. Cheng JQ, et al. Frequent mutations of NF2 and allelic loss from chromosome band 22q12 in malignant mesothelioma: evidence for a two-hit mechanism of NF2 inactivation Genes Chromosomes. Cancer 1999; 24: 238 – 42. 41. Zeiger MA, et al. Loss of heterozygosity on the short arm of chromosome 3 in mesothelioma cell lines and solid tumors. Genes Chromosomes Cancer 1994; 11: 15 – 20. 42. Pylkkanen L, et al. Reduced Fhit protein expression in human malignant mesothelioma. Virchows Arch 2004; 444: 43 – 8. 43. Balsara BR, et al. Comparative genomic hybridization and loss of heterozygosity analyses identify a common region of deletion at 15q11.1 – 15 in human malignant mesothelioma. Cancer Res 1999; 59: 450 – 4.
MESOTHELIOMAS 44. Galffy G, et al. Interleukin 8: an autocrine growth factor for malignant mesothelioma. Cancer Res 1999; 59: 367 – 71. 45. Konig JE, et al. Expression of vascular endothelial growth factor in diffuse malignant pleural mesothelioma. Virchows Arch 1999; 435: 8 – 12. 46. Ohta Y, et al. VEGF and VEGF type C play an important role in angiogenesis and lymphangiogenesis in human malignant mesothelioma tumours. Br J Cancer 1999; 81: 54 – 61. 47. Strizzi L, et al. Vascular endothelial growth factor is an autocrine growth factor in human malignant mesothelioma. J Pathol 2001; 193: 468 – 75. 48. Edwards JG, et al. Angiogenesis is an independent prognostic factor in malignant mesothelioma. Br J Cancer 2001; 85: 863 – 8. 49. Gordon GJ, et al. Identification of novel candidate oncogenes and tumor suppressors in malignant pleural mesothelioma using large-scale transcriptional profiling. Am J Pathol 2005; 166: 1827 – 40. 50. Sun X, et al. Molecular characterization of tumour heterogeneity and malignant mesothelioma cell differentiation by gene profiling. J Pathol 2005; 207: 91 – 101. 51. Kannerstein M, et al. A critique of the criteria for the diagnosis of diffuse malignant mesothelioma. Mt Sinai J Med 1977; 44: 485 – 94. 52. Aisner J. Current approach to malignant mesothelioma of the pleura. Chest 1995; 107: 332S – 344S. 53. Wadler S, et al. Cardiac abnormalities in patients with diffuse malignant pleural mesothelioma. Cancer 1986; 58: 2744 – 50. 54. Nakano T, et al. Thrombocytosis in patients with malignant pleural mesothelioma. Cancer 1986; 58: 1699 – 701. 55. Marom EM, et al. The role of imaging in malignant pleural mesothelioma. Semin Oncol 2002; 29: 26 – 35. 56. Patz EF Jr, et al. Malignant pleural mesothelioma: value of CT and MR imaging in predicting resectability. AJR Am J Roentgenol 1992; 159: 961 – 6. 57. Lorigan JG, Libshitz HI. MR imaging of malignant pleural mesothelioma. J Comput Assist Tomogr 1989; 13: 617 – 20. 58. Conlon KC, Rusch VW, Gillern S. Laparoscopy: an important tool in the staging of malignant pleural mesothelioma. Ann Surg Oncol 1996; 3: 489 – 94. 59. Wang ZJ, et al. Malignant pleural mesothelioma: evaluation with CT, MR imaging, and PET. Radiographics 2004; 24: 105 – 19. 60. Benard F, et al. Prognostic value of FDG PET imaging in malignant pleural mesothelioma. J Nucl Med 1999; 40: 1241 – 5. 61. Flores RM. The role of PET in the surgical management of malignant pleural mesothelioma. Lung Cancer 2005; 49(Suppl. 1): S27 – 32. 62. Butchart EG, et al. Pleuropneumonectomy in the management of diffuse malignant mesothelioma of the pleura. Experience with 29 patients. Thorax 1976; 31: 15 – 24. 63. Rusch VW, From the International Mesothelioma Interest Group. A proposed new international TNM staging system for malignant pleural mesothelioma. Chest 1995; 108: 1122 – 8. 64. van Meerbeeck JP et al., European Organization for Research and Treatment of Cancer Lung Cancer Cooperative Group. A Phase II study of gemcitabine in patients with malignant pleural mesothelioma. Cancer 1999; 85: 2577 – 82. 65. Ruffie P, et al. Diffuse malignant mesothelioma of the pleura in Ontario and Quebec: a retrospective study of 332 patients. J Clin Oncol 1989; 7: 1157 – 68. 66. McCormack PM, et al. Surgical treatment of pleural mesothelioma. J Thorac Cardiovasc Surg 1982; 84: 834 – 42. 67. Law MR, et al. Malignant mesothelioma of the pleura: a study of 52 treated and 64 untreated patients. Thorax 1984; 39: 255 – 9. 68. Faber LP. Surgical treatment of asbestos-related disease of the chest. Surg Clin North Am 1988; 68: 525 – 43. 69. DaValle MJ, et al. Extrapleural pneumonectomy for diffuse, malignant mesothelioma. Ann Thorac Surg 1986; 42: 612 – 8. 70. Wanebo HJ, et al. Pleural mesothelioma. Cancer 1976; 38: 2481 – 8. 71. Achatzy R, et al. The diagnosis, therapy and prognosis of diffuse malignant mesothelioma. Eur J Cardiothorac Surg 1989; 3: 445 – 7. 72. Brancatisano RP, Joseph MG, McCaughan BC. Pleurectomy for mesothelioma. Med J Aust 1991; 154: 455 – 7, 460. 73. Rusch VW. Pleurectomy/decortication and adjuvant therapy for malignant mesothelioma. Chest 1993; 103: 382S – 4S.
289
74. Branscheid D, et al. Diagnostic and therapeutic strategy in malignant pleural mesothelioma. Eur J Cardiothorac Surg 1991; 5: 466 – 72. 75. Lee TT, et al. Radical pleurectomy/decortication and intraoperative radiotherapy followed by conformal radiation with or without chemotherapy for malignant pleural mesothelioma. J Thorac Cardiovasc Surg 2002; 124: 1183 – 9. 76. Rusch VW, Venkatraman E. The importance of surgical staging in the treatment of malignant pleural mesothelioma. J Thorac Cardiovasc Surg 1996; 111: 815 – 25. 77. Worn H. Chances and results of surgery of malignant mesothelioma of the pleura (author’s transl). Thoraxchir Vask Chir 1974; 22: 391 – 3. 78. Bamler KJ, Maassen W. The percentage of benign and malign pleura-tumors among the patients of a clinic of lung surgery with special consideration of the malign pleuramesothelioma and its radical treatment, including results of a diaphragma substitution of preserved dura mater (author’s transl). Thoraxchir Vask Chir 1974; 22: 386 – 91. 79. DeLaria GA, et al. Surgical management of malignant mesothelioma. Ann Thorac Surg 1978; 26: 375 – 82. 80. Geroulanos S, et al. Malignant pleural mesothelioma: diagnosis, therapy and prognosis. Schweiz Rundsch Med Prax 1990; 79: 361 – 7. 81. Sugarbaker DJ, et al. Resection margins, extrapleural nodal status, and cell type determine postoperative long-term survival in trimodality therapy of malignant pleural mesothelioma: results in 183 patients. J Thorac Cardiovasc Surg 1999; 117: 54 – 63. 82. Stevens CW, et al. Treatment planning system evaluation for mesothelioma IMRT. Lung Cancer 2005; 49(Suppl. 1): S75 – 81. 83. Rusch VW, et al. A phase II trial of surgical resection and adjuvant high-dose hemithoracic radiation for malignant pleural mesothelioma. J Thorac Cardiovasc Surg 2001; 122: 788 – 95. 84. Rosenzweig KE, et al. A pilot trial of high-dose-rate intraoperative radiation therapy for malignant pleural mesothelioma. Brachytherapy 2005; 4: 30 – 3. 85. Boutin C, Rey F, Viallat JR. Prevention of malignant seeding after invasive diagnostic procedures in patients with pleural mesothelioma. A randomized trial of local radiotherapy. Chest 1995; 108: 754 – 8. 86. Bydder S, et al. A randomised trial of single-dose radiotherapy to prevent procedure tract metastasis by malignant mesothelioma. Br J Cancer 2004; 91: 9 – 10. 87. Steele JP, Klabatsa A. Chemotherapy options and new advances in malignant pleural mesothelioma. Ann Oncol 2005; 16: 345 – 51. 88. Tomek S, Manegold C. Chemotherapy for malignant pleural mesothelioma: past results and recent developments. Lung Cancer 2004; 45(Suppl. 1): S103 – 19. 89. Nowak AK, et al. Current chemotherapeutic treatment of malignant pleural mesothelioma. Expert Opin Pharmacother 2004; 5: 2441 – 9. 90. van Klaveren RJ, et al. Inadequacy of the RECIST criteria for response evaluation in patients with malignant pleural mesothelioma. Lung Cancer 2004; 43: 63 – 9. 91. Byrne MJ, Nowak AK. Modified RECIST criteria for assessment of response in malignant pleural mesothelioma. Ann Oncol 2004; 15: 257 – 60. 92. Scagliotti GV, et al. Phase II study of pemetrexed with and without folic acid and vitamin B12 as front-line therapy in malignant pleural mesothelioma. J Clin Oncol 2003; 21: 1556 – 61. 93. Kindler HL, et al. Gemcitabine for malignant mesothelioma: A phase II trial by the Cancer and Leukemia Group B. Lung Cancer 2001; 31: 311 – 7. 94. Aversa SM, Favaretto AG. Carboplatin and gemcitabine chemotherapy for malignant pleural mesothelioma: a phase II study of the GSTPV. Clin Lung Cancer 1999; 1: 73 – 5. 95. Favaretto AG, et al. Gemcitabine combined with carboplatin in patients with malignant pleural mesothelioma: a multicentric phase II study. Cancer 2003; 97: 2791 – 7. 96. Byrne MJ, et al. Cisplatin and gemcitabine treatment for malignant mesothelioma: a phase II study. J Clin Oncol 1999; 17: 25 – 30. 97. Nowak AK, et al. A multicentre phase II study of cisplatin and gemcitabine for malignant mesothelioma. Br J Cancer 2002; 87: 491 – 6. 98. van Haarst JM, et al. Multicentre phase II study of gemcitabine and cisplatin in malignant pleural mesothelioma. Br J Cancer 2002; 86: 342 – 5.
290
THORACIC TUMORS
99. Castagneto B, et al. Cisplatin and gemcitabine in malignant pleural mesothelioma: a phase II study. Am J Clin Oncol 2005; 28: 223 – 6. 100. Lerner HJ, et al. Malignant mesothelioma. The Eastern Cooperative Oncology Group (ECOG) experience. Cancer 1983; 52: 1981 – 5. 101. Sorensen PG, et al. Randomized trial of doxorubicin versus cyclophosphamide in diffuse malignant pleural mesothelioma. Cancer Treat Rep 1985; 69: 1431 – 2. 102. Baas P, et al. Caelyx in malignant mesothelioma: a phase II EORTC study. Ann Oncol 2000; 11: 697 – 700. 103. Oh Y, et al. Phase II study of intravenous Doxil in malignant pleural mesothelioma. Invest New Drugs 2000; 18: 243 – 5. 104. Colbert N, et al. A prospective study of detorubicin in malignant mesothelioma. Cancer 1985; 56: 2170 – 4. 105. Kaukel E, et al. A phase II study of pirarubicin in malignant pleural mesothelioma. Cancer 1990; 66: 651 – 4. 106. Magri MD, et al. Epirubicin in the treatment of malignant mesothelioma: a phase II cooperative study. Tumori 1991; 77: 49 – 51. 107. Mattson K, et al. Epirubicin in malignant mesothelioma: a phase II study of the European Organization for Research and Treatment of Cancer Lung Cancer Cooperative Group. J Clin Oncol 1992; 10: 824 – 8. 108. Eisenhauer EA, et al. Phase II study of mitoxantrone in patients with mesothelioma: a National Cancer Institute of Canada Clinical Trials Group Study. Cancer Treat Rep 1986; 70: 1029 – 30. 109. van Breukelen FJ, et al. Mitoxantrone in malignant pleural mesothelioma: a study by the EORTC Lung Cancer Cooperative Group. Eur J Cancer 1991; 27: 1627 – 9. 110. Hudis CA, Kelsen DP. Menogaril in the treatment of malignant mesothelioma: a phase II study. Invest New Drugs 1992; 10: 103 – 6. 111. Sahmoud T, et al. Etoposide in malignant pleural mesothelioma: two phase II trials of the EORTC Lung Cancer Cooperative Group. Eur J Cancer 1997; 33: 2211 – 5. 112. Tammilehto L, et al. Oral etoposide in the treatment of malignant mesothelioma. A phase II study. Ann Oncol 1994; 5: 949 – 50. 113. Falkson G, Vorobiof DA, Lerner HJ. A phase II study of m-AMSA in patients with malignant mesothelioma. Cancer Chemother Pharmacol 1983; 11: 94 – 7. 114. Kindler HL, et al. Irinotecan for malignant mesothelioma A phase II trial by the Cancer and Leukemia Group B. Lung Cancer 2005; 48: 423 – 8. 115. Alberts AS, Falkson G, van Zyl L. Ifosfamide and mesna with doxorubicin have activity in malignant mesothelioma. Eur J Cancer 1990; 26: 1002 – 990. 116. Zidar BL et al., A Southwest Oncology Group Study. A phase II evaluation of ifosfamide and mesna in unresectable diffuse malignant mesothelioma. Cancer 1992; 70: 2547 – 51. 117. Andersen MK, et al. Ifosfamide in malignant mesothelioma: a phase II study. Lung Cancer 1999; 24: 39 – 43. 118. Talbot SM, et al. High-dose ifosfamide with mesna and granuloctyecolony-stimulating factor (recombinant human G-CSF) in patients with unresectable malignant mesothelioma. Cancer 2003; 98: 331 – 6. 119. Bajorin D, Kelsen D, Mintzer DM. Phase II trial of mitomycin in malignant mesothelioma. Cancer Treat Rep 1987; 71: 857 – 8. 120. Mintzer DM, et al. Phase II trial of high-dose cisplatin in patients with malignant mesothelioma. Cancer Treat Rep 1985; 69: 711 – 2. 121. Zidar BL, et al. A phase II evaluation of cisplatin in unresectable diffuse malignant mesothelioma: a Southwest Oncology Group Study. Invest New Drugs 1988; 6: 223 – 6. 122. Mbidde EK, et al. Phase II trial of carboplatin (JM8) in treatment of patients with malignant mesothelioma. Cancer Chemother Pharmacol 1986; 18: 284 – 5. 123. Vogelzang NJ, et al. Carboplatin in malignant mesothelioma: a phase II study of the Cancer and Leukemia Group B. Cancer Chemother Pharmacol 1990; 27: 239 – 42. 124. Raghavan D, et al. Phase II trial of carboplatin in the management of malignant mesothelioma. J Clin Oncol 1990; 8: 151 – 4. 125. Giaccone G, et al. Phase II trial of ZD0473 as second-line therapy in mesothelioma. Eur J Cancer 2002; 38(Suppl. 8): S19 – 24. 126. van Meerbeeck JP, et al. A phase II EORTC study of temozolomide in patients with malignant pleural mesothelioma. Eur J Cancer 2002; 38: 779 – 83.
127. Solheim OP, et al. High-dose methotrexate in the treatment of malignant mesothelioma of the pleura. A phase II study. Br J Cancer 1992; 65: 956 – 60. 128. Vogelzang NJ, et al. Trimetrexate in malignant mesothelioma: a Cancer and Leukemia Group B Phase II study. J Clin Oncol 1994; 12: 1436 – 42. 129. Kindler HL, et al. Edatrexate (10-ethyl-deaza-aminopterin) (NSC #626715) with or without leucovorin rescue for malignant mesothelioma. Sequential phase II trials by the cancer and leukemia group B. Cancer 1999; 86: 1985 – 91. 130. Cantwell BM, Earnshaw M, Harris AL. Phase II study of a novel antifolate, N10-propargyl-5,8 dideazafolic acid (CB3717), in malignant mesothelioma. Cancer Treat Rep 1986; 70: 1335 – 6. 131. Harvey VJ, et al. Chemotherapy of diffuse malignant mesothelioma. Phase II trials of single-agent 5-fluorouracil and adriamycin. Cancer 1984; 54: 961 – 4. 132. Otterson GA, et al. Capecitabine in malignant mesothelioma: a phase II trial by the Cancer and Leukemia Group B (39807). Lung Cancer 2004; 44: 251 – 9. 133. Vogelzang NJ et al., Cancer and Leukemia Group B. Dihydro-5azacytidine in malignant mesothelioma. A phase II trial demonstrating activity accompanied by cardiac toxicity. Cancer 1997; 79: 2237 – 42. 134. Dhingra HM, et al. Phase II trial of 5,6-dihydro-5-azacytidine in pleural malignant mesothelioma. Invest New Drugs 1991; 9: 69 – 72. 135. Martensson G, Sorenson S. A phase II study of vincristine in malignant mesothelioma – a negative report. Cancer Chemother Pharmacol 1989; 24: 133 – 4. 136. Cowan JD, et al. Phase II trial of five day intravenous infusion vinblastine sulfate in patients with diffuse malignant mesothelioma: a Southwest Oncology Group study. Invest New Drugs 1988; 6: 247 – 8. 137. Kelsen D, et al. Vindesine in the treatment of malignant mesothelioma: a phase II study. Cancer Treat Rep 1983; 67: 821 – 2. 138. Boutin C, et al. Phase II trial of vindesine in malignant pleural mesothelioma. Cancer Treat Rep 1987; 71: 205 – 6. 139. Steele JP, et al. Phase II study of vinorelbine in patients with malignant pleural mesothelioma. J Clin Oncol 2000; 18: 3912 – 7. 140. Vorobiof DA, et al. Malignant pleural mesothelioma: a phase II trial with docetaxel. Ann Oncol 2002; 13: 412 – 5. 141. Belani CP, et al. Docetaxel for malignant mesothelioma: phase II study of the Eastern Cooperative Oncology Group. Clin Lung Cancer 2004; 6: 43 – 7. 142. van Meerbeeck J, et al. Paclitaxel for malignant pleural mesothelioma: a phase II study of the EORTC Lung Cancer Cooperative Group. Br J Cancer 1996; 74: 961 – 3. 143. Vogelzang NJ, et al. High-dose paclitaxel plus G-CSF for malignant mesothelioma: CALGB phase II study 9234. Ann Oncol 1999; 10: 597 – 600. 144. Eagan RT, et al. Phase II trial of diaziquone in malignant mesothelioma. Cancer Treat Rep 1986; 70: 429. 145. Falkson G, et al. Phase II trial of acivicin in malignant mesothelioma. Cancer Treat Rep 1987; 71: 545 – 6. 146. Mulatero CW, et al. A phase II study of combined intravenous and subcutaneous interleukin-2 in malignant pleural mesothelioma. Lung Cancer 2001; 31: 67 – 72. 147. Mikulski SM, et al. Phase II trial of a single weekly intravenous dose of ranpirnase in patients with unresectable malignant mesothelioma. J Clin Oncol 2002; 20: 274 – 81. 148. Baas P, et al. Thalidomide in patients with malignant pleural mesothelioma. Lung Cancer 2005; 48: 291 – 6. 149. Govindan R, et al. Gefitinib in patients with malignant mesothelioma: a phase II study by the Cancer and Leukemia Group B. Clin Cancer Res 2005; 11: 2300 – 4. 150. Mathy A, et al. Limited efficacy of imatinib mesylate in malignant mesothelioma: A phase II trial. Lung Cancer 2005; 50(1): 83 – 6. 151. Cortese JF, et al. Common EGFR mutations conferring sensitivity to gefitinib in lung adenocarcinoma are not prevalent in human malignant mesothelioma. Int J Cancer 2006; 118: 521 – 2. 152. Butnor KJ, et al. The spectrum of Kit (CD117) immunoreactivity in lung and pleural tumors: a study of 96 cases using a single-source antibody with a review of the literature. Arch Pathol Lab Med 2004; 128: 538 – 43.
MESOTHELIOMAS 153. Horvai AE, et al. c-Kit is not expressed in malignant mesothelioma. Mod Pathol 2003; 16: 818 – 22. 154. Berghmans T, et al. Activity of chemotherapy and immunotherapy on malignant mesothelioma: a systematic review of the literature with meta-analysis. Lung Cancer 2002; 38: 111 – 21. 155. Vogelzang NJ, et al. Phase III study of pemetrexed in combination with cisplatin versus cisplatin alone in patients with malignant pleural mesothelioma. J Clin Oncol 2003; 21: 2636 – 44. 156. van Meerbeeck JP, et al. A randomized phase III study of cisplatin with or without raltitrexed in patients with malignant pleural mesothelioma: an Intergroup Study of the EORTC Lung Cancer Group and the NCIC. J Clin Oncol 2005; 23: 6881 – 9. 157. Porta C, et al. Raltitrexed-Oxaliplatin combination chemotherapy is inactive as second-line treatment for malignant pleural mesothelioma patients. Lung Cancer 2005; 48: 429 – 34. 158. Manegold C, et al. Second-line (post-study) chemotherapy received by patients treated in the phase III trial of pemetrexed plus cisplatin versus cisplatin alone in malignant pleural mesothelioma. Ann Oncol 2005; 16: 923 – 7. 159. Samson MK, et al. Randomized comparison of cyclophosphamide, imidazole carboxamide, and adriamycin versus cyclophosphamide and adriamycin in patients with advanced stage malignant mesothelioma: a Sarcoma Intergroup Study. J Clin Oncol 1987; 5: 86 – 91. 160. White SC, et al. Randomised phase II study of cisplatin-etoposide versus infusional carboplatin in advanced non-small-cell lung cancer and mesothelioma. Ann Oncol 2000; 11: 201 – 6. 161. Hilaris BS, et al. Pleurectomy and intraoperative brachytherapy and postoperative radiation in the treatment of malignant pleural mesothelioma. Int J Radiat Oncol Biol Phys 1984; 10: 325 – 31. 162. Rusch V, et al. A phase II trial of pleurectomy/decortication followed by intrapleural and systemic chemotherapy for malignant pleural mesothelioma. J Clin Oncol 1994; 12: 1156 – 63. 163. Lee JD, et al. Intrapleural chemotherapy for patients with incompletely resected malignant mesothelioma: the UCLA experience. J Surg Oncol 1995; 60: 262 – 7. 164. Rice TW, et al. Aggressive multimodality therapy for malignant pleural mesothelioma. Ann Thorac Surg 1994; 58: 24 – 9. 165. Sauter ER, et al. Optimal management of malignant mesothelioma after subtotal pleurectomy: revisiting the role of intrapleural chemotherapy and postoperative radiation. J Surg Oncol 1995; 60: 100 – 5. 166. Rusch VW, Venkatraman ES. Important prognostic factors in patients with malignant pleural mesothelioma, managed surgically. Ann Thorac Surg 1999; 68: 1799 – 804. 167. Sugarbaker DJ, et al. Prevention, early detection, and management of complications after 328 consecutive extrapleural pneumonectomies. J Thorac Cardiovasc Surg 2004; 128: 138 – 46. 168. Baldini EH, et al. Patterns of failure after trimodality therapy for malignant pleural mesothelioma. Ann Thorac Surg 1997; 63: 334 – 8. 169. Maggi G, et al. Trimodality management of malignant pleural mesothelioma. Eur J Cardiothorac Surg 2001; 19: 346 – 50. 170. Pass HW, et al. A phase II trial investigating primary immunochemotherapy for malignant pleural mesothelioma and the feasibility of adjuvant immunochemotherapy after maximal cytoreduction. Ann Surg Oncol 1995; 2: 214 – 20. 171. Weder W, et al. Neoadjuvant chemotherapy followed by extrapleural pneumonectomy in malignant pleural mesothelioma. J Clin Oncol 2004; 22: 3451 – 7. 172. Boutin C, et al. Intrapleural treatment with recombinant gammainterferon in early stage malignant pleural mesothelioma. Cancer 1994; 74: 2460 – 7. 173. Monnet I, et al. Intrapleural infusion of activated macrophages and gamma-interferon in malignant pleural mesothelioma: a phase II study. Chest 2002; 121: 1921 – 7. 174. Goey SH, et al. Intrapleural administration of interleukin 2 in pleural mesothelioma: a phase I-II study. Br J Cancer 1995; 72: 1283 – 8. 175. Stam TC, et al. Intrapleural administration of tumour necrosis factoralpha (TNFalpha) in patients with mesothelioma: cytokine patterns and acute-phase protein response. Eur J Clin Invest 2000; 30: 336 – 43. 176. Castagneto B, et al. Palliative and therapeutic activity of IL-2 immunotherapy in unresectable malignant pleural mesothelioma with
177. 178.
179.
180.
181.
182.
183.
184.
185.
186.
187.
188.
189. 190.
191. 192. 193.
194.
195.
196.
197. 198.
199.
200.
291
pleural effusion: results of a phase II study on 31 consecutive patients. Lung Cancer 2001; 31: 303 – 10. Ris HB. Photodynamic therapy as an adjunct to surgery for malignant pleural mesothelioma. Lung Cancer 2005; 49(Suppl 1): S65 – 8. Schouwink H, et al. Intraoperative photodynamic therapy after pleuropneumonectomy in patients with malignant pleural mesothelioma: dose finding and toxicity results. Chest 2001; 120: 1167 – 74. van Ruth S, et al. Cytoreductive surgery combined with intraoperative hyperthermic intrathoracic chemotherapy for stage I malignant pleural mesothelioma. Ann Surg Oncol 2003; 10: 176 – 82. Ratto GB, et al. Pleural space perfusion with cisplatin in the multimodality treatment of malignant mesothelioma: a feasibility and pharmacokinetic study. J Thorac Cardiovasc Surg 1999; 117: 759 – 65. Lu C, et al. Phase II study of a liposome-entrapped cisplatin analog (L-NDDP) administered intrapleurally and pathologic response rates in patients with malignant pleural mesothelioma. J Clin Oncol 2005; 23: 3495 – 501. Sterman DH, et al. A pilot study of systemic corticosteroid administration in conjunction with intrapleural adenoviral vector administration in patients with malignant pleural mesothelioma. Cancer Gene Ther 2000; 7: 1511 – 8. Sterman DH, Kaiser LR, Albelda SM. Gene therapy for malignant pleural mesothelioma. Hematol Oncol Clin North Am 1998; 12: 553 – 68. Curran D, et al. Prognostic factors in patients with pleural mesothelioma: the European Organization for Research and Treatment of Cancer experience. J Clin Oncol 1998; 16: 145 – 52. Herndon JE, et al. Factors predictive of survival among 337 patients with mesothelioma treated between 1984 and 1994 by the Cancer and Leukemia Group B. Chest 1998; 113: 723 – 31. Edwards JG, et al. Prognostic factors for malignant mesothelioma in 142 patients: validation of CALGB and EORTC prognostic scoring systems. Thorax 2000; 55: 731 – 5. Fennell DA, et al. Statistical validation of the EORTC prognostic model for malignant pleural mesothelioma based on three consecutive phase II trials. J Clin Oncol 2005; 23: 184 – 9. Bongiovanni M, et al. p27(kip1) immunoreactivity correlates with long-term survival in pleural malignant mesothelioma. Cancer 2001; 92: 1245 – 50. Gordon GJ. Transcriptional profiling of mesothelioma using microarrays. Lung Cancer 2005; 49(Suppl 1): S99 – S103. Pass HI, et al. Gene expression profiles predict survival and progression of pleural mesothelioma. Clin Cancer Res 2004; 10: 849 – 59. Gordon GJ, et al. Validation of genomics-based prognostic tests in malignant pleural mesothelioma. Clin Cancer Res 2005; 11: 4406 – 14. Mohamed F, Sugarbaker PH. Peritoneal mesothelioma. Curr Treat Options Oncol 2002; 3: 375 – 86. Antman K, et al. Malignant mesothelioma: prognostic variables in a registry of 180 patients, the Dana-Farber Cancer Institute and Brigham and Women’s Hospital experience over two decades, 1965 – 1985. J Clin Oncol 1988; 6: 147 – 53. Park BJ, et al. Treatment of primary peritoneal mesothelioma by continuous hyperthermic peritoneal perfusion (CHPP). Ann Surg Oncol 1999; 6: 582 – 90. Deraco M, et al. Peritoneal mesothelioma treated by induction chemotherapy, cytoreductive surgery, and intraperitoneal hyperthermic perfusion. J Surg Oncol 2003; 83: 147 – 53. Sugarbaker PH, et al. A review of peritoneal mesothelioma at the Washington Cancer Institute. Surg Oncol Clin N Am 2003; 12: 605 – 21, xi. Le DT, et al. Cisplatin and irinotecan (CPT-11) for peritoneal mesothelioma. Cancer Invest 2003; 21: 682 – 9. Janne PA, et al. Open-label study of pemetrexed alone or in combination with Cisplatin for the treatment of patients with peritoneal mesothelioma: outcomes of an expanded access program. Clin Lung Cancer 2005; 7: 40 – 6. Gupta NP, et al. Malignant mesothelioma of the tunica vaginalis testis: a report of two cases and review of literature. J Surg Oncol 1999; 70: 251 – 4. Vigneswaran WT, Stefanacci PR. Pericardial mesothelioma. Curr Treat Options Oncol 2000; 1: 299 – 302.
292
THORACIC TUMORS
201. Briselli M, Mark EJ, Dickersin GR. Solitary fibrous tumors of the pleura: eight new cases and review of 360 cases in the literature. Cancer 1981; 47: 2678 – 89. 202. Sung SH, et al. Solitary fibrous tumors of the pleura: surgical outcome and clinical course. Ann Thorac Surg 2005; 79: 303 – 7.
203. Sethna K, et al. Peritoneal cystic mesothelioma: a case series. Tumori 2003; 89: 31 – 5.
Section 5 : Thoracic Tumors
24
Primary Melanoma of the Lung Richard A. Scolyer, James F. Bishop and John F. Thompson
INTRODUCTION Although most primary melanomas develop as cutaneous lesions and a few develop in recognized mucosal sites, the very occasional occurrence of primary melanomas in unusual noncutaneous sites, including the lung, is well documented.1 Primary melanoma of the lung is undoubtedly a very rare tumor. However, its precise incidence is the subject of continuing debate. Some have even questioned whether the condition actually exists. The difficulty arises because an unrecognized primary cutaneous melanoma can occasionally undergo complete regression and disappear without trace, while a lung metastasis which has arisen from it continues on to grow and may eventually present as an apparently isolated focus of melanoma.2,3 Because melanocytes are not normally found in the tracheobronchial tree, it is also difficult to explain the histogenesis of a primary lung melanoma. In patients who do have a history of primary cutaneous melanoma and who subsequently present with metastatic disease, the lungs are the only clinically detectable site of metastasis in 7–9% of cases.4 Melanoma has a predilection to metastasize to the lungs,5,6 with approximately 70% of patients who die of melanoma being found to have pulmonary metastases at autopsy.7,8 Nevertheless, the literature contains many reports of patients with melanomas that seem to be of primary lung origin. In these patients, their history and subsequent follow-up appear to exclude the possibility of metastasis to the lung from a primary site elsewhere.7,9 – 51 Supporting evidence for a diagnosis of primary lung melanoma may be obtained from histopathologic features of the tumor, from a pattern of subsequent involvement of regional lymph nodes which is consistent with that of other tumors of primary bronchial origin, and from failure to find a primary melanoma elsewhere either during the lifetime of the patient or at the time of autopsy. The occurrence of primary melanomas in other noncutaneous sites has long been recognized1,12,52 and is much more comprehensively documented in the literature. Ocular and conjunctival melanomas are the most common, but noncutaneous melanomas are also well recognized to arise in the urethra, the vulva, the vagina, the leptomeninges, the adrenal
gland, the nasopharynx, the oropharynx, the esophagus, the gall bladder, and occasionally in other parts of the gastrointestinal tract including the stomach, the small bowel, the large bowel, and the anal canal.53 An important difference, however, is that melanocytes have been identified in most of these sites in normal individuals, providing a plausible histogenetic basis for the origin of the melanomas, whereas melanocytes are not usually found in lung tissue.
Criteria for Diagnosis To define cases of primary lung melanomas with greater certainty, minimal criteria for the diagnosis have been proposed. The first such proposal was in 1967 by Jensen and Egedorf,16 who suggested that a diagnosis of primary lung melanoma should be made in the following circumstances: • no history suggestive of a previous melanoma (cutaneous or ocular) • no demonstrable melanoma in any other organ at the time of surgery • a solitary tumor in the surgical specimen from the lung • tumor morphology compatible with a primary tumor • no evidence at autopsy of a primary melanoma elsewhere Other authors have since endorsed these criteria, with minor variations and additions.17,26,31,43,54 However, detailed staging investigations and prolonged follow-up are probably necessary before accepting a definitive diagnosis of primary pulmonary melanoma because, as recently reported by de Wilt et al.,44 some patients presenting with apparently isolated pulmonary melanoma may subsequently develop cutaneous or nodal recurrences suggesting that the tumor originated from a regressed primary cutaneous melanoma. Although the criteria proposed by Jensen and Egedorf are undoubtedly appropriate and desirable, it is nevertheless clear that there are instances in which all the criteria cannot be satisfied yet in which the likelihood is high that the lung tumor is a primary melanoma.38
Incidence of Primary Lung Melanoma In a recent review from the Sydney Melanoma Unit in Sydney, Australia, de Wilt and colleagues identified 27 patients
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
294
THORACIC TUMORS
who presented with pulmonary melanoma as their first sign of disease and in whom no other melanoma metastases were found.44 This was from a total of 19 000 patients with melanoma treated over a 50-year period. Following a detailed analysis of clinical, pathologic and follow-up data, the authors concluded that seven of these patients were likely to have had primary pulmonary melanomas. Reviewing 10 134 patients with primary lung tumors treated at the Mayo Clinic over a 10-year period, Miller and Allen40 reported three patients with primary lung melanomas. In a report from Japan published in 1998, Sekine et al. identified one case of primary pulmonary melanoma from 3481 primary lung tumors resected over a 20-year period.45 Each of these reports emphasizes the rarity of primary lung melanoma, but cannot be used to quantify its incidence since the patients seen at each of the institutions were referred for treatment and the data were therefore not population based. Although there have been numerous case reports claiming a diagnosis of primary lung melanoma, the patients described in many of the early publications did not fulfill the criteria subsequently accepted as being necessary to definitively establish the diagnosis. The first two recorded cases, reported by Todd in 1888,9 were certainly based on insufficient evidence to warrant the diagnosis. Similarly, other early reports by Kunkel and Torrey in 1916,10 and by Carlucci and Schleussner in 194211 were deficient because full autopsies were not performed to exclude the possibility of a primary melanoma elsewhere. It was not until 1963 that a patient fulfilling subsequently accepted criteria for the diagnosis of primary lung melanoma was reported.7 In a 1987 publication, Alghanem et al.31 considered it appropriate to accept only seven previously reported cases as true primary lung melanomas, on the basis of stringent criteria for diagnosis. Reporting a series of eight new cases in 1997, Wilson and Moran suggested that there were fewer than 25 previous cases acceptable as primary pulmonary melanomas.46 As far as could be determined from a comprehensive review of the literature by the present authors, the total number of reported cases for which the diagnosis had been established by appropriate criteria stood at 53 in 2005.
HISTOGENESIS A histogenetic basis for the development of primary lung melanomas is not obvious, but several explanations have been proposed. Although melanocytes have not been demonstrated in the lower respiratory tract, some investigators hypothesize that there can be aberrant migration of these potential precursor cells to the lung, and suggest that this could explain the development of primary melanomas at this site.16 This seems a plausible explanation because embryologically the respiratory tract develops from an outgrowth of the primitive foregut between the pharynx and esophagus, in the region of the larynx, and melanocytes have been identified in the mucosa of these sites.55 – 57 Because squamous metaplasia is occasionally observed in melanoma-affected epithelium, a proposed alternative explanation is that epithelial cells undergo metaplastic transformation into melanocytes.14 However, it appears more
likely that the squamous metaplasia is a consequence of bronchial involvement by melanoma. A condition described as “melanogenic metaplasia” of mucous glands in the oral cavity has also been reported,58 and it is possible that a similar process might occur in the mucous glands of the tracheobronchial tree. Another proposal, perhaps more plausible, is that neuroendocrine (Kulchitsky) precursor cells have the potential to undergo melanocytic differentiation; both cell types are histogenetically related, being of neural crest origin.38 This theory is supported by the occasional occurrence of melanocytic differentiation in carcinoid tumors.59 A similar theory has been proposed to explain the origin of primary adrenal melanomas,30,60 – 62 since melanocytes have not been identified in this location either. The occurrence of melanocytic differentiation within peripheral nerve sheath tumors, such as melanotic schwannomas, raises the possibility that primary pulmonary melanomas may also arise from native pulmonary neural elements.
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS The cases of lung melanoma reported in the literature are divided approximately equally into those which presented as a polypoid obstructing lesion within the tracheobronchial tree and those which presented as a mass within the lung parenchyma. The tumor is almost always unifocal, but multifocal primary lung melanomas have been described.26,60 Primary pleural melanoma has also been reported.21 As expected, the presenting symptoms of a patient with a primary lung melanoma are determined by the site and size of the tumor. Initial recognition of an abnormality is sometimes made on a chest X-ray performed for an unrelated reason.18,36 A definitive histologic diagnosis of melanoma can usually be obtained for endobronchial lesions by bronchoscopy and biopsy, and for lesions in the peripheral lung parenchyma by fine needle aspiration biopsy. However, a diagnosis of primary (rather than metastatic) lung melanoma is unlikely to be made at this time, and may not be reached until after full staging investigations, pathologic examination of the resected tumor, an appropriate period of follow-up, and possibly at autopsy.
PATHOLOGY In 1968, Allen and Drash suggested that the surgical pathologist must always consider the possibility of melanoma when examining a lung tumor if the diagnosis is not to be missed.17 Nearly 40 years later this advice may still be appropriate. Melanoma remains “the great imitator” and unless the possibility is considered, primary lung melanoma may be misdiagnosed as a large cell carcinoma or as a sarcoma of the lung. Having considered the possibility of a melanoma, confirmation or exclusion of this possibility should then be achievable by careful examination for the classic features of melanoma, with support from immunohistochemistry and/or electron microscopy.
PRIMARY MELANOMA OF THE LUNG
295
cutaneous melanoma, there is the additional possibility of systemic dissemination via the bloodstream to such sites as brain, liver, adrenal gland, and so on. Metastatic disease involving the pleura, pericardium, and heart also occurs.35,36
TREATMENT
Figure 1 Primary pulmonary melanoma occurring in a male aged 61 years. Pleomorphic epithelioid melanoma cells are present, some of which are pigmented, beneath bronchial respiratory-type mucosa. There is focal mucosal surface erosion by the tumor. (H&E stained section, original magnification ×100).
A number of histologic features have been suggested as being indicative of primary lung melanoma,17,28,36 including the following: • obvious melanoma cells, confirmed by immunohistochemical staining for S-100 and HMB-45 and possibly by electron microscopy (see Figure 1) • evidence of junctional change • “nesting” of cells beneath the bronchial epithelium • invasion of the intact (i.e. nonulcerated) bronchial epithelium by melanoma cells However, these features were not demonstrated in many of the previously reported cases of primary pulmonary melanoma46 and, furthermore, are probably not as specific as has been suggested. “In situ melanoma” changes, which the latter three criteria represent, may be absent in primary melanomas occurring at other sites, particularly in ulcerated primary cutaneous melanomas. In addition, similar changes are sometimes observed in epidermotropic melanomas metastasizing to the skin,63 and an intraepithelial growth of melanoma metastatic to the lung has been documented previously.64 Indeed, the results of one recent series of 15 patients suggest that distinguishing primary from metastatic melanoma is best performed on the basis of clinical behavior, particularly the pattern of metastatic spread, rather than on histopathologic criteria.44
Patterns of Metastasis Primary melanoma of the lung metastasizes in a pattern consistent with that of other primary lung tumors – indeed, this is one of the points of evidence raised in support of the existence of lung melanoma as an entity. Thus, involvement of regional lymph nodes in the lung hilum and mediastinum is likely to be observed. As with primary
On the basis of experience with primary melanomas arising in other sites, the treatment of choice for primary lung melanoma is radical surgical excision. This will usually involve formal lobectomy or pneumonectomy. For primary cutaneous melanoma, the prognosis for the patient does not appear to be significantly worse if regional lymph node dissection is delayed until the metastatic disease in regional nodes becomes clinically apparent. For primary lung melanoma, however, there is little possibility of performing satisfactory delayed regional lymph node clearance in the lung hilum and mediastinum following lobectomy and pneumonectomy. It therefore seems logical to clear these nodes at the time of the initial definitive lung surgery whenever possible. Lymphatic mapping and sentinel lymphadenectomy have been performed in patients with primary and secondary lung tumors, including melanomas, and may provide more accurate pathologic staging.65 The major difficulty with this theoretically ideal approach is that the diagnosis of primary rather than secondary lung melanoma may not be established with reasonable certainty until after detailed pathologic examination of the resected tumor. In the absence of any evidence to indicate otherwise, the treatment of locally recurrent melanoma from a lung primary must be based on the standard treatment principles established for other forms of melanoma. If radical surgical excision is possible, it provides the best form of palliation, and may even achieve cure. If whole body imaging by computed tomography (CT) scanning or more reliably by positron emission tomography (PET) scanning61 does not reveal any evidence of metastatic disease elsewhere, radical surgery is certainly indicated. Because melanoma tissue always has a very high glucose uptake, whole body PET scans with fluorine-18-fluorodeoxyglucose61,62,66 – 71 has largely replaced CT and magnetic resonance imaging (MRI) scans as the staging investigation of choice in most major melanoma treatment centers. This imaging modality is therefore likely to assist in determining whether the focus of melanoma in the lung is a primary or a secondary tumor. If a diagnosis of primary lung melanoma is confirmed, PET scanning should also demonstrate whether distant metastasis has occurred and, if it has, should avoid inappropriate surgery as a treatment for primary lung melanoma. If surgical clearance of the recurrent disease is not feasible, systemic chemotherapy may be considered, or radiotherapy if the tumor mass in the chest is causing troublesome symptoms. Foci of metastatic disease outside the chest must similarly be treated on their merits, according to general principles for the treatment of metastatic melanoma, since there are no data to indicate that any different form of treatment is likely to be more effective for primary lung melanomas in this situation.
296
THORACIC TUMORS
PROGNOSIS The very limited information which can be gleaned from published reports of patients with primary lung melanomas indicates that the prognosis is generally very poor, with the principal determinant of outcome being the presence or absence of local (peribronchial and hilar) lymph node metastases. However, in one recent series of 15 patients who presented with isolated pulmonary melanoma with no known primary tumor, the overall actuarial survival was 42%.44 This is remarkably high when compared with other studies reporting resection of pulmonary metastatic melanomas in which actuarial 5-year survival rates of about 20% were observed.72 – 74 In addition, a number of long-term survivors after treatment of apparently primary pulmonary melanoma by radical surgery have also been reported.14,15,18,44 Longterm survival after treatment of primary lung melanoma by chemotherapy, immunotherapy, or radiotherapy alone, however, has not been documented. It is possible that some tumors diagnosed as primary lung melanomas may actually be metastases from totally regressed primary cutaneous melanomas. Thus, the results of treating foci of metastatic melanoma from an occult primary site warrant review. Several studies have suggested that the survival outcome for patients with metastatic disease from an occult primary tumor does not differ significantly from the outcome for patients with metastatic disease from a known primary site,75 – 79 but a recent study from the Sydney Melanoma Unit indicated a somewhat more favorable prognosis compared with patients with known sites of primary disease.80 Thus, if no other focus of primary or secondary melanoma in the body can be demonstrated by the best available imaging techniques, radical surgical treatment for a suspected primary lung melanoma is entirely logical and offers the patient the best chance of cure, even if it is subsequently shown to be a deposit of metastatic melanoma.
REFERENCES 1. Dasgupta TK, Brasfield RD, Paglia MA. Primary melanomas in unusual sites. Surg Gynecol Obstet 1969; 128: 841 – 8. 2. Smith JL Jr, Stehlin JS Jr. Spontaneous regression of primary malignant melanomas with regional metastases. Cancer 1965; 18: 1399 – 415. 3. Milton GW, Lane Brown MM, Gilder M. Malignant melanoma with an occult primary lesion. Br J Surg 1967; 54: 651 – 8. 4. Gromet MA, et al. The thorax as the initial site for systemic relapse in malignant melanoma: a prospective survey of 324 patients. Cancer 1979; 44: 776 – 84. 5. Sutton FD, Vestal RE, Creagh CE Jr. Varied presentations of metastatic pulmonary melanoma. Chest 1974; 65: 415 – 9. 6. Webb WR, Gamsu G. Thoracic metastasis in malignant melanoma. A radiographic survey of 65 patients. Chest 1977; 71: 176 – 81. 7. Salm R. A primary malignant melanoma of the bronchus. J Pathol Bacteriol 1963; 85: 121 – 6. 8. Dasgupta T, Brasfield R. Metastatic melanoma. A clinicopathological study. Cancer 1964; 17: 1323 – 39. 9. Todd FW. Two cases of melanotic tumors in the lungs. JAMA 1888; 11: 53 – 4. 10. Kunkel OF, Torrey E. Report of a case of primary melanotic sarcoma of the lung presenting difficulties in differentiating from tuberculosis. N Y State J Med 1916; 16: 198 – 201. 11. Carlucci GA, Schleussner RC. Primary melanoma of the lung: case report. J Thorac Surg 1942; 11: 643 – 9.
12. Allen AC, Spitz S. Malignant melanoma; a clinicopathological analysis of the criteria for diagnosis and prognosis. Cancer 1953; 6: 1 – 45. 13. Hsu CW, Wu SC, Ch’En CS. Melanoma of lung. Chin Med J 1962; 81: 263 – 6. 14. Reed RJ, Kent EM. Solitary pulmonary melanomas: two case reports. J Thorac Cardiovasc Surg 1964; 48: 226 – 31. 15. Reid JD, Mehta VT. Melanoma of the lower respiratory tract. Cancer 1966; 19: 627 – 31. 16. Jensen OA, Egedorf J. Primary malignant melanoma of the lung. Scand J Respir Dis 1967; 48: 127 – 35. 17. Allen MS, Drash EC Jr. Primary melanoma of the lung. Cancer 1968; 21: 154 – 9. 18. Taboada CF, et al. Primary melanoma of the lung. Chest 1972; 62: 629 – 31. 19. Walter P, Fernandes C, Florange W. Melanome malin primitif pulmonaire. Ann Anat Pathol (Paris) 1972; 17: 91 – 9. 20. Mori K, Cho H, Som M. Primary “flat” melanoma of the trachea. J Pathol 1977; 121: 101 – 5. 21. Smith S, Opipari MI. Primary pleural melanoma. A first reported case and literature review. J Thorac Cardiovasc Surg 1978; 75: 827 – 31. 22. Adebonojo SA, Grillo IA, Durodola JI. Primary malignant melanoma of the bronchus. J Natl Med Assoc 1979; 71: 579 – 81. 23. Robertson AJ, et al. Primary melanocarcinoma of the lower respiratory tract. Thorax 1980; 35: 158 – 9. 24. Weshler Z, et al. Bronchial malignant melanoma. J Surg Oncol 1980; 15: 243 – 8. 25. Gephardt GN. Malignant melanoma of the bronchus. Hum Pathol 1981; 12: 671 – 3. 26. Verweij J, Breed WP, Jansveld CA. Primary tracheo-bronchial melanoma. Neth J Med 1982; 25: 163 – 6. 27. Roldan JdeG, Pla RV, JAC M. Nodulo puylmonar soitario en el curso de melanoma maligno. A proposito de cuatro observaciones. Arch Bronconeumol 1983; 19: 128 – 31. 28. Angel R, Prados M. Primary bronchial melanoma. J La State Med Soc 1984; 136: 13 – 5. 29. Cagle P, et al. Pulmonary melanoma. Primary vs metastatic. Chest 1984; 85: 125 – 6. 30. Carstens PH, Kuhns JG, Ghazi C. Primary malignant melanomas of the lung and adrenal. Hum Pathol 1984; 15: 910 – 4. 31. Alghanem AA, Mehan J, Hassan AA. Primary malignant melanoma of the lung. J Surg Oncol 1987; 34: 109 – 12. 32. Demeter SL, Fuenning C, Miller JB. Primary malignant melanoma of the lower respiratory tract: endoscopic identification. Cleve Clin J Med 1987; 54: 305 – 8. 33. Santos F, et al. Primary bronchopulmonary malignant melanoma. Case report. Scand J Thorac Cardiovasc Surg 1987; 21: 187 – 9. 34. Andre N, et al. Malignant endobronchial melanoma, apparently primary. Apropos of a case. Rev Pneumol Clin 1988; 44: 143 – 5. 35. Bagwell SP, et al. Primary malignant melanoma of the lung. Am Rev Respir Dis 1989; 139: 1543 – 7. 36. Bertola G, et al. Primary lung melanoma. Ital J Surg Sci 1989; 19: 187 – 9. 37. Cohen RE, et al. Pulmonary blastoma with malignant melanoma component. Arch Pathol Lab Med 1990; 114: 1076 – 8. 38. Jennings TA, et al. Primary malignant melanoma of the lower respiratory tract. Report of a case and literature review. Am J Clin Pathol 1990; 94: 649 – 55. 39. Sanchez Navarro JJ, et al. Melanoma maligno primitivo de pulmon. Rev Clin Esp 1991; 188: 68 – 9. 40. Miller DL, Allen MS. Rare pulmonary neoplasms. Mayo Clin Proc 1993; 68: 492 – 8. 41. Barzo P, et al. Primary malignant melanoma of the lung and lower respiratory tract. Orv Hetil 1994; 135: 245 – 9. 42. Pasquini E, et al. Primary bronchial malignant melanoma. A case report. Pathologica 1994; 86: 546 – 8. 43. Farrell DJ, et al. Primary malignant melanoma of the bronchus. Thorax 1996; 51: 223 – 4. 44. de Wilt JHW, et al. Isolated melanoma in the lung when there is no known primary site: metastatic disease or primary lung tumour? Melanoma Res. 2005; 15(6): 531 – 7. 45. Sekine I, et al. Rare pulmonary tumors – a review of 32 cases. Oncology 1998; 55: 431 – 4.
PRIMARY MELANOMA OF THE LUNG 46. Wilson RW, Moran CA. Primary melanoma of the lung: a clinicopathologic and immunohistochemical study of eight cases. Am J Surg Pathol 1997; 21: 1196 – 202. 47. Ost D, et al. Primary pulmonary melanoma: case report and literature review. Mayo Clin Proc 1999; 74: 62 – 6. 48. Ozdemir N, et al. Primary malignant melanoma of the lung in oculocutaneous albino patient. Eur J Cardiothorac Surg 2001; 20: 864 – 7. 49. Testini M, et al. Ileal intussusception due to intestinal metastases from primary malignant melanoma of the lung. Am Surg 2002; 68: 377 – 9. 50. Dountsis A, et al. Primary malignant melanoma of the lung: a case report. World J Surg Oncol 2003; 1: 26. 51. Lie CH, et al. Primary pulmonary malignant melanoma presenting with haemoptysis. Melanoma Res 2005; 15: 219 – 21. 52. Scotto J, Fraumeni JF, Lee JA Jr. Melanomas of the eye and other noncutaneous sites: epidemiologic aspects. J Natl Cancer Inst 1976; 56: 489 – 91. 53. Ross MI, Stern SJ. Mucosal melanomas. In Balch CM, et al. (eds) Cutaneous Melanoma, 3rd ed. St Louis, Missouri: Quality Medical Publishing, 1998: 195 – 206. 54. Carter D, Eggleston J. Tumors of the lower respiratory tract. In Atlas of Tumor Pathology, Series 2, Fascicle 17. Washington, District of Columbia: Armed Forces Institute of Pathology, 1979: 220. 55. Fowler M, Sutherland HDA. Malignant melanoma of the oesophagus. J Pathol Bacteriol 1952; 64: 473 – 7. 56. Goldman JL, et al. The presence of melanocytes in the human larynx. Laryngoscope 1972; 82: 824 – 35. 57. Busuttil A. Dendritic pigmented cells within human laryngeal mucosa. Arch Otolaryngol 1976; 102: 43 – 4. 58. Batsakis JG, et al. The pathology of head and neck tumors: mucosal melanomas, part 13. Head Neck Surg 1982; 4: 404 – 18. 59. Grazer R, et al. Melanin-containing peripheral carcinoid of the lung. Am J Surg Pathol 1982; 6: 73 – 8. 60. Rosenberg LM, Polanco GB, Blank S. Multiple tracheobronchial melanomas with ten-year survival. JAMA 1965; 192: 717 – 9. 61. Damian DL, et al. Positron emission tomography in the detection and management of metastatic melanoma. Melanoma Res 1996; 6: 325 – 9. 62. Gritters LS, et al. Initial assessment of positron emission tomography using 2-fluorine-18-fluoro-2-deoxy-D-glucose in the imaging of malignant melanoma. J Nucl Med 1993; 34: 1420 – 7. 63. Heenan PJ. Local recurrence of melanoma. Pathology 2004; 36: 491 – 5. 64. Littman CD. Metastatic melanoma mimicking primary bronchial melanoma. Histopathology 1991; 18: 561 – 3.
297
65. Faries MB, et al. Lymphatic mapping and sentinel lymphadenectomy for primary and metastatic pulmonary malignant neoplasms. Arch Surg 2004; 139: 870 – 6; discussion 876 – 7. 66. Boni R, et al. Staging of metastatic melanoma by whole-body positron emission tomography using 2-fluorine-18-fluoro-2-deoxy-D-glucose. Br J Dermatol 1995; 132: 556 – 62. 67. Rigo P, et al. Oncological applications of positron emission tomography with fluorine-18 fluorodeoxyglucose. Eur J Nucl Med 1996; 23: 1641 – 74. 68. Valk PE, et al. Cost-effectiveness of PET imaging in clinical oncology. Nucl Med Biol 1996; 23: 737 – 43. 69. Wagner JD, et al. Initial assessment of positron emission tomography for detection of nonpalpable regional lymphatic metastases in melanoma. J Surg Oncol 1997; 64: 181 – 9. 70. Gulec SA, et al. The role of fluorine-18 deoxyglucose positron emission tomography in the management of patients with metastatic melanoma: impact on surgical decision making. Clin Nucl Med 2003; 28: 961 – 5. 71. Dalrymple-Hay MJ, et al. Pulmonary metastatic melanoma – the survival benefit associated with positron emission tomography scanning. Eur J Cardiothorac Surg 2002; 21: 611 – 4; discussion 614 – 5. 72. La Hei ER, et al. Surgical resection of pulmonary metastatic melanoma: a review of 83 thoracotomies. Asia Pac Heart 1996; 5: 111 – 4. 73. Harpole DH Jr, et al. Analysis of 945 cases of pulmonary metastatic melanoma. J Thorac Cardiovasc Surg 1992; 103: 743 – 8; discussion 748 – 50. 74. Gorenstein LA, et al. Improved survival after resection of pulmonary metastases from malignant melanoma. Ann Thorac Surg 1991; 52: 204 – 10. 75. Dasgupta T, Bowden L, Berg JW. Malignant melanoma of unknown primary origin. Surg Gynecol Obstet 1963; 117: 341 – 5. 76. Giuliano AE, Moseley HS, Morton DL. Clinical aspects of unknown primary melanoma. Ann Surg 1980; 191: 98 – 104. 77. Reintgen DS. et al. Metastatic malignant melanoma with an unknown primary. Surg Gynecol Obstet 1983; 156: 335 – 40. 78. Norman J, et al. Metastatic melanoma with an unknown primary. Ann Plast Surg 1992; 28: 81 – 4. 79. Katz KA, et al. Melanoma of unknown primary: experience at Massachusetts General Hospital and Dana-Farber Cancer Institute. Melanoma Res 2005; 15: 77 – 82. 80. Vijuk G, Coates AS. Survival of patients with visceral metastatic melanoma from an occult primary lesion: a retrospective matched cohort study. Ann Oncol. 1998; 9: 419 – 22.
Section 5 : Thoracic Tumors
25
Large Cell Neuroendocrine Carcinoma William D. Travis, Lee M. Krug and Valerie Rusch
INTRODUCTION Large cell neuroendocrine carcinoma (LCNEC) of the lung was first described in 1991 as a form of high-grade non–small cell neuroendocrine carcinoma.1 It is part of a family of pulmonary neuroendocrine tumors that includes the low-grade typical carcinoid (TC), intermediate-grade atypical carcinoid (AC), the high-grade LCNEC and small cell lung carcinoma (SCLC).2,3 In the 2004 World Health Organization (WHO) classification, LCNEC is classified as a variant of large cell carcinoma (LCC). Within large cell carcinoma there are four major categories of neuroendocrine differentiation (NED) that can occur (see Table 1): (i) LCNEC which have neuroendocrine features by light microscopy as well as immunohistochemistry and/or electron microscopy, (ii) large cell carcinoma with neuroendocrine morphology (LCNEM) that have neuroendocrine morphology but lack neuroendocrine differentiation by electron microscopy or immunohistochemistry, (iii) large cell carcinomas with neuroendocrine differentiation (LCC-NED) that have no neuroendocrine pattern by light microscopy but show neuroendocrine differentiation by immunohistochemistry or electron microscopy and (iv) classic LCC that lacks both neuroendocrine morphology by light microscopy and neuroendocrine differentiation by immunohistochemistry or electron microscopy.1,4 In the past, LCNEC have been classified as a variety of other lung tumors including AC, the intermediate subtype of SCLC, large cell carcinoma and large cell neuroendocrine tumor.1,2 These tumors can also be confused with poorly differentiated adenocarcinomas, squamous cell carcinomas and basaloid carcinomas.2 These cases also frequently are confused with LCC or non–small cell carcinomas with neuroendocrine differentiation that lack neuroendocrine morphology.2 The concept of LCNEC has been widely accepted as reflected by the growing number of publications on this subject. Nevertheless, the literature has to be read critically because some authors have lumped the above subsets
and have also included combined small cell carcinoma/large cell carcinoma into a single category of LCNEC.5 – 8 This complicated our attempt to summarize some of the published data in Table 2. Until these distinct subsets of large cell carcinoma with neuroendocrine morphology (LCCNEM)/differentiation have been carefully characterized, it is premature to analyze them together. Because of the rarity of LCNEC, little is known about its clinical behavior and how these tumors should be treated. No single institution has sufficient cases to address this issue. This chapter will review the existing literature on LCNEC and summarize the limited knowledge about therapy. Brief mention will be made about the scant data regarding LCC-NEM. Large cell carcinoma or non–small cell lung carcinomas (NSCLC) with neuroendocrine differentiation will not be specifically addressed in this chapter. The formation of an International Registry of pulmonary neuroendocrine tumors intended to address the lack of knowledge about LCNEC is also presented.
BIOLOGY AND EPIDEMIOLOGY LCNEC accounts for about 3% of surgically resected lung cancers (range 1–5%, see Table 1).25 The vast majority of patients are cigarette smokers and most have over a 50 pack/year history of smoking.1 There is a strong male predominance that probably reflects a predominance of smoking in males compared to females (see Table 1). In the spectrum of neuroendocrine lung tumors, LCNEC express many molecular abnormalities similar to SCLC but there are considerable differences with TC and AC. Identification of molecular alterations in tumors has become an attractive way to identify novel therapeutic approaches. As we do not have effective therapies for LCNEC, understanding of the molecular changes in the entire spectrum of pulmonary neuroendocrine tumors is important. Since these tumors are so uncommon, there are few molecular studies that have examined large numbers of LCNEC. The larger studies usually include the entire spectrum of pulmonary neuroendocrine tumors so cases of TC, AC, and SCLC are included with
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
LARGE CELL NEUROENDOCRINE CARCINOMA Table 1 Neuroendocrine differentiation in large cell carcinoma.
Tumor name Large cell carcinoma (LCC) Large cell neuroendocrine carcinoma (LCNEC) Large cell carcinoma with neuroendocrine morphology (LCNEM) Large cell carcinoma with neuroendocrine differentiation (LCC-NED)
Neuroendocrine morphology
Neuroendocrine special studies: immunohistochemistry or electron microscopy
No
No
Yes
Yes
Yes
No
No
Yes
the LCNEC. Also since very few institutions have frozen tissue banks, most molecular studies of LCNEC have been retrospective studies performed on tissue samples that are formalin-fixed and paraffin-embedded, limiting the type of molecular studies that can be performed. As there are a large number of molecular pathways that have been studied in LCNEC mostly using small numbers of cases, we will focus on a few major studies in this summary. In LCNEC Onuki et al. demonstrated a high percentage of p53 abnormalities by immunohistochemical overexpression (75%), loss of heterozygosity (67%) and mutation analysis (59%).26 These findings were very similar to those seen in SCLC, but significantly higher than those seen in TC and AC.26 Similar findings have been demonstrated by others with p53 expression ranging between 40 and 86% and p53 mutations from 27 to 59%.27 – 32 Interestingly, Onuki et al. found that 58% of the point mutations found in highgrade neuroendocrine tumors were G:C to T:A or other transversions.26 The G:C to T:A transversions are associated with carcinogens found in cigarette smoke, consistent with the high frequency of heavy cigarette smoking in LCNEC patients.33 LCNEC have frequent abnormalities of the p16INK4 / cyclin D1/Rb pathway involved with regulation of G1 arrest in the cell cycle. Lack of Rb expression was found by Beasley et al. and Igarashi et al. in 49–68% of LCNEC, similar to SCLC (84–87%), but significantly more often than in TC (0%) and AC (0–21%).34,35 Cyclin D1 overexpression and loss of p16 staining were found in 9.5 – 32% and 18 – 22% of LCNEC, respectively.34,35 Igarashi also demonstrated overexpression of cyclin B1 in 84% of LCNEC and SCLC. Since an intact Rb pathway demonstrated p16 positive, cyclin D1 positive and Rb positive, in 78 to 90% of LCNEC, the Rb pathway is disrupted.34,35 The finding that virtually all cyclin D1 positive tumors are Rb positive indicates that cyclin D1 overexpression characteristically occurs only in the presence of intact Rb. Loss of Rb is the most frequent mechanism of Rb cell cycle pathway deregulation in LCNEC and SCLC. The frequent expression of cyclin B1 in high-grade neuroendocrine carcinomas is consistent with the concept that regulation of cyclin B1 expression and G2/M arrest are consistently compromised in LCNEC and SCLC.35
299
LCNEC have a high index (1.3 to 6.8%) of apoptosis compared to carcinoids that have a variable apoptotic index and SCLC that have almost no apoptosis.27 LCNEC and SCLC have a high Bcl-2/Bax ratio in contrast to TC and AC that have predominant Bax expression.27,36 These findings are consistent with the concept that the high rate of cell division in LCNEC could be worsened by abrogation of cell death which is favored by high Bcl-2 and low Bax levels leading to a short doubling time and tumor aggressiveness.27 c-KIT protein expression by immunohistochemistry was demonstrated in 55 and 61% of LCNEC by Araki and Casali, et al.17,37 Differing criteria for interpretation of results have been used with at least 50% of tumor cells with 2+ positive cytoplasmic and membrane staining of tumor cells used by Casali et al.17 while Pelosi et al. separated the results of membranous versus cytoplasmic patterns of c-KIT staining and found membrane immunoreactivity in 77% and cytoplasmic reactivity in 44% of LCNEC, using a cutoff of 5% or more immunoreactivity for a positive result.38 Casali et al. found a significantly worse prognosis (p = 0.046) as well as a higher rate of recurrence (0.037) for patients with c-KIT positive tumors.17 However, neither Pelosi et al. nor Araki, et al. found any prognostic significance to c-KIT expression in LCNEC or SCLC.37,38 In contrast to lung carcinoids, MEN 1 gene mutations are very rare in LCNEC. Debelenko et al. found a somatic frameshift in the MEN 1 gene (1226delC) in one of 13 tumors. This represented the first mutation observed outside the spectrum of neoplasms associated with MEN 1. On the other allele, neither a deletion nor mutation was detected and wild-type mRNA sequence was expressed, suggesting that the typical two-hit mechanism of MEN 1 gene inactivation did not occur.39
PATHOLOGY Gross Features Most LCNEC are peripheral (66–100%) with the remainder being centrally located.18,20,40 The average size is 3–4.0 cm with a range from 0.9 up to 12 cm.3,11,18,20,40 They are usually circumscribed nodular masses with a necrotic, tan, red cut surface. In larger tumors, the necrosis tends to be confluent and more extensive.20
Histologic Features The diagnostic criteria for LCNEC are (i) neuroendocrine morphology with organoid nesting, palisading, or rosettelike structures (see Figure 1), (ii) high mitotic rate greater than 10 mitoses 2 mm−2 (average 60–80 mitoses per 2 mm−2 ), (iii) non–small cell cytologic features including large cell size, low nuclear/cytoplasmic ratio, nucleoli, or vesicular chromatin, (see Figure 2) and (iv) neuroendocrine differentiation by immunohistochemistry (chromogranin, CD56 or synaptophysin) or electron microscopy (see Figure 3).2,3 Combined LCNEC consists of an LCNEC with components of adenocarcinoma, squamous cell carcinoma, giant cell carcinoma and/or spindle cell carcinoma.2,3 Most often this represents a component of adenocarcinoma, but squamous cell, giant cell or spindle cell carcinoma can be present.
300
THORACIC TUMORS
Figure 1 Large cell neuroendocrine carcinoma shows organoid nesting pattern with prominent palisading at the periphery of the tumor cell nests. A few rosettelike structures are present. Focal necrosis is seen.
Figure 3 The tumor cells strongly express CD56 with a membranous pattern of staining, indicating neuroendocrine differentiation.
pathologic characteristics of LCNEM are much different from LCNEC. The only major difference was reported by Iyoda et al. who found the mitotic rate of LCC-NEM was significantly higher than that of LCNEC.11
Immunohistochemistry/Electron microscopy
Figure 2 The tumor cells demonstrate rosettelike structures, abundant cytoplasm, nuclei with vesicular chromatin and scattered nucleoli. Several mitotic figures are present in this high-power field.
If there is a component of SCLC, the tumor becomes a combined SCLC and LCNEC. It is very difficult to make the diagnosis of LCNEC based on small biopsies or cytology (see section “Cytology” below). This is because of the problems in recognizing the neuroendocrine morphologic pattern and demonstrating neuroendocrine differentiation by immunohistochemistry in small pieces of tissue. Therefore, in the vast majority of cases a definite diagnosis of LCNEC and SCLC will require a surgical lung biopsy. Large Cell Carcinoma with Neuroendocrine Morphology
Except for the lack of neuroendocrine differentiation by immunohistochemistry, little data exists to suggest that the
The presence of neuroendocrine differentiation by immunohistochemistry or electron microscopy is required for the diagnosis of LCNEC.2,3 By immunohistochemistry, LCNEC stain positively with cytokeratin antibodies such as AE1/AE3 and CAM5.2.40,41 LCNEC stain immunohistochemically with chromogranin (55–82%), CD56/NCAM (Neural cell adhesion molecules) (73–100%), synaptophysin (40–91%), and pancytokeratin (100%).1,9,13,42 A panel of neuroendocrine markers is useful rather than staining with a single antibody. Currently the most useful antibodies include chromogranin, CD56 and synaptophysin. Thyroid transcription factor 1 (TTF-1) is positive in 41–75% of LCNEC.41,43,44 The adenocarcinoma component of combined LCNEC and adenocarcinomas is likely to express TTF-1.41 A variety of different criteria have been applied regarding the minimum positive staining for neuroendocrine markers. Most studies have required only one neuroendocrine marker (excluding neuron-specific enolase [NSE]) to be positive,13,21 but others have required two positive stains.9 Some have required as much as 10%21 or as little as a single definite positive cell13 In general, the distribution and intensity of immunohistochemical staining for neuroendocrine markers is less for LCNEC than it is for typical and AC tumors. However, neuroendocrine markers may be diffusely and strongly positive. Staining for neuroendocrine markers is often focal, since these are poorly differentiated tumors. Takei et al., found less than half of their cases stained diffusely.13 Takei found all three neuroendocrine markers, synaptophysin, NCAM (CD56), and chromogranin to be
LARGE CELL NEUROENDOCRINE CARCINOMA
301
Table 2 Large cell neuroendocrine carcinoma: clinical characteristics.
5-year survival (%)
Author
Year
n
% cases
Median Age
Male (%)
Smoking history %
Jiang9 Travis3 Garcia-Yuste10 Iyoda11 Mazieres12 Takei13 Hage14 Kakinuma15 Zacharias8 Ab’Saber16 Casali17 Daddi18 Doddoli19 Oshiro20 Paci21
1998 1998 2000 2001 2002 2002 2003 2003 2003 2004 2004 2004 2004 2004 2004
22 37 22 50 18 87 7 38 15 24 33 18 20 38 48
2.9 NA NA 2.4 1.6 3.1 1 4.1 NA 5 NA 1 NA NA 3.5
63 58 67 64 63 68 64 63 64 58 65 69 62 66 64
(51 – 77) (21 – 75) (47 – 70) (38 – 82) (49 – 78) (37 – 82) (54 – 77) (51 – 74) (45 – 79) (44 – 84) (42 – 80) (58 – 77) (43 – 80) (45 – 82) (39 – 81)
83 67 77 84 100 89 71 85 60 58 94 72 90 95 87
NA 100 NA NA 94 98 100 NA NA NA 97 94.4 100 100 87
44.8 27 21 35.3 NA 57 0 NA 52 13 monthsb 51 37.5 36 NA 21
Battafarano5 Filosso22 Kozuki23 Rossi24
2005 2005 2005 2006
45 18 12 83
2.1 1.2 NA 1.7
67(35 – 90)c 63 (48 – 71) 60 (43 – 77) 67 (41 – 89)
56c 61 92 88
NA 78 92 96
30.2 35 33 28
Overall
Stage I
Stage II
Stage III
Stage IV
NA 35 NA NA NA 67 NA NA 88a NA NA NA
NA 0 NA NA NA 75 NA
NA 0 NA NA NA 4 NA
NA 0 NA NA NA 0 NA NA
NA NA NA
NA NA NA
NA NA NA
NA IIB:18%
NA 0
NA NA
NA NA NA 23
NA NA NA 8
NA NA NA NA
NA 28a
54 NA IA:66% IB:10% NAc NA NA 33
25
Note: n, number of patients; NA, not applicable. a Lumped large cell neuroendocrine carcinoma and large cell carcinoma with neuroendocrine morphology together. b Median Survival. c lumped large cell neuroendocrine carcinoma, large cell carcinoma with neuroendocrine morphology, and large cell carcinoma with neuroendocrine differentiation together (personal communication).
positive in 68% of cases, 2 markers in 85% and at least one marker in 100% of cases.13 NSE is not regarded to be a reliable marker for neuroendocrine differentiation since it will stain up to 60% of non–small cell carcinomas.45 Hormonal markers such as adrenocorticotropic hormone (ACTH) and calcitonin are positive in only the minority of cases. A higher percentage of LCNEC stain for keratin and carcinoembryonic antigen (CEA), in contrast to TC and AC.1 By electron microscopy, LCNEC show dense-core granules and may have cytoplasmic lumina and/or desmosomes.1
Cytology It is very difficult to diagnose LCNEC based on cytology specimens; however, several groups have addressed this subject.15,46 – 50 The difficulty in making this diagnosis is reflected by the study of Kakinuma et al. where the diagnosis of LCNEC was not made in any of the 20 cases that they reported.15 Instead the most common diagnosis rendered was carcinoma, histologic type undetermined, or, less often, squamous cell carcinoma, AC, small cell carcinoma, and large cell carcinoma.15 Wiatrowska et al. found the most characteristic features were (i) a bloody, inflammatory and necrotic background, (ii) flattened three-dimensional clusters of large cells with peripheral palisading, and many single cells, (iii) cytologic characteristics including a moderate to high nuclear to cytoplasmic ratio, and nuclei that are large, oval/round to polygonal and show nucleoli in most cases.49 Mitoses were seen in only one-third of cases.49 Immunohistochemistry
for neuroendocrine markers was positive in 87% of cases using a panel of chromogranin (5/16 positive, 31%) and synaptophysin (12/16 positive, 75%). Kakinuma et al. reported bronchial brush cytology in 20 LCNEC and compared the findings with poorly differentiated adenocarcinomas, squamous cell carcinomas and small cell carcinomas.15 Characteristic features included: abundant necrotic debris (90%), large tumor cell size (90%), naked nuclei (90%) and nuclear streaking (90%). Nuclear features consisted of thin nuclear membranes, finely granular nuclear chromatin and one or a few nucleoli. Less than one half of the cases showed Indian filing and rosettelike structures. These features were less frequent in poorly differentiated adenocarcinomas and squamous cell carcinomas that more often showed thick nuclear membranes.15 Immunohistochemistry for neuroendocrine markers was less sensitive in cytology specimens compared to biopsy specimens. Hiroshima et al. reported cytology of 14 touch imprint and 11 curettage specimens from 25 histologically confirmed LCNECs.46 Their specimens were characterized by medium to large-sized, round, or polygonal tumor cells with nuclear pleomorphism. Naked nuclei were frequent. The nuclear chromatin was finely or coarsely granular. Most cases had one or two nucleoli, but they were inconspicuous in some cases. Tumor cells were in clusters with some rosettelike structures or single cells. A necrotic background and nuclear streaking were common. The diagnosis of LCNEC could be suspected in cases where the neuroendocrine morphologic pattern was identifiable.
302
THORACIC TUMORS
Differential Diagnosis LCNEC must be distinguished from AC, SCLC, large cell carcinoma, LCC-NED, LCC-NEM and classical large cell carcinoma (LCC) with no neuroendocrine features. Mitotic counts are one of the most useful criteria for distinguishing AC from LCNEC.1 According to the new WHO criteria for AC, the upper limit of mitoses should be 10 mitoses 2 mm−2 .2 This contrasts with LCNEC, which should have a mitotic count greater than 11 2 mm−2 but typically ranges between 50 and 100 mitoses per 10 high-power fields (hpf).2 Necrosis in LCNEC is generally more extensive than in AC where it usually consists of punctated foci within organoid nests of tumor cells. Nuclei of AC usually show a finely granular chromatin while most LCNEC have a vesicular or coarse chromatin. Separation of LCNEC from SCLC requires consideration of multiple histologic features such as cell size, nucleoli, chromatin pattern, and nuclear/cytoplasmic ratio, rather than a single criterion. Artifacts such as introduced by frozen sections, can distort cellular morphology resulting in confusion with SCLC. The difficulty in making this distinction is reflected in the retrospective review of a group of previously diagnosed SCLC where up to 44% of cases were reclassified as LCNEC.13,51 If a large cell carcinoma has no neuroendocrine pattern by light microscopy, but immunohistochemistry or electron microscopy demonstrates neuroendocrine features, the tumor is classified as LCC-NED. These cases are similar to the 10–15% of non–small cell lung carcinomas in which neuroendocrine differentiation (NSCLC-NED) can be found by electron microscopy and/or immunohistochemistry despite the absence of neuroendocrine morphology by light microscopy.25,52 – 66 All the published studies are retrospective and there is no consensus on whether these patients have a better or worse prognosis or if their tumors are more or less responsive to chemotherapy compared to NSCLC without neuroendocrine differentiation. Separation of LCNEC from LCC is based on whether or not a light microscopic neuroendocrine pattern is present. In most cases, this is not difficult because the neuroendocrine morphologic features are so distinctive. However, in some tumors the morphologic neuroendocrine features may be more subtle and separation from LCC or LCC-NE may be more difficult. Further data regarding the spectrum of clinical and pathologic features of LCNEC are awaiting larger series of cases. Hopefully this will further define the differences in survival and response to therapy for LCNEC compared to AC, LCNEM, LCC-NE, LCC, and SCLC.
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS In previously reported series (see Table 1), LCNEC patients have a median age of 62 years (range 33–87 years).1 Ectopic hormone production and paraneoplastic syndromes are uncommon.1 The most common presenting symptom is chest pain, followed by hemoptysis, dyspnea, cough, fever, and weight
loss. Up to 24% may be asymptomatic.8 Paraneoplastic syndromes are typically absent,10,13 a significant clinical difference with SCLC. Because surgical resection is usually required to establish the diagnosis, the stage distribution in most series is weighted towards early stage resectable tumors rather than advanced disease. Several studies have reported computed tomography (CT) findings of LCNEC.20,67,68 The largest study by Oshiro et al. included 28 cases where most tumors were in the lung periphery (84%) and the upper lobes (63%).20 Endobronchial growth was seen in 5% of cases, obstructive pneumonia in 8% and pleural effusion in 24%.20 With high resolution CT, the tumor-lung interface was well defined in 74% of cases and lobulation was present in 79%. Spiculation is reported between 32 and 73% of cases.20,67 Cavitation was rare (3%).20 The enhancement pattern on contrast-enhanced CT is more inhomogeneous in larger tumors.20 Serum tumor markers have been reported in a few series. Kozuki et al. found elevated serum levels of serum soluble fragments of cytokeratin 19 (CYFRA) in 67% of patients, NSE in 63%, lactate dehydrogenase (LDH) in 55%, SLX (sialyl Lewis X-i) in 50%, CEA in 42%, ProGRP in 33%, while no patient had elevations in Carbohydrate Antigen 19-9 (CA 19-9) or Squamous cell carcinoma (SCC) related antigen. All patients with elevated CYFRA, NSE, and ProGRP (pro-gastrin related polypeptide) had stage IV disease.23 Takei et al. found elevated serum CEA in 49% of patients while NSE and pro-gastrinreleasing peptide were elevated in 19 and 11% of patients, respectively.13
Large Cell Carcinoma with Neuroendocrine Morphology There is little information available regarding the clinical characteristics of patients with LCC-NEM.8,11,16 However, the data suggests that the clinical features such as age, gender predilection, smoking history, stage distribution and survival are very similar.8,11,16 Iyoda et al. demonstrated that LCCNEM had similar clinical properties to LCNEC with the exception that LCC-NEM had a more significantly elevated serum tissue polypeptide antigen (TPA) compared to LCNEC and a higher LDH than classic LCC.
TREATMENT Surgery The vast majority of patients with LCNEC have had surgical resections usually consisting of lobectomy or pneumonectomy. Smaller numbers of patients have only had wedge or segmental resections, sleeve resections, or excisional biopsies of metastatic lesions. Even fewer cases are diagnosed by small biopsies such as bronchoscopic or needle biopsies. Part of the reason for this is that the diagnosis of LCNEC is very difficult to establish in small specimens due to the requirement of identifying the neuroendocrine morphologic pattern as well as the need for immunohistochemistry to confirm neuroendocrine differentiation.
LARGE CELL NEUROENDOCRINE CARCINOMA
Zacharias et al. reported a series of patients with LCNEC where they performed a meticulous systematic nodal dissection in the majority of their patients. This careful approach to staging was suggested as a possible cause for the relatively favorable survival found in their stage I patients compared to other studies.8 Surgical complications do not appear to be increased or significantly different from those the operations performed for other non–small cell carcinomas. Mazieres et al. reported a 5% perioperative mortality.12 Most studies do not report major operative complications, but acute respiratory failure and hemothorax have been reported.12,19
Chemotherapy Little information exists regarding chemotherapy treatment for the entity of LCNEC. The controversy centers on whether to treat LCNEC, based on the neuroendocrine features, in the same manner as small cell lung cancer. Certainly the aggressive natural history and propensity to metastasize are similar to small cell lung cancer, yet LCNEC clearly does not demonstrate the same chemosensitivity. Previously published reports about LCNEC are not edifying in this regard since essentially all are retrospective series with a focus on patients who underwent surgery. Details regarding chemotherapy treatment, such as the agents used or the response rates, are generally not provided. Only small series, which are essentially case reports, include these specifics. For example, in the study by Mazieres et al., 13 patients with relapsed disease received chemotherapy with etoposide plus cisplatin or carboplatin.12 Partial responses were noted in 2 of the 10 evaluable patients. Kozuki and colleagues described seven patients treated with chemotherapy.23 Three patients with stage IIIB disease received concomitant radiation with cisplatin plus either etoposide or docetaxel and all had a partial response. The other four, treated with chemotherapy alone, received single agent vinorelbine, cisplatin plus etoposide, carboplatin plus gemcitabine, and cisplatin, docetaxel, and gemcitabine. No responses occurred among these patients. Salvage regimens including paclitaxel, docetaxel, irinotecan, gemcitabine, vinorelbine, or amrubicin were equally ineffective. Five patients were treated with gefitinib, an epidermal growth factor tyrosine kinase inhibitor, and one (a male with a 114 pack/year smoking history) achieved a partial response. The variability in regimens and the patient characteristics in these small studies make generalizations, regarding optimal chemotherapy for LCNEC, impossible to ascertain. What is apparent, however, is that response rates and survival are poor. Only recently has a prospective trial, selectively enrolling patients with LCNEC (a Japanese phase II trial of irinotecan and cisplatin), just begun. Although the diagnosis of LCNEC is rare which may necessitate multicenter participation, similar clinical trials of chemotherapy regimens and novel agents need to be conducted. Adjuvant Chemotherapy
In the rare cases of surgically resected small cell lung cancer, patients are generally treated with adjuvant chemotherapy based on historic reports of high relapse rates without
303
therapy and suggestions of benefit from older trials using outdated chemotherapy regimens.69 In non–small cell lung cancer, multiple large randomized trials have now confirmed that adjuvant chemotherapy after surgery confers a survival advantage.70,71 However, due to the limited number of cases, the benefit of chemotherapy after surgical resection has not been established specifically for LCNEC. A Japanese group attempted to discern the role of adjuvant chemotherapy in LCNEC by compiling data on 16 patients who received postoperative chemotherapy and 57 who did not.72 For most patients, the chemotherapy regimens administered included combinations typically used for small cell lung cancer, such as cisplatin or carboplatin with etoposide, or cyclophosphamide, doxorubicin, and vincristine. For all patients, the 5-year survival was 62% for stage I, 18% for stage II, and 17% for stage III. The authors note that the 5-year survival for the five patients with stage I disease who received adjuvant chemotherapy was 100% while it was 51% for the 23 patients who did not. Postoperative chemotherapy did not affect survival for other stages. An Italian retrospective series further supports the use of adjuvant chemotherapy for LCNEC. In their analysis of 83 cases, small cell-based adjuvant chemotherapy with a regimen such as platinum and etoposide was the most important variable correlating with survival in univariate and multivariate analysis. Interpretation of these reports is limited by the retrospective nature and the small sample size. However, these data, along with the known poor natural history and the routine use of adjuvant chemotherapy for SCLC and NSCLC, suggest that treatment with etoposide and cisplatin postoperatively is appropriate in LCNEC. Filosso and colleagues investigated whether adjuvant treatment with octreotide offers any benefit.22 Retrospectively, 18 patients were identified who had surgery for LCNEC. Ten patients had positive octreotide scans preoperatively and were given octreotide as adjuvant therapy. No patient received adjuvant chemotherapy, but patients with greater than stage IB disease received radiation. Of the nine patients that developed recurrent disease, only one had received octreotide. After recurrence, patients were treated with a platinum plus etoposide or paclitaxel but survival was only 8 months. Further studies of adjuvant octreotide in select patients may be warranted.
Radiation Radiation has been administered to a subset of patients in some studies,5,11,13,19,21 but there is insufficient information to establish whether this modality is effective or not. In some cases, it has been given only in a palliative setting.13 No studies have attempted to specifically determine the optimal approach to radiation and to document tumor responsiveness.
AUTHORS’ RECOMMENDATIONS LCNEC is an aggressive disease with distinct pathologic and biologic features shared by both SCLC and NSCLC. Since a surgical specimen is required to establish a definitive pathologic diagnosis in virtually all cases, most patients
304
THORACIC TUMORS
will undergo a surgical resection. Because of the aggressive behavior of this tumor, adjunctive chemotherapy or radiation will be given in many cases. Treatment with chemotherapy has marginal benefit. The optimal chemotherapy regimen for LCNEC remains unknown, but regimens used for small cell lung cancer, such as etoposide and cisplatin, seem to have the most data supporting their use. In line with standard therapy for both small cell and non–small cell lung cancer, patients with resected LCNEC should be offered adjuvant chemotherapy. Clinical trials designed specifically for patients with LCNEC are sorely needed to better define the role of chemotherapy and other treatment modalities. An effort is under way by the authors of this chapter to establish an International Registry of Pulmonary Neuroendocrine Tumors. Hopefully, through international collaborations the problems of rarity of this tumor and inconsistent approaches to therapy can be overcome so that optimal therapy can be determined in the future.
PROGNOSIS The survival for patients with LCNEC is poor. Overall 5-year survival ranges from 15 to 57% (see Table 1). The variation in survival is probably due to several factors including the distribution of lower vs higher stage or the thoroughness of the staging methods in the study. Given the lack of clear responses to chemotherapy or radiation, it is unlikely that survival has been due to variation in approach to treatment with these modalities. It is likely that the favorable survival data, stage for stage, in some series can be attributed to the careful approach to surgical staging such as systematic nodal dissection.8 However, even the studies with more favorable outcomes for LCNEC have not been able to demonstrate a significant difference in survival for SCLC treated in the same fashion. Several studies have demonstrated that survival for LCNEC is significantly worse than that for nonneuroendocrine non–small cell carcinomas. With such poorly differentiated and aggressive tumors, these survival differences are best observed in comparing survival of patients with lower stage tumors, because the higher stage patients die so rapidly there is little chance to see survival differences in a small series of patients. Jiang et al. found a 58.8 and 44.8% one and 5-year survival in 22 patients with a significantly worse prognosis than patients with other non–small cell carcinomas (p = 0.046).9 Iyoda also found that the survival for LCNEC was significantly worse than that for classical large cell carcinoma.11 Takei et al. found that in patients with only stage I tumors, survival for LCNEC, poorly differentiated NSCLC, and LCC were 67, 88 and 92% with significant differences between LCNEC and NSCLC (p = 0.003), LCNEC and LCC (p = 0.03) but not between LCNEC and SCLC.13 However, they found LCNEC to have no significant difference in survival compared with all stages of patients with poorly differentiated NSCLC, LCC, or SCLC. Of the 40% of patients who developed recurrence in the study by Takei et al., distant metastases occurred in
56% in the brain, liver, bone and lung; and locoregional recurrence occurred in 35% either in mediastinal or supraclavicular lymph nodes or in the bronchial stump.13 In 9% of cases, both distant and locoregional recurrences occurred simultaneously.13 Recurrence occurred within the first 6 months in 50% of patients, between 7 and 12 months in 32%, between 13 and 24 months in 9% and after 24 months in 9% of patients.13
REFERENCES 1. Travis WD, et al. Neuroendocrine tumors of the lung with proposed criteria for large-cell neuroendocrine carcinoma. An ultrastructural, immunohistochemical, and flow cytometric study of 35 cases. Am J Surg Pathol 1991; 15: 529 – 53. 2. Travis WD, et al. Pathology and Genetics: Tumours of the Lung, Pleura, Thymus and Heart. Lyon, France: IARC Press, 2004. 3. Travis WD, et al. Survival analysis of 200 pulmonary neuroendocrine tumors with clarification of criteria for atypical carcinoid and its separation from typical carcinoid. Am J Surg Pathol 1998; 22: 934 – 44. 4. Travis WD, et al. Collaboration with L.H. Sobin and pathologists from 14 countries. Histological Typing of Lung and Pleural Tumors. Berlin, Germany: Springer-Verlag, 1999. 5. Battafarano RJ, et al. Large cell neuroendocrine carcinoma: an aggressive form of non-small cell lung cancer. J Thorac Cardiovasc Surg 2005; 130: 166 – 72. 6. Dresler CM, et al. Clinical-pathologic analysis of 40 patients with large cell neuroendocrine carcinoma of the lung. Ann Thorac Surg 1997; 63: 180 – 5. 7. Rusch VW, Klimstra DS, Venkatraman ES. Molecular markers help characterize neuroendocrine lung tumors. Ann Thorac Surg 1996; 62: 798 – 809. 8. Zacharias J, et al. Large cell neuroendocrine carcinoma and large cell carcinomas with neuroendocrine morphology of the lung: prognosis after complete resection and systematic nodal dissection. Ann Thorac Surg 2003; 75: 348 – 52. 9. Jiang SX, et al. Large cell neuroendocrine carcinoma of the lung: a histologic and immunohistochemical study of 22 cases. Am J Surg Pathol 1998; 22: 526 – 37. 10. Garcia-Yuste M, et al. Prognostic factors in neuroendocrine lung tumors: a Spanish Multicenter Study. Spanish Multicenter Study of Neuroendocrine Tumors of the Lung of the Spanish Society of Pneumonology and Thoracic Surgery (EMETNE-SEPAR). Ann Thorac Surg 2000; 70: 258 – 63. 11. Iyoda A, et al. Clinical characterization of pulmonary large cell neuroendocrine carcinoma and large cell carcinoma with neuroendocrine morphology. Cancer 2001; 91: 1992 – 2000. 12. Mazieres J, et al. Large cell neuroendocrine carcinoma of the lung: pathological study and clinical outcome of 18 resected cases. Lung Cancer 2002; 37: 287 – 92. 13. Takei H, et al. Large cell neuroendocrine carcinoma of the lung: a clinicopathologic study of eighty-seven cases. J Thorac Cardiovasc Surg 2002; 124: 285 – 92. 14. Hage R, et al. Pulmonary large-cell neuroendocrine carcinoma (LCNEC). Eur J Cardiothorac Surg 2003; 23: 457 – 60. 15. Kakinuma H, et al. Diagnostic findings of bronchial brush cytology for pulmonary large cell neuroendocrine carcinomas: comparison with poorly differentiated adenocarcinomas, squamous cell carcinomas, and small cell carcinomas. Cancer 2003; 99: 247 – 54. 16. Ab’ Saber AM, et al. Neuroendocrine and biologic features of primary tumors and tissue in pulmonary large cell carcinomas. Ann Thorac Surg 2004; 77: 1883 – 90. 17. Casali C, et al. The prognostic role of c-kit protein expression in resected large cell neuroendocrine carcinoma of the lung. Ann Thorac Surg 2004; 77: 247 – 52. 18. Daddi N, et al. Surgical treatment of neuroendocrine tumors of the lung. Eur J Cardiothorac Surg 2004; 26: 813 – 7.
LARGE CELL NEUROENDOCRINE CARCINOMA 19. Doddoli C, et al. Large cell neuroendocrine carcinoma of the lung: an aggressive disease potentially treatable with surgery. Ann Thorac Surg 2004; 77: 1168 – 72. 20. Oshiro Y, et al. CT findings of surgically resected large cell neuroendocrine carcinoma of the lung in 38 patients. AJR Am J Roentgenol 2004; 182: 87 – 91. 21. Paci M, et al. Large cell neuroendocrine carcinoma of the lung: a 10year clinicopathologic retrospective study. Ann Thorac Surg 2004; 77: 1163 – 7. 22. Filosso PL, et al. Large-cell neuroendocrine carcinoma of the lung: A clinicopathologic study of eighteen cases and the efficacy of adjuvant treatment with octreotide. J Thorac Cardiovasc Surg 2005; 129: 819 – 24. 23. Kozuki T, et al. Complexity in the treatment of pulmonary large cell neuroendocrine carcinoma. J Cancer Res Clin Oncol 2005; 131: 147 – 51. 24. Rossi G, et al. The role of chemotherapy and receptor tyrosine kinases KIT, PDGFR-alpha, PDGFR-beta, and Met in large cell neuroendocrine carcinoma of the lung. J Clin Oncol 2005; 23: 8774 – 85. 25. Iyoda A, et al. Pulmonary large cell carcinomas with neuroendocrine features are high-grade neuroendocrine tumors. Ann Thorac Surg 2002; 73: 1049 – 54. 26. Onuki N, et al. Genetic changes in the spectrum of neuroendocrine lung tumors. Cancer 1999; 85: 600 – 7. 27. Brambilla E, et al. Apoptosis-related factors p53, Bcl2, and Bax in neuroendocrine lung tumors. Am J Pathol 1996; 149: 1941 – 52. 28. Hiroshima K, et al. Genetic alterations in early-stage pulmonary large cell neuroendocrine carcinoma. Cancer 2004; 100: 1190 – 8. 29. Iyoda A, et al. Pulmonary large cell neuroendocrine carcinoma demonstrates high proliferative activity. Ann Thorac Surg 2004; 77: 1891 – 5. 30. Jiang SX, et al. The significance of frequent and independent p53 and bcl-2 expression in large-cell neuroendocrine carcinomas of the lung. Mod Pathol 1999; 12: 362 – 9. 31. Przygodzki RM, et al. Analysis of p53, K-ras-2, and C-raf-1 in pulmonary neuroendocrine tumors. Correlation with histological subtype and clinical outcome. Am J Pathol 1996; 148: 1531 – 41. 32. Sampietro G, et al. Gene product immunophenotyping of neuroendocrine lung tumors. No linking evidence between carcinoids and small-cell lung carcinomas suggested by multivariate statistical analysis. Appl Immunohistochem Mol Morphol 2000; 8: 49 – 56. 33. Hollstein M, et al. p53 mutations in human cancers. Science 1991; 253: 49 – 53. 34. Beasley MB, et al. The P16/cyclin D1/Rb pathway in neuroendocrine tumors of the lung. Hum Pathol 2003; 34: 136 – 42. 35. Igarashi T, et al. Divergent cyclin B1 expression and Rb/p16/cyclin D1 pathway aberrations among pulmonary neuroendocrine tumors. Mod Pathol 2004; 17: 1259 – 67. 36. Kobayashi Y, et al. Molecular markers for reinforcement of histological subclassification of neuroendocrine lung tumors. Cancer Sci 2004; 95: 334 – 41. 37. Araki K, et al. Frequent overexpression of the c-kit protein in large cell neuroendocrine carcinoma of the lung. Lung Cancer 2003; 40: 173 – 80. 38. Pelosi G, et al. CD117 immunoreactivity in high-grade neuroendocrine tumors of the lung: a comparative study of 39 large-cell neuroendocrine carcinomas and 27 surgically resected small-cell carcinomas. Virchows Arch 2004; 445: 449 – 55. 39. Debelenko LV, et al. MEN1 gene mutation analysis of high-grade neuroendocrine lung carcinoma. Genes Chromosomes Cancer 2000; 28: 58 – 65. 40. Rossi G, et al. TTF-1, cytokeratin 7, 34betaE12, and CD56/NCAM immunostaining in the subclassification of large cell carcinomas of the lung. Am J Clin Pathol 2004; 122: 884 – 93. 41. Sturm N, et al. Thyroid transcription factor 1 and cytokeratins 1, 5, 10, 14 (34betaE12) expression in basaloid and large-cell neuroendocrine carcinomas of the lung. Hum Pathol 2001; 32: 918 – 25. 42. Lantuejoul S, et al. Neural cell adhesion molecules (NCAM) and NCAM-PSA expression in neuroendocrine lung tumors. Am J Surg Pathol 1998; 22: 1267 – 76. 43. Folpe AL, et al. Thyroid transcription factor-1: immunohistochemical evaluation in pulmonary neuroendocrine tumors. Mod Pathol 1999; 12: 5 – 8.
305
44. Sturm N, et al. Expression of thyroid transcription factor-1 in the spectrum of neuroendocrine cell lung proliferations with special interest in carcinoids. Hum Pathol 2002; 33: 175 – 82. 45. Said JW, et al. Immunoreactive neuron-specific enolase, bombesin, and chromogranin as markers for neuroendocrine lung tumors. Hum Pathol 1985; 16: 236 – 40. 46. Hiroshima K, et al. Cytological characteristics of pulmonary large cell neuroendocrine carcinoma. Lung Cancer 2005; 48: 331 – 7. 47. Iyoda A, et al. Imprint cytologic features of pulmonary large cell neuroendocrine carcinoma: Comparison with classic large cell carcinoma. Oncol Rep 2004; 11: 285 – 8. 48. Nicholson SA, Ryan MR. A review of cytologic findings in neuroendocrine carcinomas including carcinoid tumors with histologic correlation. Cancer 2000; 90: 148 – 61. 49. Wiatrowska BA, Krol J, Zakowski MF. Large-cell neuroendocrine carcinoma of the lung: proposed criteria for cytologic diagnosis. Diagn Cytopathol 2001; 24: 58 – 64. 50. Yang YJ, et al. Diagnosis of high-grade pulmonary neuroendocrine carcinoma by fine-needle aspiration biopsy: nonsmall-cell or small-cell type? Diagn Cytopathol 2001; 25: 292 – 300. 51. Travis WD, et al. Reproducibility of neuroendocrine lung tumor classification. Hum Pathol 1998; 29: 272 – 9. 52. Howe MC, et al. Neuroendocrine differentiation in non-small cell lung cancer and its relation to prognosis and therapy. Histopathology 2005; 46: 195 – 201. 53. Hiroshima K, et al. Prognostic significance of neuroendocrine differentiation in adenocarcinoma of the lung. Ann Thorac Surg 2002; 73: 1732 – 5. 54. Carnaghi C, et al. Clinical significance of neuroendocrine phenotype in non-small-cell lung cancer. Ann Oncol 2001; 12(Suppl 2): S119 – 23. 55. Baldi A, et al. Neuroendocrine differentiation in non-small cell lung carcinomas. In Vivo 2000; 14: 109 – 14. 56. Abbona G, et al. Chromogranin A gene expression in non-small cell lung carcinomas. J Pathol 1998; 186: 151 – 6. 57. Hage R, et al. Neural cell adhesion molecule expression: prognosis in 889 patients with resected non-small cell lung cancer. Chest 1998; 114: 1316 – 20. 58. Kwa HB, et al. Prognostic factors in resected non-small cell lung cancer: an immunohistochemical study of 39 cases. Lung Cancer 1996; 16: 35 – 45. 59. Schleusener JT, et al. Neuroendocrine differentiation is an independent prognostic factor in chemotherapy-treated nonsmall cell lung carcinoma. Cancer 1996; 77: 1284 – 91. 60. Senderovitz T, Skov BG, Hirsch FR. Neuroendocrine characteristics in malignant lung tumors: implications for diagnosis, treatment, and prognosis. Cancer Treat Res 1995; 72: 143 – 54. 61. Linnoila RI, Piantadosi S, Ruckdeschel JC. Impact of neuroendocrine differentiation in non-small cell lung cancer. The LCSG experience. Chest 1994; 106: 367S – 371S. 62. Carles J, et al. Neuroendocrine differentiation as a prognostic factor in non- small cell lung cancer. Lung Cancer 1993; 10: 209 – 19. 63. Pujol JL, et al. Neural cell adhesion molecule and prognosis of surgically resected lung cancer. Am Rev Respir Dis 1993; 148: 1071 – 5. 64. Skov BG, et al. Prognostic impact of histologic demonstration of chromogranin A and neuron specific enolase in pulmonary adenocarcinoma. Ann Oncol 1991; 2: 355 – 60. 65. Sundaresan V, et al. Neuroendocrine differentiation and clinical behaviour in non- small cell lung tumours. Br. J Cancer 1991; 64: 333 – 8. 66. Berendsen HH, et al. Clinical characterization of non-small-cell lung cancer tumors showing neuroendocrine differentiation features. J Clin Oncol 1989; 7: 1614 – 20. 67. Jung KJ, et al. Large cell neuroendocrine carcinoma of the lung: clinical, CT, and pathologic findings in 11 patients. J Thorac Imaging 2001; 16: 156 – 62. 68. Shin AR, et al. Large cell neuroendocrine carcinoma of the lung: radiologic and pathologic findings. J Comput Assist Tomogr 2000; 24: 567 – 73. 69. Murren J, Glatstein E, Pass HI. Small cell lung cancer. In DeVita VT, Hellman S, Rosenberg SA. (eds) Cancer, Principles and Practice of Oncology. Philadelphia, Pennsylvania: Lippincott, Williams & Wilkins, 2001: 983 – 1018.
306
THORACIC TUMORS
70. Arriagada R, et al. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med 2004; 350: 351 – 60. 71. Winton T, et al. Vinorelbine plus cisplatin vs. observation in resected non-small-cell lung cancer. N Engl J Med 2005; 352: 2589 – 97.
72. Iyoda A, et al. Adjuvant chemotherapy for large cell carcinoma with neuroendocrine features. Cancer 2001; 92: 1108 – 12.
Section 5 : Thoracic Tumors
26
Carcinoid Tumors of the Lung
Simon Chowdhury, Paul Cane, James F. Spicer and Peter G. Harper
INTRODUCTION Pulmonary carcinoid tumors are rare, malignant, neuroendocrine tumors. They used to be grouped under the misnomer “bronchial adenoma” with tumors of the bronchial glands such as adenoid cystic carcinoma and mucoepidermoid carcinoma. Carcinoid tumors are now thought to arise from or show differentiation toward the neuroendocrine cells found lining the bronchus (Kulchitski cells) and are therefore regarded as a type of neuroendocrine tumor. The physiological purpose of the bronchial neuroendocrine cells is still uncertain. They may play a role in lung development, as they are more prominent in the lungs of infants. Neuroendocrine cells are also noted to become more numerous in association with chronic pulmonary diseases.1
BIOLOGY AND EPIDEMIOLOGY Pulmonary carcinoids comprise 1–2% of all primary lung cancers,2,3 and approximately 25% of all carcinoid tumors are located in the respiratory tract,4,5 with an estimated annual incidence of 2.5 cases per one million population.6 There is an equal gender distribution and no proven link to smoking tobacco has been documented; however, several studies report that smoking may be a risk factor in the development of atypical carcinoids.7 – 9 The tumors occur over a wide age range, including the pediatric population, and the average age of incidence is 55 years.10 The current World Health Organisation classification of neuroendocrine tumors of the lung comprises typical and atypical carcinoid tumors, large cell neuroendocrine carcinoma and small cell carcinoma.11 It has been suggested that these tumors are all related, forming a spectrum ranging from the fairly indolent typical carcinoid to the very aggressive small cell carcinoma with atypical carcinoids and large cell neuroendocrine carcinomas falling in between. There is, however, growing molecular evidence to suggest that carcinoid tumors are a distinct tumor type from small cell and large cell neuroendocrine carcinomas. Comparative genomic hybridization has shown that a large proportion of typical and atypical carcinoid tumors show loss of part of the
long arm of chromosome 11.12 Carcinoids frequently possess mutations in the multiple endocrine neoplasia 1 (MEN 1 ) gene and absence of the gene product menin.13 These changes are not normally found in other lung tumors. In addition to abnormalities of the MEN 1 locus, atypical carcinoid tumors tend to possess further genetic abnormalities that are only rarely seen in typical carcinoids. Atypical carcinoids have been shown to possess loss of heterozygosity at 3p, 9p and 17p. These abnormalities may also be found in small cell carcinomas.14 In support of the theory that carcinoid tumors are a distinct tumor type from small cell carcinoma, gene expression profiling shows that carcinoid tumors appear more similar to neural tumors than either normal bronchial epithelium or small cell carcinoma.15
PATHOLOGY Carcinoid tumors may arise both centrally and peripherally in the lung. Two-thirds arise centrally, the typical appearance being a firm tumor mass growing into, and possibly obstructing, the lumen of a bronchus. Growth into the lung parenchyma adjacent to the bronchus produces a dumb-bell like appearance (see Figure 1). There may be secondary consolidation or bronchiectasis due to obstruction. One-third of tumors are peripheral, appearing as firm, circumscribed tan or yellow tumors within the lung parenchyma. Rarely, the tumor may contain melanin pigment rendering it as a black or brown mass (pigmented carcinoids). The Tumor-Node-Metastasis (TNM) staging system for non-small cell lung cancer (NSCLC) is used to stage pulmonary carcinoids as nodal involvement and distant metastasis have a large influence on the prognosis of these tumors. Atypical carcinoid tumors behave more aggressively than typical carcinoid tumors, metastasizing more commonly both to regional lymph nodes and systemically. A large retrospective analysis of 142 cases of pulmonary carcinoid (typical n = 128, atypical n = 14) showed nodal involvement and distant metastases occurred in 10 and 3% respectively in the typical group, and 57 and 21% in the atypical group.6 Both typical and atypical carcinoid tumors can metastasise to regional N1, N2 and N3 lymph nodes but atypical carcinoids
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
308
THORACIC TUMORS
of necrosis. Atypical carcinoids have 2–10 mitotic figures per 2 mm2 of microscopic section and/or possess areas of necrosis. Large cell neuroendocrine carcinoma or small cell carcinoma have, by definition, more than 10 mitotic figures per 2 mm2 and usually around 50 per 2 mm2 .11 It has been noted that the airways surrounding carcinoid tumors show increased numbers of intramucosal neuroendocrine cells. This condition is called diffuse idiopathic pulmonary neuroendocrine cell hyperplasia (DIPNECH). DIPNECH has also been noted in the absence of a tumor and has been postulated to be a preinvasive form of carcinoid tumor. Rarely, small tumorlike masses of neuroendocrine cells are found in the lungs that measure less than 5 mm in diameter. Such lesions are called carcinoid tumorlets rather than carcinoid tumors because of their very low rate of metastasis. They are often found incidentally and may be multiple.11
CLINICAL FEATURES
Figure 1 Central carcinoid tumor. The tumor is partly within the bronchus and partly in the adjacent lung producing a “dumb-bell” shape. A benign lymph node is seen below the tumor and early consolidation is present distal to the obstructed bronchus. (Inset: Microscopic appearance showing nested arrangement of cells with characteristic finely granular chromatin.)
more commonly involve N2 and contralateral mediastinal lymph nodes.6 The most common sites of distant metastases are liver, bone, brain, adrenal, and soft tissue.7 Microscopically, the tumors are typically comprised of fairly bland cells arranged in nests or chords (see Figure 1 inset). Other histological patterns may be found including solid, papillary, apocrine and rosette forming. The cell nuclei are round or oval with a characteristic finely granular chromatin pattern. The cells may be spindle shaped, particularly more in peripheral tumors. Although not typically found in cytological specimens such as bronchial brushings, washings or sputum because the tumor resides submucosally, fine-needle aspiration of lymph node metastases may yield cellular preparations with characteristic cytological appearances. Some tumors contain a population of sustentacular cells as seen in paraganglioid carcinoid tumors.16 Immunohistochemistry for neuroendocrine markers such as chromograffin and synaptophysin usually show strong staining of neurosecretory granules in the tumor cell’s cytoplasm. Thyroid transcription factor-1 (TTF-1) is a marker used to show that tumors are likely to have originated in the lung or thyroid. However, this marker may be positive or negative in lung carcinoids,17 and as TTF-1 may also be positive in neuroendocrine tumors from other sites, it has little value in these tumors. Typical carcinoids have, by definition, fewer than two mitotic figures per 2 mm2 of microscopic section and no areas
The majority of patients with pulmonary carcinoids are symptomatic with the proportion of patients with symptoms ranging from 52 to 92% in selected series.18,19 Patients commonly present with cough, hemoptysis, recurrent chest infections, chest discomfort or pain, wheezing, and shortness of breath. Symptoms from pulmonary carcinoids are notoriously nonspecific and often lead to delayed or misdiagnosis. In one series, 14.2% of patients had been treated for asthma for up to 3 years before the tumor was discovered.18 However, some patients are asymptomatic and a large Danish study showed that 24% of all typical carcinoids and 7% of all atypical carcinoids were found coincidentally at autopsy.20 The carcinoid “syndrome” is rare in patients with pulmonary carcinoids with an incidence of around 2%.5 The release of tumor vasoactive substances into the systemic circulation produces the characteristic symptoms of carcinoid syndrome such as diarrhea, flushing, palpitations, and wheezing. The flushing associated with pulmonary carcinoids has a greater tendency to cause diffuse body involvement than with other carcinoids, and after repeated flushing, patients may develop a constant red or cyanotic color. The severity of symptoms in patients experiencing the carcinoid syndrome is related to the location of the tumor (accessibility to the systemic circulation) and tumor mass. 5-hydroxytryptamine (5-HT) is the most commonly detected peptide responsible for the carcinoid syndrome in pulmonary carcinoids. In addition to the carcinoid syndrome, pulmonary carcinoids may be associated with ectopic production of adrenocorticotrophic hormone (ACTH) and growth hormone (GH)-releasing hormone causing Cushing’s syndrome and acromegaly respectively.21,22
DIAGNOSIS Computed tomography (CT) scanning and/or chest radiography (CXR) often reveals the primary lesion with a mean size of 3 cm although much larger lesions are also seen.23 Typical and atypical carcinoids have similar imaging features.
CARCINOID TUMORS OF THE LUNG
Central bronchial carcinoids are seen as an endobronchial nodule or hilar or perihilar mass with a close relationship to the bronchus. Calcification is common and is well visualized in CT. Associated atelectasis, air trapping, obstructing pneumonitis, and mucoid impaction may also be seen. Somatostatin receptors (SSRs), especially subtype 2 and 5, are located on the cell membranes of carcinoid tumors. Somatostatin analogues (for example, octreotide, lanreotide, and pentetreotide) have a high affinity for the SSR 2 and 5 subtypes. Radio-labelled somatostatin analogues, such as 111 In-pentetreotide, are potentially useful for imaging carcinoid tumors. The sensitivity of this technique is reported to range from 80 to 90%.24 However, specificity of this technique is low with “false-positives” in other tumors, granulomas, and autoimmune disease. This technique does provide useful information on tumor localization and is also predictive of response to octreotide therapy.25 Iodine-131-metaiodobenzylguanidine (131 I-MIBG, MIBG scan) is an analogue of a biogenic amine precursor and is taken up by chromaffin cells and stored in neurosecretory granules. 131 I-MIBG may be used for the imaging of neuroendocrine tumors. The sensitivity of 131 I-MIBG is reported to be slightly lower than 111 In-pentetreotide (70 vs 85%). However, a combination of these two scans increased sensitivity to 95%.24 Positron emission tomography (PET) scanning using 18 Flabelled fluorodeoxyglucose (18 FDG) is widely used in imaging a variety of malignancies particularly NSCLC. Increased FDG uptake in carcinoid tumors is limited owing to their low proliferative activity and high degree of differentiation. This has led to the development of several tracers directed toward the specific characteristics of carcinoid tumors, for example, 6-[18 F] fluorodopamine (18 F-dopa) and 11 C-labelled 5-HTP, for use in PET imaging. Promising initial results have been described in diagnosing small lesions and lymph node metastases.26 Apart from its role in the staging of carcinoid tumors PET may also have a role in follow-up and monitoring therapy. Despite characteristic features using the imaging techniques discussed above, biopsy is mandatory to provide definitive diagnosis and important prognostic information. Transbronchial biopsy is the usual means of obtaining a tissue diagnosis for the majority of tumors that are centrally located. For these tumors, especially when rigid bronchoscopy is used, a positive biopsy rate of almost 100% can be achieved. Bronchoscopy is not without complication and cases of carcinoid crisis after biopsy have been reported. In the past, major postbiopsy hemorrhage was feared but in a review of 587 bronchoscopic biopsies only four had hemorrhage requiring emergency surgery.27 Bronchoscopy is less effective for peripheral lesions and CT guided biopsy is needed for these cases.
MANAGEMENT It is important to differentiate between typical and atypical carcinoids because of their distinct natural histories. This is reflected in 5-year survival rates of resected patients found to have atypical carcinoids of 40–69% versus 87–100% for typical carcinoids.5
309
SURGERY Surgery remains the treatment of choice and offers the only chance of cure. In patients with central typical carcinoid tumors, bronchial sleeve resection or sleeve lobectomy should be considered. Local recurrence is rare and survival is excellent. In peripherally located typical carcinoid tumors, segmentectomy for patients with inadequate pulmonary function tests is indicated. Otherwise lobectomy is preferable to minimize the risk of recurrence. In cases of endobronchial localization of typical carcinoids, bronchoplastic parenchyma-sparing surgery is the standard surgical procedure. Intraoperative lymph node evaluation should be performed in all patients with typical carcinoid tumors and, if positive, should lead to a complete lymph node dissection. Patients with atypical carcinoids have a higher risk of relapse and thus a more extensive surgical approach, such as lobectomy or pneumonectomy with lymph node dissection, should be undertaken.5 Parenchyma-sparing surgery is not considered sufficient and the same surgical treatment as in NSCLC is advocated. Endobronchial laser treatment can be useful both to treat airway obstruction prior to surgery and also to look behind the tumor to assess the extent of surgery required.19 Endobronchial treatment alone cannot be considered curative as most carcinoids spread extraluminally but can be used as an alternative treatment option in patients unfit to undergo surgical resection. Such patients then require careful follow-up with high-resolution CT and endobronchial ultrasonography (EBUS). The excellent 5- and 10-year survival rates of typical pulmonary carcinoid tumors, with or without regional lymph node involvement, does not justify adjuvant treatment with either chemotherapy or radiotherapy. Atypical pulmonary carcinoid tumors have poorer 5- and 10-year survival rates and, although there are no randomized trials, adjuvant chemotherapy should be considered on an individual basis. The role and effectiveness of adjuvant radiotherapy also remains to be defined. In a study of patients with N2 disease, adjuvant radiotherapy was not shown to be beneficial.28
METASTATIC DISEASE The percentage of patients who develop metastatic disease is extremely variable depending on the series analyzed and histological subtype as already described. In a large series of 206 patients with resected stage I pulmonary carcinoids, Kaplan and colleagues demonstrated locoregional failures in 8% of patients at 5 years and distant metastases in 7%.29 Patients with stage I atypical carcinoids had a locoregional recurrence rate of 23% by 5 years and a rate of distant metastasis of 23%. Metastases may occur late up to decades after the initial diagnosis. Overall the prognosis of patients with metastatic pulmonary carcinoid tumors is poor with a 5-year survival of 22% from the start of treatment reported.30 Treatment for metastatic disease includes somatostatin analogues, interferons, radionucleotides, chemotherapy, combinations of these treatments, radiotherapy, and local treatments. However, because of the rarity of the tumor and the
310
THORACIC TUMORS
lack of prospectively designed studies, it is difficult to draw definitive conclusions about optimal treatment.
BIOTHERAPY Somatostatin interferes with the release of hormones and neurotransmitters through activation of membrane receptors. Its short half-life (2–4 minutes) limits its clinical application and has led to the development of somatostatin analogues such as octreotide and lanreotide, which have longer halflives. These drugs are now the drugs of choice to control the symptoms of patients with carcinoid syndrome. In addition to controlling symptoms, they have been assessed for their antitumor effects. In general they have a poor tumoricidal effect, decreasing tumor size in 0–17% of patients.31 However, both somatostatin analogues, have demonstrated a tumorostatic effect (30–100%)32 although no prospective study has shown that this results in increased survival. There is limited data on the use of somatostatin analogues for treating pulmonary carcinoids. In a subgroup from a recent study of patients with metastatic pulmonary carcinoids, four patients were treated with somatostatin analogues, all of whom developed progressive disease. One patient benefited with relief from carcinoid syndrome.30 Recently, several radioactive-labelled somatostatin analogues have been developed for therapeutic use in neuroendocrine tumors.33,34 The first reports of these studies are encouraging, with tumor reduction of around 30% and further studies are awaited. Interferon-α is now widely used in the treatment of metastatic carcinoid tumors, although its exact mechanism of action remains unknown. A reduction in tumor size occurs in a small number of patients (0–20%).31 However, like octreotide, interferon-α appears to have a tumorostatic effect (30–70% of patients) that may lead to prolonged survival. Because of their separate tumorostatic effects and potential synergy, interferon-α has been combined with somatostatin analogues. This combination has not been shown to be superior to the use of either agent alone in a randomized study treating metastatic neuroendocrine gastroenteropancreatic tumors.35 Interferon-α alone or in combination with other biotherapies has been evaluated in patients with metastatic pulmonary carcinoids.30 Twenty-seven patients (two received adjuvant interferon-α) were treated with interferon-α alone or in combination with octreotide and/or interferon-γ . Twentyone patients progressed both radiologically and biologically. In four patients, the disease was stable for a median of 15 months. There was no difference in the rate of stable disease among patients with or without carcinoid syndrome, and neither could a correlation be found between responses and any of the immunohistochemical analyses, including Ki67 expression.30 The effect of interferon-α and octreotide on tumor growth appears to be more limited in patients with pulmonary carcinoids than seen with carcinoids from other sites.
CHEMOTHERAPY There is no general consensus on when, or even if, chemotherapy should be started in patients with malignant
carcinoids including those of pulmonary origin. Chemotherapy for metastatic carcinoid tumors has in general been disappointing.31 The response rates to single agent chemotherapy, such as doxorubicin, 5-fluorouracil (5-FU), dacarbazine, cisplatin, etoposide, streptozotocin, and carboplatin are poor.31 Combination chemotherapies for pulmonary carcinoids are usually platinum- or streptozotocin-based. The Eastern Cooperative Oncology Group (ECOG) investigated combination chemotherapy in patients with carcinoid tumors.36,37 In the first of these studies, 118 patients were treated with either streptozotocin and 5-FU or streptozotocin and cyclophosphamide. The response rates were 21 and 24% respectively. Subgroup analysis revealed that two out of 17 (12%) with pulmonary carcinoid tumors responded to therapy. In a recently published ECOG study, 176 patients with advanced carcinoid tumors were randomized to either 5FU and streptozotocin or 5-FU and doxorubicin.37 There were no differences in the response rate (15.9 vs 16%); however, the 5-FU/streptozotocin combination was superior to 5-FU/doxorubicin for median survival (24.3 months vs 15.7 months; p = 0.0267). This led the authors to conclude that 5-FU/streptozotocin should be used when chemotherapy is judged to be an option for selected patients with carcinoid tumors. Twenty-two patients with pulmonary carcinoid tumors were enrolled but subgroup analysis was not performed. Streptozotocin combinations have been evaluated in patients with metastatic pulmonary carcinoids.30 Out of seven patients who received 5-FU/streptozotocin, only one had stable disease for 8 months. In all other patients the disease progressed. Streptozocin-doxorubin was administered to two patients, both of whom achieved radiologically stable disease at eight and 10 months respectively.30 While 5FU/streptozotocin has activity in carcinoid tumors, toxicities including cytopenias and nephrotoxicity are significant. This coupled with the apparent decreased response rates in pulmonary carcinoids has led some authors to conclude that streptozotocin-based therapy has no role in the treatment of pulmonary carcinoids.38 Because of the shared histopathologic features between small cell lung cancer (SCLC) and pulmonary carcinoids, the latter are often treated with small cell-based chemotherapy (i.e., platinum-based combinations) when surgery is not possible. However, response rates are not as high as that seen with SCLC. Wirth and colleagues treated 15 patients with pulmonary carcinoid tumors using etoposide and platinumbased regimens.38 The response rate to chemotherapy was 20% (3 out of 15 patients treated). In a recent study from Sweden, cisplatin and etoposide were used to treat eight patients. Two patients with typical carcinoids responded for 6 and 8 months respectively and one patient with atypical carcinoid had stable disease for 7 months.30 There was no correlation between response to treatment and Ki67 expression. The discrepancy in response to therapy suggests that pulmonary carcinoids are not as closely related to SCLC as previously thought. As stated earlier, it is now believed that carcinoid tumors are a distinct tumor type from small cell
CARCINOID TUMORS OF THE LUNG
carcinoma. Gene expression profiling shows that carcinoid tumors appear more similar to neural tumors than either normal bronchial epithelium or small cell carcinoma.15 The divergent pathophysiology between pulmonary carcinoids and SCLC may therefore explain the marked differences in response to chemotherapy and radiotherapy. The limitations of the above studies are clear. The majority of patients recruited into studies of treatment of metastatic carcinoid tumors have gastrointestinal rather than pulmonary carcinoids. Thus, the numbers of patients in studies are small and studies are often retrospective with variations in treatment regimens. Despite these limitations, it is clear that advanced pulmonary carcinoids can respond to chemotherapy used for SCLC. Several questions remain regarding the optimal nonsurgical management of this uncommon tumor. These include whether there are true differences in response rates between atypical and typical carcinoids; is there a role for adjuvant therapy; and what is the optimal treatment for patients with locally advanced unresectable disease? Given the clinical and molecular differences between pulmonary carcinoids and SCLC, further characterization of these tumors is needed with specific emphasis on identifying potential therapeutic targets for future clinical studies.
LOCAL TREATMENT OF METASTASES The local treatment of liver metastases of carcinoid tumors is attractive because of their slow growth and often localized growth pattern. Surgical resection is rarely feasible because metastases are usually diffuse at the time of diagnosis. Hepatic artery embolization may not only ameliorate symptoms but also reduce tumor burden. An objective or biochemical response rate of up to 50% and a median progression-free survival of 12 months have been reported in cases failing systemic therapy.39,40 Reports on chemoembolization show slightly better biochemical and tumor response rates. Side effects include liver pain, fever, renal toxicity and elevation of liver function tests. Radiofrequency ablation (RFA) can be used to treat nodules up to 4 cm in diameter. The use of RFA to treat metastases of carcinoid tumors has been reported in small series. Local tumor control has been reported in almost all patients, with a symptomatic and biochemical response in about 60–80%.41 However, local recurrence and development of new metastases is frequent. Radiotherapy may be used to provide local control at sites outside of the liver. It can palliate unresectable primary bronchial tumors as well as providing symptomatic benefit when used to treat brain and bone metastases.
CONCLUSION Pulmonary carcinoids are rare tumors. Localized disease should be treated with surgery including lymphadenectomy when necessary as this offers the only chance of cure. There are a number of therapeutic options available to treat metastatic disease. However, because of the rarity of the tumor and the lack of prospectively designed studies, it is difficult to draw definitive conclusions about its treatment. The
311
combination of platinum with etoposide provides a pragmatic treatment option, balancing efficacy with acceptable toxicity for many cases of metastatic disease. Decisions regarding treatment should be made in multidisciplinary meetings focused on “tailor-made” therapy based on the individual patient’s needs. Because pulmonary carcinoids are uncommon tumors, efforts should be made to treat these patients in specialized centers and for these centers to join together in multicenter studies.
REFERENCES 1. Gosney JR. Pulmonary neuroendocrine cell system in pediatric and adult lung disease. Microsc Res Tech 1997; 37(1): 107 – 13. 2. Harpole DH Jr, et al. Bronchial carcinoid tumors: a retrospective analysis of 126 patients. Ann Thorac Surg 1992; 54(1): 50 – 4; discussion 54 – 5. 3. Vadasz P, et al. Diagnosis and treatment of bronchial carcinoid tumors: clinical and pathological review of 120 operated patients. Eur J Cardiothorac Surg 1993; 7(1): 8 – 11. 4. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carcinoid tumors. Cancer 2003; 97(4): 934 – 59. 5. Hage R, et al. Update in pulmonary carcinoid tumors: a review article. Ann Surg Oncol 2003; 10(6): 697 – 704. 6. Fink G, et al. Pulmonary carcinoid: presentation, diagnosis, and outcome in 142 cases in Israel and review of 640 cases from the literature. Chest 2001; 119(6): 1647 – 51. 7. Beasley MB, et al. Pulmonary atypical carcinoid: predictors of survival in 106 cases. Hum Pathol 2000; 31(10): 1255 – 65. 8. Erasmus JJ, et al. Evaluation of primary pulmonary carcinoid tumors using FDG PET. AJR Am J Roentgenol 1998; 170(5): 1369 – 73. 9. Kayser K, et al. Carcinoid tumors of the lung: immuno- and ligandohistochemistry, analysis of integrated optical density, syntactic structure analysis, clinical data, and prognosis of patients treated surgically. J Surg Oncol 1996; 63(2): 99 – 106. 10. Hurt R, Bates M. Carcinoid tumours of the bronchus: a 33 year experience. Thorax 1984; 39(8): 617 – 23. 11. Travis WD, Colby TV, Corrin, B. WHO histological classification of tumors. Histological Typing of Lung and Pleural Tumors. Berlin, Germany: Springer-Verlag, 1999. 12. Walch AK, et al. Typical and atypical carcinoid tumors of the lung are characterized by 11q deletions as detected by comparative genomic hybridization. Am J Pathol 1998; 153(4): 1089 – 98. 13. Debelenko LV, et al. Identification of MEN1 gene mutations in sporadic carcinoid tumors of the lung. Hum Mol Genet 1997; 6(13): 2285 – 90. 14. Onuki N, et al. Genetic changes in the spectrum of neuroendocrine lung tumors. Cancer 1999; 85(3): 600 – 7. 15. Anbazhagan R, et al. Classification of small cell lung cancer and pulmonary carcinoid by gene expression profiles. Cancer Res 1999; 59(20): 5119 – 22. 16. Wick MR. Immunohistology of neuroendocrine and neuroectodermal tumors. Semin Diagn Pathol 2000; 17(3): 194 – 203. 17. Travis WDB, et al. World Health Organisation classification of tumors. Pathology and Genetics. Tumors of the Lung, Pleura, Thymus and Heart. Lyon, France: IARC Press, 2004. 18. Filosso PL, et al. Bronchial carcinoid tumors: surgical management and long-term outcome. J Thorac Cardiovasc Surg 2002; 123(2): 303 – 9. 19. Schreurs AJ, et al. A twenty-five-year follow-up of ninety-three resected typical carcinoid tumors of the lung. J Thorac Cardiovasc Surg 1992; 104(5): 1470 – 5. 20. Skuladottir H, et al. Pulmonary neuroendocrine tumors: incidence and prognosis of histological subtypes. A population-based study in Denmark. Lung Cancer 2002; 37(2): 127 – 35. 21. Aniszewski JP, et al. Cushing syndrome due to ectopic adrenocorticotropic hormone secretion. World J Surg 2001; 25(7): 934 – 40. 22. Melmed S, et al. Medical management of acromegaly due to ectopic production of growth hormone-releasing hormone by a carcinoid tumor. J Clin Endocrinol Metab 1988; 67(2): 395 – 9.
312
THORACIC TUMORS
23. Jeung MY, et al. Bronchial carcinoid tumors of the thorax: spectrum of radiologic findings. Radiographics 2002; 22(2): 351 – 65. 24. Taal BG, et al. Combined diagnostic imaging with 131I-metaiodobenzylguanidine and 111In-pentetreotide in carcinoid tumours. Eur J Cancer 1996; 32A(11): 1924 – 32. 25. Janson ET, et al. [111In-DTPA-D-Phe1]octreotide scintigraphy in patients with carcinoid tumours: the predictive value for somatostatin analogue treatment. Eur J Endocrinol 1994; 131(6): 577 – 81. 26. Hoegerle S, et al. Whole-body 18F dopa PET for detection of gastrointestinal carcinoid tumors. Radiology 2001; 220(2): 373 – 80. 27. Dusmet ME, McKneally MF. Pulmonary and thymic carcinoid tumors. World J Surg 1996; 20(2): 189 – 95. 28. Quaedvlieg PF, et al. Epidemiology and survival in patients with carcinoid disease in The Netherlands. An epidemiological study with 2391 patients. Ann Oncol 2001; 12(9): 1295 – 300. 29. Kaplan B, et al. Outcomes and patterns of failure in bronchial carcinoid tumors. Int J Radiat Oncol Biol Phys 2003; 55(1): 125 – 31. 30. Granberg D, et al. Experience in treatment of metastatic pulmonary carcinoid tumors. Ann Oncol 2001; 12(10): 1383 – 91. 31. Zuetenhorst JM, Taal BG. Metastatic carcinoid tumors: a clinical review. Oncologist 2005; 10(2): 123 – 31. 32. Oberg K. Carcinoid tumors: molecular genetics, tumor biology, and update of diagnosis and treatment. Curr Opin Oncol 2002; 14(1): 38 – 45. 33. Buscombe JR, Caplin ME, Hilson AJ. Long-term efficacy of highactivity 111In-pentetreotide therapy in patients with disseminated neuroendocrine tumors. J Nucl Med 2003; 44(1): 1 – 6.
34. de Jong M, et al. Radiolabelled peptides for tumour therapy: current status and future directions. Plenary lecture at the EANM 2002. Eur J Nucl Med Mol Imaging 2003; 30(3): 463 – 9. 35. Faiss S, et al. 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 2003; 21(14): 2689 – 96. 36. Moertel CG, Hanley JA. Combination chemotherapy trials in metastatic carcinoid tumor and the malignant carcinoid syndrome. Cancer Clin Trials 1979; 2(4): 327 – 34. 37. Sun W, et al. Phase II/III study of doxorubicin with fluorouracil compared with streptozocin with fluorouracil or dacarbazine in the treatment of advanced carcinoid tumors: Eastern Cooperative Oncology Group Study E1281. J Clin Oncol 2005; 23(22): 4897 – 904. 38. Wirth LJ, et al. Outcome of patients with pulmonary carcinoid tumors receiving chemotherapy or chemoradiotherapy. Lung Cancer 2004; 44(2): 213 – 20. 39. Eriksson BK, et al. Liver embolizations of patients with malignant neuroendocrine gastrointestinal tumors. Cancer 1998; 83(11): 2293 – 301. 40. Ruszniewski P, et al. Hepatic arterial chemoembolization in patients with liver metastases of endocrine tumors. A prospective phase II study in 24 patients. Cancer 1993; 71(8): 2624 – 30. 41. Hellman P, et al. Radiofrequency tissue ablation using cooled tip for liver metastases of endocrine tumors. World J Surg 2002; 26(8): 1052 – 6.
Section 5 : Thoracic Tumors
27
Bronchioloalveolar Carcinoma of the Lung Gregory J. Riely and Vincent A. Miller
HISTORICAL BACKGROUND
BIOLOGY AND EPIDEMIOLOGY Epidemiology
The entity now described as bronchioloalveolar carcinoma (synonyms include bronchiolar carcinoma, alveolar carcinoma, pulmonary adenomatosis, and bronchoalveolar carcinoma; hereforth referred to as BAC) was first described in the late nineteenth and early twentieth centuries (reviewed in Ref. 1). Malassez is credited with the first report of what has become multifocal BAC in 1876. In 1903, Musser was the first to characterize pneumonic BAC. The first application of the term BAC came in 1960 by Dr Averill A. Liebow.1 In his report, he described the pathology of patients with “peripheral pulmonary neoplasms that pursue a rather characteristic course when permitted to progress untreated”. Acknowledging some controversy in describing BAC, Liebow defined it as “well-differentiated adenocarcinomas primary in the periphery of the lung. . .with a tendency to spread chiefly within the confines of the lung by aerogenous and lymphatic routes, the walls of the distal air spaces often acting as supporting stroma for the neoplastic cells”. Since there were such widely varied definitions of BAC, he noted that the one finding that grouped these patients together was that patients with BAC were more likely to have spread within the lung, with many patients dying from respiratory failure without metastatic disease. For the first century of study, BAC was diagnosed using a variety of criteria. Some standardization was achieved with the 1981 WHO lung tumor classification, which designated BAC as an “adenocarcinoma in which cylindrical tumor cells grow upon the preexisting alveoli”.2 The more recent and specific definition of BAC comes from the 1999 WHO criteria (repeated again in 2004).2 These criteria define BAC as “an adenocarcinoma with a pure bronchioloalveolar growth pattern and no evidence of stromal, vascular, or pleural invasion”. Specific note is made in this classification system that the diagnosis of BAC cannot be made with cytologic material alone.
Estimates of the proportion of patients with non-small cell lung cancer (NSCLC) who have BAC range widely and depend heavily on the time period during which the epidemiologic study was performed, the pathologic definition of BAC used, and the means of detection of NSCLC (i.e. screen detected or symptom related). While evaluations of the Surveillance Epidemiology and End Results (SEER) population-based cancer registry have supported a relatively stable incidence of BAC, both in absolute number and as a proportion of patients with NSCLC, the uncertainty of what can be called BAC makes interpretation of these findings difficult.3 More recent studies, with pathologic reevaluation of lung adenocarcinomas compared with contemporary lung adenocarcinomas, suggest that the number of patients with BAC and the proportion of patients with lung adenocarcinomas who have BAC is rising.4 – 6 The only clearly identified risk factor for BAC is a history of cigarette smoking. While the odds ratio for BAC in patients with a smoking history is not as high as that found for either non-BAC adenocarcinomas or squamous cell carcinoma, smoking remains the only clear risk factor.7 An estimated 60–70% of patients with BAC are either former or current smokers.7 Since nearly 40% of patients with BAC are never smokers and have not had long-term exposure to mutagenic carcinogens, this population provides an ideal venue to study the molecular biology of lung cancer.
Biology To understand the biology of BAC, a number of animal models have been explored. Ovine pulmonary adenomatosis (OPA; also known as jaagsiekte and sheep pulmonary adenomatosis or ovine pulmonary carcinoma) is a widely studied model of BAC that was written about in Liebow’s initial description of BAC (reviewed in Refs 8 and 9). Similarities in sheep and human disease include: presence
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
314
THORACIC TUMORS
of peripheral, multifocal lesions, the production of large amounts of mucous, and tumors that originate from type II pneumocytes and Clara cells. By WHO criteria, OPA would be classified as mixed adenocarcinoma showing acinar, papillary, and bronchioloalveolar growth patterns. OPA is caused by a retrovirus (JSRV; the Jaagsiekte sheep retrovirus) and its transforming element was recently identified as the Env protein.10 Despite its histologic similarities, no virus has been isolated from human BAC. The human disease does not appear to be transmissible. Explorations of mechanisms of tumorigenesis of BAC have focused on atypical adenomatous hyperplasia (AAH) as a possible precursor lesion. AAH is “a localized proliferation of mild to moderately atypical cells lining involved alveoli. . .”.2 While etiologic and molecular evidence that AAH is a precursor of BAC and invasive adenocarcinoma is lacking, there is evidence that genetic changes become more prevalent with greater invasiveness of tumors. The best example of this is the identification of allelic loss corresponding to Noguchi classification.11 As described below, the Noguchi classification of tumors grades adenocarcinomas with BAC components into subgroups: type A (localized bronchioloalveolar carcinoma, LBAC), type B (LBAC with focal alveolar collapse), and type C (invasive adenocarcinoma with BAC). Aoyagi et al. determined the allelic loss of 66 tumors that had been classified as type A, B, or C.12 They found that allelic loss increased from 17% in type A tumors, to 39% in type B tumors, and 96% in type C tumors. More recently, Yoshida et al. noted that only 3% (1/35) patients with AAH had activating mutations in epidermal growth factor receptor (EGFR), while 11% (4/37) patients with BAC had such mutations.13 EGFR mutations were seen in 42% (13/31) of invasive adenocarcinomas. In contrast, 27% (8/30) of AAH had K-RAS mutations, whereas 16% (5/30) of BAC tumors and 10% (3/30) of invasive adenocarcinomas had K-RAS mutations. These data support a possible role for BAC as an intervening stage of tumorigenesis between AAH and invasive adenocarcinoma. In such a model, EGFR mutations arise later in tumorigenesis while K-RAS mutations occur relatively early. Another oncogene, SCCRO (squamous cell carcinoma related oncogene), also occurs as a relatively late lesion, with SCCRO expression detected by immunohistochemistry uncommon in pure BAC, but found in 29% of adenocarcinomas.14 Similar to our understanding of the molecular genetics of NSCLC in general, the molecular genetics of BAC tumors appears to be quite heterogeneous, with no identified singlegene mutation leading to this phenotype. The most commonly identified somatic mutations described in lung cancer are those in K-RAS, with approximately 25% of patients with NSCLC having K-RAS mutations. In looking specifically at BAC patients, Rusch et al. found K-RAS mutations in 10% (2/20) of BAC tumors (both occurred in smokers with codon 12 mutations).15 Separately, Marchetti et al. looked at 58 BAC tumors (10 mucinous BAC, 40 nonmucinous, and 8 with “sclerosing” BAC).16 They found that 36% of BAC tumors had K-RAS mutations. In contrast, just 26% of adenocarcinomas had evidence of K-RAS mutations. Of the 10 mucinous BACs evaluated, all had mutations in K-RAS,
whereas only 23% (9/40) of nonmucinous BAC had K-RAS mutations. On the basis of clinical responses to erlotinib and gefitinib in patients with adenocarcinoma and BAC, recent studies have focused on mutations in EGFR. EGFR is a receptor tyrosine kinase which, upon dimerization, leads to activation of several intracellular kinases leading to cell survival and proliferation. In studying patients with radiographic responses to gefitinib, Lynch et al. examined EGFR sequence from nine patients with known sensitivity to gefitinib, four of whom had BAC.17 They found that three of the four patients with BAC who responded to gefitinib had mutations in EGFR while none of seven patients without response had mutations. Of a total of eight tumors with mutations identified (an additional five were found in patients with invasive adenocarcinomas) four tumors had exon 19 deletions while two tumors had point mutations leading to substitution of a leucine for an arginine at position 858 (L858R), and one patient each had a point mutation leading to substitution of a glycine with a cysteine at position 719 (G719C) and leucine for glutamine at position 861 (L861Q). Separately, Paez et al. sequenced exons 2–25 of EGFR in 119 primary NSCLC obtained from patients in the United States and Japan.18 They identified mutations or deletions in the EGFR kinase domain in 16 tumors. The authors noted similarities between patients with mutations and the clinical profile of patients sensitive to gefitinib.19 They then sequenced the tumor EGFR from five patients with sensitivity to gefitinib (one of whom had BAC) and four patients who had progressive disease while on gefitinib. All tumors from the five patients with response had somatic mutations in the EGFR tyrosine kinase domain, while no mutations were identified in the four patients who did not respond to gefitinib. Finally, Pao et al. showed that NSCLC patients with radiographic and clinical response to erlotinib had similar, characteristic mutations in EGFR.20 Examining seven tumors from patients with sensitivity to erlotinib, five harbored mutations in EGFR (three with L858R mutations and two with exon 19 deletions). All of these tumors with EGFR mutations were adenocarcinomas with BAC features. In looking at never smokers with adenocarcinoma, the rate of mutations was 7/15 (47%) while in former or current smokers, the rate was significantly lower 4/81 (5%). The mechanisms of EGFR mutation-mediated oncogenesis are currently being explored. In assays with small interfering RNA and drug inhibition experiments, cell lines with EGFR mutations appear to be dependent on signaling through EGFR.21,22 More recently, Greulich et al. were able to demonstrate ligand-independent transformation by mutant EGFR using retroviral transformation of NIH 3T3 cells.23 In these mutant EGFR transformed cells, the EGFR kinase is constitutively active with intact downstream signaling through Shc, STAT3, and AKT. Additional studies to better understand mutant EGFR kinase activity are ongoing.
PATHOLOGY The pathology of BAC has been controversial from the time of the original description of this disease. The greatest
BRONCHIOLOALVEOLAR CARCINOMA OF THE LUNG
315
Table 1 Systems of classification of bronchioloalveolar carcinoma and adenocarcinomas.
Noguchi11
Ebright36
Terasaki35
Higashiyama34
WHO2
Type A – localized BAC (LBAC) Type B – LBAC with foci of alveolar collapse
BAC BAC with focal invasion
100% BAC >50% BAC
Type C – LBAC with foci of fibroblast proliferation
Adenocarcinoma with >15% BAC features Adenocarcinoma – –
BAC Adenocarcinoma with BAC Adenocarcinoma without BAC – – –
BAC Adenocarcinoma mixed subtype Adenocarcinoma
Type D – poorly differentiated adenocarcinoma Type E – tubular adenocarcinoma Type F – papillary adenocarcinoma
disagreements arise over what histologic patterns should be correctly be called BAC, whether evidence of an invasive adenocarcinoma component precludes the diagnosis of BAC, and the material required to make the diagnosis (i.e. cytology vs histologic section). The strictest criteria for diagnosis of BAC have emanated from the WHO lung tumor classification system. In 1999 (and repeated in 2004) these guidelines defined BAC as “growth of neoplastic cells along preexisting alveolar structures (lepidic growth) without evidence of stromal, vascular, or pleural invasion”.2 These guidelines specifically preclude diagnosis of BAC based upon cytologic specimens, stating that “the diagnosis of BAC requires thorough histologic evaluation to exclude the presence of invasive growth”. The clinical utility of the WHO guidelines is limited by the fact that the majority of advanced lung cancers are diagnosed based solely upon cytology. While some pathologists have great confidence in a diagnosis of BAC based upon cytology, systematic studies have failed to agree upon specific cytologic features that can make the diagnosis.24 – 33 Some features that have been suggested to support a cytologic diagnosis of BAC include absence of three-dimensional clusters, neoplastic cells in flat sheets, orderly arrangement of cells with round uniform nuclei, predominance of mucinous cells, absence of nuclear overlap, absence of irregular nuclear membranes, fine granular chromatin, and presence of nuclear grooves.25 Since the criteria proposed by the WHO do not correlate with clinical outcomes, some investigators have developed more descriptive categories of BAC. Noguchi et al. studied 236 small peripheral adenocarcinomas and subdivided them into six categories (see Table 1).11 Of the BAC subtypes (types A –C), patients with type A and type B tumors had a 100% 5-year survival. Patients with types A –C tumors had an improved survival when compared with patients with types D, E, or F. Higashiyama subclassified BAC by examining 206 peripheral lung adenocarcinomas and classifying them based on the proportion of BAC (see Table 1).34 Similar to Noguchi’s series, among 17 patients with 100% BAC (type IV) none had nodal involvement and 5-year overall survival was 100%. Ebright et al. evaluated 100 resected tumors with some component of BAC and classified them based on the degree of invasive adenocarcinoma.34 They classified tumors into four groups: pure BAC, bronchioloalveolar carcinoma with focal invasion (BWFI), adenocarcinoma with >15% bronchioloalveolar features (AWBF), and adenocarcinoma without evidence of BAC features. Overall survival was improved for patients with any component of BAC,
<50% BAC 0% BAC – –
– – –
emphasizing the importance of examining and subclassifying all NSCLC containing any element of BAC. The largest pathologic review performed examined 484 adenocarcinomas and divided them into BAC, adenocarcinoma with BAC, and adenocarcinoma without BAC.35 Patients with pure BAC had no vascular, lymphatic, or pleural invasion. While there was more invasive disease in the adenocarcinoma with BAC features group (5.5% vascular invasion, 14.8% lymphatic invasion, and 1.9% pleural invasion), the adenocarcinoma group without BAC showed the most invasion with 84.9, 61.4, and 60.8% of tumors with vascular, lymphatic, and pleural invasion respectively. Unfortunately, no data was provided about clinical outcomes. There is no consensus on the systematic description of mixed subtype adenocarcinomas. Until agreement about definitions can be reached, the clinical significance of varying proportions of BAC remains unknown. As therapeutic approaches are developed that specifically target lung adenocarcinomas with or without BAC features, it has become clear that pathology reporting of lung tumors that contain mixed adenocarcinoma and BAC components should describe the proportion of BAC component within adenocarcinomas. Currently, at our institution, all NSCLC pathology reports indicate the type and proportion of any subtypes of adenocarcinoma, including papillary, BAC, and others.
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS The clinical presentation of patients with BAC is similar to patients with other NSCLC tumors. The most common symptoms (see Table 2) at the time of presentation are cough, hemoptysis, chest pain, dyspnea, weight loss, and bronchorrhea.37 – 41 A symptom that is relatively unique to BAC is bronchorrhea, the production of large amounts of sputum, which is often quite bothersome to the patient. The majority of retrospective reviews have suggested that Table 2 Presenting symptoms in bronchioloalveolar carcinoma.
Symptoms Asymptomatic Cough Dyspnea Hemoptysis Chest pain Pneumonia Bronchorrhea
Frequency37 – 41 (%) 32 – 71 20 – 42 4 – 27 9 – 17 6 – 26 2 – 19 3 – 11
316
THORACIC TUMORS
patients with BAC are more likely than patients with other forms of adenocarcinoma to be women and to have never smoked cigarettes.37 – 47 The most comprehensive retrospective report to date described the experience of the Lung Cancer Study Group (LCSG) from 1977 to 1988. They identified 235 patients with BAC histology and compared their clinical characteristics with those of 710 patients with other adenocarcinomas and 608 patients with squamous cell tumors.43 While patients with BAC were more likely than patients with squamous cell tumors to be women (42 vs 14%, p < 0.001), the difference when compared to other adenocarcinomas was less striking (42 vs 37%). Patients with BAC were more likely to present at an earlier stage (with smaller tumors and less frequent nodal involvement), with no history of smoking, and with no weight loss or weight loss <10%. The median age at diagnosis is not significantly different from patients with other forms of adenocarcinoma. Just as patients with BAC have diverse clinical characteristics, the pattern of disease that is observed in such patients can vary widely. The pattern of disease can be described as multifocal, pneumonic, or nodular. Multifocal disease refers to multiple sites of BAC within the lung. Pneumonic BAC has a consolidative radiographic appearance and is often bilateral. Nodular BAC is a multifocal process, which has the radiographic appearance of multiple nodules.
RADIOLOGY Given the potential importance of distinguishing and classifying BAC from other adenocarcinomas, a number of investigators have attempted to correlate radiographic features with pathologic findings at the time of surgery. Most work has focused on small peripheral tumors where the goal is to differentiate pure BAC from invasive adenocarcinoma. With more frequent high resolution computed tomography (CT) screening, it became apparent that patients with BAC had unique radiographic findings, most commonly ground glass opacities (GGOs).48 In multifocal BAC, high resolution CT can reveal a variety of patterns including GGOs, areas of consolidation, centrilobular and peripheral nodules, and air bronchograms.49 Kim et al. studied 224 resected small peripheral adenocarcinomas. In correlating CT scan findings with the pathology of the tumors by Noguchi classification, they noted a correlation between the Noguchi classification and the mean proportion of GGO. Noguchi type A tumors had a mean proportion of GGO of 49% while Noguchi type C tumors had a mean of 23% GGO and Noguchi E tumors had a mean of 8% GGO.50 Similarly, others have demonstrated that the higher proportion of GGO on CT images of the tumor, the more likely it is to represent pure BAC.51 Quantification of the proportion of GGO is difficult to reproduce and would benefit from development of standardized criteria and methodology. Nakata et al. used NIH image software to reproducibly determine the percentage of GGO on high resolution CT scans of tumors <3 cm.52 By doing this, they found that among lesions with more than 90% GGO component, 87% had pure BAC as defined by the WHO, while in lesions with less than 50% GGO in the tumor, only
5% were pure BAC. In the intermediate range of 50–89%, 44% (20/45) tumors were pure BAC. Nomori et al. compared the Hounsfield units histograms of CT images of 10 patients with nonmucinous BAC and 9 patients with AAH, arguing that AAH could be discerned from BAC based upon this relatively simple measurement.53 Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) imaging of BAC has only recently been explored. Initial reports all described a relatively low FDG avidity of BAC, with many lesions showing no uptake.54 – 56 Further subclassification of patients with BAC revealed that the greater percentage of BAC within a tumor, the more likely that nodule will not show FDG avidity.57 In a recent series of 22 peripheral tumors found as part of a screening project, of 7 tumors that were not FDG avid, 4 were BAC. Similarly, in a study of 192 early stage lung tumors, nine tumors were found to be PET negative, and four of these tumors were BAC.58 PET scanning of patients with known pure BAC demonstrated have found that 42–63% of tumors are FDG avid compared with 97% of patients with NSCLC.56,59 – 61 Some investigators are exploring other tracers that might be more reliably imaged in BAC.
TREATMENT Surgery Patients with early stage disease are generally offered surgical resection of the tumor. Some controversy exists over the procedure of choice for patients with a unifocal presentation of BAC. Most early stage NSCLC has been treated with lobectomy based upon the LCSG trial that showed a higher rate of recurrence for NSCLC patients treated with wedge resection as compared with those who had a lobectomy.62 Given the propensity for some BAC tumors to present at a relatively early stage and the decreased likelihood of lymph node involvement, wedge resections may be appropriate for some BAC patients. In a retrospective review of 33 patients with stage I BAC, no significant difference in survival was noted for patients treated with wedge or lobectomy.42 A prospective evaluation of the approach to patients with small, peripheral adenocarcinomas with features of BAC (including GGOs) is necessary. A second controversy in the management of BAC surrounds patients with multifocal BAC. BAC patients frequently present with multiple nodules without lymph node involvement or develop metachronous lesions in short periods of time. While these patients might be classified as having stage IV disease and therefore not approached surgically, these nodules could also represent multiple primary tumors.63 Thus, some groups have explored surgical resection in this setting. While Daly et al. reported a dismal 5-year survival for patients with bilateral multifocal BAC who underwent resection, a more recent series, which included both BAC and adenocarcinoma with BAC features, found that there was no significant difference in survival between patients with multifocal and unifocal disease.36,64 These data suggest that if, after rigorous staging evaluation, a multifocal BAC appears to represent individual primary tumors, surgical resection is a reasonable approach.
BRONCHIOLOALVEOLAR CARCINOMA OF THE LUNG
A more radical approach to management of multifocal BAC is lung transplantation.65,66 De Perrot et al. did a worldwide survey of lung transplantations and identified a total of 26 patients who had lung transplantation for the diagnosis of multifocal BAC.65 They found that, of 22 patients who survived the postoperative period, 13 had disease recurrence at a median of 12 months, with 9 patients dying at a median of 22 months after transplantation. Similarly, Zorn et al. reported a prospective study of lung transplantation in patients with multifocal BAC. Among eight patients who underwent transplantation, six had disease recurrence and four died of BAC.66 Disease recurrence in patients with BAC who have undergone transplantation is thought to be because of recurrence of host tumor and not related to new BAC forming from donor lung tissue.67 Lung transplantation remains an investigative approach for the treatment of BAC.
Systemic Therapy Knowledge about systemic therapy for BAC has largely relied upon single institution retrospective reviews and subset analyses of larger trials. Feldman et al. performed a retrospective examination of response to chemotherapy in patients with BAC treated at the Mayo Clinic from 1975 to 1985.68 They identified 25 patients with BAC and compared those with 223 patients with other adenocarcinoma subtypes. There was no difference in response rate to chemotherapy, time to progression, or overall survival. Similarly, Breathnach et al. collected information on 28 patients with BAC and compared their course with 124 patients with other forms of NSCLC.69 BAC patients treated with systemic chemotherapy (most with cisplatin based combination chemotherapy regimens) had a median survival of 12 months as compared with 8 months for patients other forms of NSCLC or 10 months for patients with other adenocarcinomas. The 1-year survival for patients with BAC was 48%. A Japanese retrospective identified 16 patients with BAC and 70 control patients with other variants of NSCLC.70 They noted a 50% response rate to chemotherapy for BAC patients compared with a 27% response rate in controls (p = 0.076). The median survival for patients with BAC was 10 months compared with 8 months for controls (p = 0.025). Subset analysis of Eastern Cooperative Oncology Group (ECOG) 1594, a randomized trial of a four platinum based doublet regimens, identified just 17 patients with BAC, though no specific pathologic criteria were reported.71 Of the 17 patients, the response rate was 6% compared to the overall response rate of 20%. The median survival for this group was 12 months. The first prospective trial exclusively for patients with BAC was Southwest Oncology Group (SWOG) 9714.72 This trial enrolled 58 patients from 1997 to 2000 and treated patients with a 96-hour infusion of paclitaxel every 21 days. The definition of BAC was that chosen by local pathologists, but subsequent central pathology review was performed. Three patients enrolled and treated were found to have adenocarcinoma without any BAC features. Further correlation between hospital pathology and central pathology review was not reported. The response rate for this treatment was 14%, with 40% of patients having stable disease (see
317
Table 3 Trials of systemic therapy in advanced/metastatic BAC.
Study
Agent
S9714
Paclitaxel 96 hours Paclitaxel 3 hours Gefitinib 500 mg Erlotinib 150 mg
E08956 S0126 Miller
Response rate (%)
Median progression-free survival (months)
Median overall survival (months)
14
5
12
11 17 24
2.2 ∼3.5 –
NR ∼12 >12
Table 3). The median progression-free survival was 5 months and the median overall survival was 12 months. Survival at three years was 13%. Two-thirds of patients had one grade 3 toxicity, with 43% having grades 3 or 4 neutropenia. Patients with mucinous BAC (n = 16) had a median survival of 17 months as compared with 9 months for nonmucinous BAC and 18 months for adenocarcinoma with BAC features (n = 17). A second trial of paclitaxel in BAC was undertaken by the European Organization for Research and Treatment of Cancer (EORTC).73 Enrollment in this trial of BAC was limited to patients who fulfilled the following criteria: they did not have a primary adenocarcinoma elsewhere, had a peripheral tumor within the lung parenchyma, histologic appearance with cells lining the alveolar septa with preservation of basic pulmonary architecture. Owing to a poor response to treatment, EORTC 08956 was discontinued after only 19 patients had been treated for a median of three cycles. After central review of all radiology, partial responses were seen in just 2/17 (11%) patients. The median progression-free survival was 2 months. Recent data have examined the role of EGFR tyrosine kinase inhibitors, erlotinib and gefitinib, in the treatment of BAC. In the trials examining the treatment of NSCLC with erlotinib or gefitinib, a subset of patients had dramatic radiographic responses.74,75 A retrospective review of patients treated with gefitinib identified improved response rates for patients with adenocarcinoma with BAC features (38 vs 14% for other adenocarcinomas), nonsmokers (36 vs 8% for former or current smokers), and patients with Karnofsky performance status ≥80% (22 vs 8% for patients with ≤70%).19 On the basis of the initial clinical impression that gefitinib was particularly effective in patients with BAC, SWOG conducted a phase II trial of gefitinib in patients with metastatic, recurrent, or inoperable BAC.76 Patients who had no prior treatment had a 20% response rate, compared with 14% for previously treated patients (p = 0.6). Other subsets with favorable response rates were never smokers (25 vs 17% for former/current smokers; p = 0.6), presence of rash during treatment (23 vs 0% for patients without rash; p = 0.2) and female patients (32 vs 8% for males subjects; p = 0.06). Simultaneously, investigators at Memorial Sloan-Kettering Cancer Center commenced a multi-institutional phase II trial of erlotinib in patients with BAC. To be eligible, all patients need pathologic evidence of adenocarcinoma with some bronchioloalveolar features or pure BAC. Patients had
318
THORACIC TUMORS
measurable disease and were either chemotherapy na¨ıve or had refractory disease after one therapeutic regimen. In this trial, partial responses have been seen in 24% (19/78) of patients and median survival approximates 1 year. Of note, even in this more homogeneous pathologic group, the response rate for patients classified as never smokers (less than 100 cigarettes per lifetime) was superior to that of current smokers or former heavy smokers. A variety of treatments whose aim is to treat bronchorrhea, an often debilitating symptom of BAC, have also been reported. Treatment of the underlying malignancy is the most effective treatment of this symptom; however, symptomatic relief can be obtained with some agents. Limited reports support the efficacy of macrolide antibiotics, epidermal growth factor tyrosine kinase inhibitors, and inhaled indomethacin for the reduction of bronchorrhea.77 – 81 No prospective studies have been reported for the treatment of this troublesome symptom.
Investigational Therapies While use of epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKI)s results in dramatic but temporary efficacy in some patients, newer agents either with broader utility or which target another subset of BAC, unresponsive to EGFR-TKIs, are needed. Following the lead of the described trials, which established that BAC-specific trials are feasible, a number of treatments are currently under investigation. Building on the observation that there is a high frequency of somatic mutations in p53 in BAC, ECOG (E6597) investigated a trial of adenovirus p53 administered by bronchoalveolar lavage every two weeks.82 In the first 27 patients treated, only one partial response (PR) was noted, but 17 had stable disease (some with progression in the untreated lung). A clinical trial investigating autologous tumor vaccines derived from harvesting tumor tissue from patients is currently being conducted by SWOG (S0310). This trial seeks to enroll 117 patients with BAC or adenocarcinoma with BAC features. In this trial, surgically resected tumors or malignant pleural effusions are collected and processed to make patient-specific vaccines by infection with an adenoviral vector that induces secretion of granulocyte-macrophage colony-stimulating factor (GMCSF). Other clinical trials in early stages include a trial of bortezomib, a trial of cetuximab (through ECOG) and a trial of pemetrexed (through SWOG). However, these trials are, to a large extent, based on empiricism and not driven by an understanding of the molecular biology of this disease. Parallel efforts at understanding the biology of BAC (including mouse models of EGFR mutations and others) and discovering new drug targets in BAC are ongoing.
AUTHORS RECOMMENDATIONS In the management of patients with BAC, we begin with a complete description of the pathology for each patient through close consultation with the pathologist. At a minimum, the pathology report for all patients should include the type of NSCLC (e.g. squamous, adenocarcinoma, large cell,
etc.) and if any subtypes (i.e. BAC) are present, an estimation of the relative proportions should be clearly stated. If multiple foci of disease are present, the pathology from each site of disease should be rigorously compared. If adequate tissue is available, molecular analysis for EGFR mutations can also be helpful in the pathologic analysis. Initial treatment decisions focus on whether the patient can be rendered disease-free by way of surgery. Complete staging evaluation including CT of the chest and upper abdomen, PET scan, magnetic resonance imaging (MRI) of the brain and, when necessary, mediastinoscopy are performed to identify whether patients with single lesions have distant metastases and determine whether those patients with multifocal disease can or should have surgical treatment as their primary treatment. If surgical treatment is not feasible, discussion of systemic therapies focuses on the goal of prolonging life and controlling symptoms. Choice of therapy can be aided by understanding the patient’s smoking history and EGFR and K-RAS mutation status, when available. In patients with minimal or no smoking history or those with known EGFR mutations, we begin treatment with erlotinib alone or erlotinib with cytotoxic chemotherapy.83 Additional systemic therapy options are guided by good clinical practices in the treatment of general NSCLC patients. In patients suitable for diseasespecific clinical trials, adequate tissue should be mandated to allow detailed pathologic description, EGFR and K-RAS determination, and assessment of other molecular markers of interest.
REFERENCES 1. Liebow AA. Bronchiolo-alveolar carcinoma. Advances in Internal Medicine 1960; 10: 329 – 58. 2. Travis WD, et al. Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon, France: IARC Press, 2004. 3. Read WL, et al. The epidemiology of bronchioloalveolar carcinoma over the past two decades: analysis of the SEER database. Lung Cancer 2004; 45: 137 – 42. 4. Auerbach O, Garfinkel L. The changing pattern of lung carcinoma. Cancer 1991; 68: 1973 – 7. 5. Barsky SH, et al. Rising incidence of bronchioloalveolar lung carcinoma and its unique clinicopathologic features. Cancer 1994; 73: 1163 – 70. 6. Falk RT, et al. Epidemiology of bronchioloalveolar carcinoma. Cancer Epidemiol Biomarkers Prev 1992; 1: 339 – 44. 7. Rolen KA, et al. Bronchoalveolar Carcinoma (BAC) of the Lung is Related to Cigarette Smoking: A Case Control Study from Rhode Island. Proc Am Soc Clin Oncol 2003; 22: 674 (abstract 2711). 8. Palmarini M, Fan H. Molecular biology of jaagsiekte sheep retrovirus. Curr Top Microbiol Immunol 2003; 275: 81 – 115. 9. Palmarini M, Fan H, Sharp JM. Sheep pulmonary adenomatosis: a unique model of retrovirus-associated lung cancer. Trends Microbiol 1997; 5: 478 – 83. 10. Wootton SK, Halbert CL, Miller AD. Sheep retrovirus structural protein induces lung tumours. Nature 2005; 434: 904 – 7. 11. Noguchi M, et al. Small adenocarcinoma of the lung. Histologic characteristics and prognosis. Cancer 1995; 75: 2844 – 52. 12. Aoyagi Y, et al. Accumulation of losses of heterozygosity and multistep carcinogenesis in pulmonary adenocarcinoma. Cancer Res 2001; 61: 7950 – 4. 13. Yoshida Y, et al. Mutations of the epidermal growth factor receptor gene in atypical adenomatous hyperplasia and bronchioloalveolar carcinoma of the lung. Lung Cancer 2005; 50(1): 1 – 8. 14. Sarkaria IS, et al. SCCRO expression correlates with invasive progression in bronchioloalveolar carcinoma. Ann Thorac Surg 2004; 78: 1734 – 41.
BRONCHIOLOALVEOLAR CARCINOMA OF THE LUNG 15. Rusch VW, et al. Ras oncogene point mutation: an infrequent event in bronchioloalveolar cancer. J Thorac Cardiovasc Surg 1992; 104: 1465 – 9. 16. Marchetti A, et al. Bronchioloalveolar lung carcinomas: K-ras mutations are constant events in the mucinous subtype. J Pathol 1996; 179: 254 – 9. 17. Lynch TJ, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 2004; 350: 2129 – 39. 18. Paez JG, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004; 304: 1497 – 500. 19. Miller VA, et al. Bronchioloalveolar pathologic subtype and smoking history predict sensitivity to gefitinib in advanced non-small-cell lung cancer. J Clin Oncol 2004; 22: 1103 – 9. 20. Pao W, et al. EGF receptor gene mutations are common in lung cancers from “never smokers” and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc Natl Acad Sci U S A 2004; 101: 13306 – 11. 21. Sordella R, et al. Gefitinib-sensitizing EGFR mutations in lung cancer activate anti-apoptotic pathways. Science 2004; 305: 1163 – 7. 22. Tracy S, et al. Gefitinib induces apoptosis in the EGFRL858R nonsmall-cell lung cancer cell line H3255. Cancer Res 2004; 64: 7241 – 4. 23. Greulich H, et al. Oncogenic transformation by inhibitor-sensitive and -resistant EGFR mutants. PLoS Med 2005; 2: e313. 24. Atkins KA. The diagnosis of bronchioloalveolar carcinoma by cytologic means. Am J Clin Pathol 2004; 122: 14 – 6. 25. Ohori NP, Santa Maria EL. Cytopathologic diagnosis of bronchioloalveolar carcinoma: does it correlate with the 1999 World Health Organization definition? Am J Clin Pathol 2004; 122: 44 – 50. 26. Auger M, Katz RL, Johnston DA. Differentiating cytological features of bronchioloalveolar carcinoma from adenocarcinoma of the lung in fineneedle aspirations: a statistical analysis of 27 cases. Diagn Cytopathol 1997; 16: 253 – 7. 27. Lozowski W, Hajdu SI. Cytology and immunocytochemistry of bronchioloalveolar carcinoma. Acta Cytol 1987; 31: 717 – 25. 28. MacDonald LL, Yazdi HM. Fine-needle aspiration biopsy of bronchioloalveolar carcinoma. Cancer 2001; 93: 29 – 34. 29. Morishita Y, et al. Small-sized adenocarcinoma of the lung. Cytologic characteristics and clinical behavior. Cancer 2001; 93: 124 – 31. 30. Silverman JF, et al. Fine needle aspiration cytology of bronchioloalveolar-cell carcinoma of the lung. Acta Cytol 1985; 29: 887 – 94. 31. Tao LC, et al. Bronchiolo-alveolar carcinoma: a correlative clinical and cytologic study. Cancer 1978; 42: 2759 – 67. 32. Tao LC, et al. Cytologic diagnosis of bronchioloalveolar carcinoma by fine-needle aspiration biopsy. Cancer 1986; 57: 1565 – 70. 33. Zaman SS, et al. Distinction between bronchioloalveolar carcinoma and hyperplastic pulmonary proliferations: a cytologic and morphometric analysis. Diagn Cytopathol 1997; 16: 396 – 401. 34. Higashiyama M, et al. Prognostic value of bronchiolo-alveolar carcinoma component of small lung adenocarcinoma. Ann Thorac Surg 1999; 68: 2069 – 73. 35. Terasaki H, et al. Lung adenocarcinoma with mixed bronchioloalveolar and invasive components: clinicopathological features, subclassification by extent of invasive foci, and immunohistochemical characterization. Am J Surg Pathol 2003; 27: 937 – 51. 36. Ebright MI, et al. Clinical pattern and pathologic stage but not histologic features predict outcome for bronchioloalveolar carcinoma. Ann Thorac Surg 2002; 74: 1640 – 46; discussion 1646 – 7. 37. Carretta A, et al. Evaluation of radiological and pathological prognostic factors in surgically-treated patients with bronchoalveolar carcinoma. Eur J Cardiothorac Surg 2001; 20: 367 – 71. 38. Martini N, et al. Bronchiolar carcinoma: a review of 152 patients. Clinical Bulletin 1973; 3: 98 – 101. 39. Okubo K, et al. Bronchoalveolar carcinoma: clinical, radiologic, and pathologic factors and survival. J Thorac Cardiovasc Surg 1999; 118: 702 – 9. 40. Regnard JF, et al. Bronchioloalveolar lung carcinoma: results of surgical treatment and prognostic factors. Chest 1998; 114: 45 – 50. 41. Volpino P, et al. Comparative analysis of clinical features and prognostic factors in resected bronchioloalveolar carcinoma and adenocarcinoma of the lung. Anticancer Res 2003; 23: 4959 – 65.
319
42. Breathnach OS, et al. Bronchioloalveolar carcinoma of the lung: recurrences and survival in patients with stage I disease. J Thorac Cardiovasc Surg 2001; 121: 42 – 7. 43. Grover FL, Piantadosi S. Recurrence and survival following resection of bronchioloalveolar carcinoma of the lung – The Lung Cancer Study Group experience. Ann Surg 1989; 209: 779 – 90. 44. Liu YY, et al. Prognosis and recurrent patterns in bronchioloalveolar carcinoma. Chest 2000; 118: 940 – 7. 45. Rena O, et al. Stage I pure bronchioloalveolar carcinoma: recurrences, survival and comparison with adenocarcinoma of the lung. Eur J Cardiothorac Surg 2003; 23: 409 – 14. 46. Sakurai H, et al. Bronchioloalveolar carcinoma of the lung 3 centimeters or less in diameter: a prognostic assessment. Ann Thorac Surg 2004; 78: 1728 – 33. 47. Volpino P, et al. Bronchioloalveolar carcinoma: clinical, radiographic, and pathological findings. Surgical results. J Cardiovasc Surg (Torino) 2001; 42: 261 – 7. 48. Jang HJ, et al. Bronchioloalveolar carcinoma: focal area of groundglass attenuation at thin-section CT as an early sign. Radiology 1996; 199: 485 – 8. 49. Akira M, et al. High-resolution CT findings of diffuse bronchioloalveolar carcinoma in 38 patients. AJR Am J Roentgenol 1999; 173: 1623 – 9. 50. Kim EA, et al. Quantification of ground-glass opacity on highresolution CT of small peripheral adenocarcinoma of the lung: pathologic and prognostic implications. AJR Am J Roentgenol 2001; 177: 1417 – 22. 51. Matsuguma H, et al. Objective definition and measurement method of ground-glass opacity for planning limited resection in patients with clinical stage IA adenocarcinoma of the lung. Eur J Cardiothorac Surg 2004; 25: 1102 – 6. 52. Nakata M, et al. Objective radiologic analysis of ground-glass opacity aimed at curative limited resection for small peripheral non-small cell lung cancer. J Thorac Cardiovasc Surg 2005; 129(6): 1226 – 31. 53. Nomori H, et al. Differentiating between atypical adenomatous hyperplasia and bronchioloalveolar carcinoma using the computed tomography number histogram. Ann Thorac Surg 2003; 76: 867 – 71. 54. Hidaka N, et al. Expression of E-cadherin, alpha-catenin, beta-catenin, and gamma-catenin in bronchioloalveolar carcinoma and conventional pulmonary adenocarcinoma: an immunohistochemical study. Mod Pathol 1998; 11: 1039 – 45. 55. Kim BT, et al. Localized form of bronchioloalveolar carcinoma: FDG PET findings. AJR Am J Roentgenol 1998; 170: 935 – 9. 56. Smith GT, et al. FDG PET for evaluation of Bronchioloalveolar Cell Carcinoma (BAC) of the Lung. Clin Positron Imaging 1998; 1: 260. 57. Yap CS, et al. FDG-PET imaging in lung cancer: how sensitive is it for bronchioloalveolar carcinoma? Eur J Nucl Med Mol Imaging 2002; 29: 1166 – 73. 58. Marom EM, et al. T1 lung cancers: sensitivity of diagnosis with fluorodeoxyglucose PET. Radiology 2002; 223: 453 – 9. 59. Heyneman LE, Patz EF. PET imaging in patients with bronchioloalveolar cell carcinoma. Lung Cancer 2002; 38: 261 – 6. 60. Higashi K, et al. Comparison of fluorine-18-FDG PET and thallium-201 SPECT in evaluation of lung cancer. J Nucl Med 1998; 39: 9 – 15. 61. Gould MK, et al. Accuracy of positron emission tomography for diagnosis of pulmonary nodules and mass lesions: a meta-analysis. JAMA 2001; 285: 914 – 24. 62. Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy versus limited resection for T1 N0 non-small cell lung cancer. Lung Cancer Study Group. Ann Thorac Surg 1995; 60: 615 – 22; discussion 622 – 3. 63. Martini N, Melamed MR. Multiple primary lung cancers. J Thorac Cardiovasc Surg 1975; 70: 606 – 12. 64. Daly RC, et al. Bronchoalveolar carcinoma: factors affecting survival. Ann Thorac Surg 1991; 51: 368 – 76; discussion 376 – 7. 65. de Perrot M, et al. Role of lung transplantation in the treatment of bronchogenic carcinomas for patients with end-stage pulmonary disease. J Clin Oncol 2004; 22: 4351 – 6. 66. Zorn GL Jr, et al. Pulmonary transplantation for advanced bronchioloalveolar carcinoma. J Thorac Cardiovasc Surg 2003; 125: 45 – 8. 67. Garver RI Jr, et al. Recurrence of bronchioloalveolar carcinoma in transplanted lungs. N Engl J Med 1999; 340: 1071 – 4. 68. Feldman ER, Eagan RT, Schaid DJ. Metastatic bronchioloalveolar carcinoma and metastatic adenocarcinoma of the lung: comparison
320
69.
70.
71. 72.
73.
74.
75.
76.
THORACIC TUMORS
of clinical manifestations, chemotherapeutic responses, and prognosis. Mayo Clin Proc 1992; 67: 27 – 32. Breathnach OS, et al. Clinical features of patients with stage IIIB and IV bronchioloalveolar carcinoma of the lung. Cancer 1999; 86: 1165 – 73. Fujimoto N, et al. Clinical investigation of bronchioloalveolar carcinoma: a retrospective analysis of 53 patients in a single institution. Anticancer Res 1999; 19: 1369 – 73. Schiller JH, et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med 2002; 346: 92 – 8. West HL, et al. Advanced bronchioloalveolar carcinoma: a phase II trial of paclitaxel by 96-hour infusion (SWOG 9714): a Southwest Oncology Group study. Ann Oncol 2005; 16: 1076 – 80. Scagliotti GV, et al. A phase II study of paclitaxel in advanced bronchioloalveolar carcinoma (EORTC trial 08956). Lung Cancer 2005; 50(1): 91 – 6. Fukuoka M, et al. 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 2003; 21: 2237 – 46. Kris MG, et al. Efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: a randomized trial. JAMA 2003; 290: 2149 – 58. Gumerlock P, et al. Mutational Analysis of K-RAS and EGFR Implicates K-RAS as a Resistance Marker in the Southwest Oncology
77.
78. 79.
80.
81.
82.
83.
Group (SWOG) Trial S0126 of Bronchioloalveolar Carcinoma (BAC) Patients (pts) Treated with Gefitinib. Proc Am Soc Clin Oncol 2005; 24: 16s (abstract 7008). Milton DT, et al. Prompt control of bronchorrhea in patients with bronchioloalveolar carcinoma treated with gefitinib (Iressa). Support Care Cancer 2005; 13: 70 – 2. Suga T, et al. Bronchioloalveolar carcinoma with bronchorrhea treated with erythromycin. Eur Respir J 1994; 7: 2249 – 51. Takao M, et al. Successful treatment of persistent bronchorrhea by gefitinib in a case with recurrent bronchioloalveolar carcinoma: a case report. World J Surg Oncol 2003; 1: 8. Tamaoki J, et al. Inhaled indomethacin in bronchorrhea in bronchioloalveolar carcinoma: role of cyclooxygenase. Chest 2000; 117: 1213 – 4. Homma S, et al. Successful treatment of refractory bronchorrhea by inhaled indomethacin in two patients with bronchioloalveolar carcinoma. Chest 1999; 115: 1465 – 8. Carbone DP, et al. Adenovirus p53 administered by bronchoalveolar lavage in patients with bronchioloalveolar cell lung carcinoma (BAC). Proc Am Soc Clin Oncol 2003; 22: 620 (abstract 2492). Herbst RS, et al. TRIBUTE: a phase III trial of erlotinib hydrochloride (OSI-774) combined with carboplatin and paclitaxel chemotherapy in advanced non-small-cell lung cancer. J Clin Oncol 2005; 23: 5892 – 9.
Section 5 : Thoracic Tumors
28
Primary Adenoid Cystic Carcinoma of the Lung John G. Devlin and Corey J. Langer
BACKGROUND Cancer claims one in four newly diagnosed individuals in the United States every year.1 Despite recent advances in chemotherapy, targeted therapies, and radiation and surgical techniques, lung cancer, in particular, remains a tremendous burden, both in terms of worldwide human suffering and health-care expenditures. In the United States, cancers of the lung and bronchus rank first amongst all malignancies in total number of fatalities, with 164 470 expected in 2005, and second in number of incident cases, with 174 920 new diagnoses anticipated for 2005.1 The disease occurs more frequently in males than in females, though the incidence in the latter has risen, possibly related to changes in smoking patterns of the populations studied.2 Demographic studies indicate that patients of African descent are affected disproportionately.1 Tobacco remains the predominant risk factor, with duration of exposure more significant than average daily consumption. Other potential risk factors include exposure to secondhand smoke, radon gas, arsenic, nickel, asbestos, chromium, prior radiation to the chest, and perhaps inflammatory lung lesions with scarring, though the etiology of some cases remains elusive.2 Histologically, the majority of cases are non–small cell lung cancers (NSCLC), consisting of adenocarcinoma, squamous cell carcinoma, bronchoalveolar, and large cell variants.2 Staging is achieved by the Tumor-Node-Metastasis (TNM) classification.3 Small cell lung cancer (SCLC) makes up most of the remaining cases. This is a clinically important distinction, given the dramatic differences in prognosis, behavior, and treatment between small and non–small cell lung cancers. Rarely there have been reports of less common primary tumors of the lung and bronchus with carcinomatous, sarcomatous, neuroendocrine, or even multilineage origin. Adenoid cystic carcinoma is an uncommon epithelial tumor, first reported by Billroth4 in the mid-19th century; it typically arises in the minor salivary glands, although it has also been described, albeit rarely, in the breast, cervix, skin, prostate, upper aerodigestive tract, and lung.5 – 8
This tumor has been known in past years as basaloid squamous carcinoma, cylindroma, pseudoadenomatous basal cell carcinoma, adenomyoepithelioma, and even adenoma (among other less common terms), making interpretation of the literature confusing and difficult. Although low in incidence, recognition of this uncommon bronchial tumor has important clinical implications, given the very different natural history and treatment of this tumor compared with more common primary lung tumors.2 This chapter takes into account hundreds of reported cases of this rare tumor in roughly 60 years of world medical literature.
HISTOGENESIS AND PATHOLOGY Bronchial adenoid cystic carcinomas are typically submucosal in location, where they have been reportedly found in association with mucous glands,9 a likely site of origin, although others have postulated a dual origin of ductal and myoepithelial cells, similar to their salivary gland counterparts.10 Grossly, typical adenoid cystic carcinomas are small, infiltrative lesions with a poorly defined capsule, usually tan in color, but often pink or gray.11 Histologically, perineural invasion is a prominent feature, though angiolymphatic invasion is rare.2,10,11 Tumors are usually composed of small uniform cells with densely compact nuclei and very little basophilic cytoplasm; mitoses or necrosis are rarely seen.11 Cells may be arranged in solid sheets, in tubular clusters, or in the classic cribiform structure in a somewhat “cylindromatous” arrangement.11 Clinical tumor behavior and prognosis may correlate with the pattern of tumor clustering (see below),12,13 though this has not been consistently observed.10 Periodic-acid Schiff (PAS) positive mucin is often found intracellularly, but is even more prominent in the common extracellular lumen.10,11,14 The surrounding stroma is characteristically myxoid or extensively hyalinized, and is likely composed of excess basement membrane material.11 Light microscopic examination typically yields evidence of bronchial submucosal tumor spread.15 – 17 A predominantly
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
322
THORACIC TUMORS
cribiform pattern is noted in resected metastases,10 which resembled the primary tumor in nearly every case. Immunohistochemical studies have shown that the tubular and cribiform subtypes stain with antibodies to keratin (similar to normal tracheal duct epithelium and myoepithelial cells), Schwann (S-100) protein, human secretory component (SC), and lactoferrin (similar to normal tracheobronchial serous acinar cells), while the solid subtype typically lacks these stains.12 Thus, the solid subtype probably represents the least differentiated of the three subtypes,12 which is consistent with its more chaotic histopathologic activity; necrosis, mitoses, angiolymphatic invasion, and a higher synthetic-phase fraction are often observed.18 Nomori et al.12 found a strong correlation between the histologic tumor grade (defined as grade I being tubular or cribiform with no solid component, grade II identical but with solid component <20%, grade III identical but with solid component >20%) and the gross tumor growth pattern, with the grade I tumors more likely to show an intraluminal growth pattern, and the solid component/grade III tumors far more likely to invade the tracheobronchial infrastructure (see Table 1). In contrast to the more common NSCLCs, expression of p53, cyclooxygenase-2 (COX-2), and Her2-neu (erb2) has generally not been observed in adenoid cystic bronchial carcinoma,18 although some authors have proposed that “dedifferentiated” variants may exist that overexpress Her-2-neu.19 Expression of CD117, marking the presence and function of the proto-oncogene c-KIT, has been reported in all 13 patients evaluated by Albers et al.,9 which is consistent with that of adenoid cystic carcinomas in the upper aerodigestive tract. In the same study, CD117 expression did not correlate with Ki67 expression or with tumor grade.9 Mutations in the Wnt signaling pathway have also been described.20 Moran et al.10 reported tumor staining with actin, keratin, and vimentin, suggesting a myoepithelial component, which should differentiate adenoid cystic carcinomas from histologically similar adenocarcinoma. The poorly differentiated, solid subtype of adenoid cystic carcinoma is often confused with the more common basaloid squamous carcinoma, probably as a result of similar histopathologic and immunohistochemical features (though the latter are somewhat variable). Emanuel et al.21 reported that immunohistochemistry for p63 may help distinguish between the two. Ultrastructural studies22 – 24 have also shown subtle differences between the two tumors. Hewan-Lowe and Dardick22 found in a review of six cases (only one of which was bronchial) that adenoid cystic carcinoma is
more likely to display oligocilia, a combination of large and small compressed lumina, and cytoplasmic filaments that displace organelles, whereas the basaloid squamous carcinoma is more likely to show focal squamous differentiation, including keratin “pearl” formation. Additionally, basaloid squamous carcinomas (particularly esophageal variants) are often associated with dysplasia in overlying or adjacent mucosa,25 and direct contiguity is sometimes discovered between the tumor and the overlying squamous carcinoma in situ.26 Ultrastructural studies of exclusively bronchial adenoid cystic carcinomas are not available.
CLINICAL CHARACTERISTICS Adenoid cystic carcinoma is estimated to comprise 0.2% of all primary pulmonary tumors.27,28 In a study of 50 patients by Xu et al.,29 adenoid cystic carcinomas occurred with the same frequency in the trachea as squamous cell carcinomas (together the two most common primary tracheal tumors), and represented the most common primary bronchial tumor as well, sharing that distinction with carcinoid tumors. The natural history of adenoid cystic carcinomas is generally very different from that of tumors of different histology arising in the same anatomic location. Classically, these tumors are felt to be rather indolent, albeit progressive, with a slow growth pattern over a protracted period of time, usually over years to decades.2 Indeed, 5-year survival rates are probably an inadequate way to describe prognosis, and recurrence may ultimately be reported if patients are followed up for a long enough interval.2 Regional nodal involvement at diagnosis is uncommon; it was estimated to be 2%, though the rate of distant metastasis was 26%.30 In other series, the rates have been higher. The lungs are preferentially involved at some point in the course of this disease in about 40% of patients, though long-term survival of years or even decades despite lung metastases has been reported.2 Involvement of visceral organs or bone portends a prognosis that is far worse.2,10 Median age of onset appears to be in the fifth decade of life, with a range from 29–76 years, though most patients with this uncommon bronchial tumor have presented between ages 35 and 50 years. There may be a slight predilection for female gender,2,9 though most studies of bronchial variants report a relatively even distribution. Further demographic study has not been done, though Asians and Caucasians are the best-represented racial groups in the studies that have reported ethnicity.
Table 1 Growth pattern and prognosis of 12 patients based on histologic grade for adenoid cystic carcinoma of the tracheobronchial tree (Nomori et al.12 )
Tumor grade Growth pattern Prognosis
Grade I (tubular or cribiform with no solid component)
Grade II (tubular or cribiform with solid <20%)
Grade III (>20% solid component)
50% entirely intraluminal, 50% partially infiltrating All 4 patients alive (4 – 86 months postoperatively)
100% partially infiltrating
33% partially infiltrating, 66% extensively infiltrating 1 patient alive (49 months postoperatively); 2 patients deceased (41 – 78 months postoperatively)
4 patients alive (36 – 87 months postoperatively); 1 patient deceased (63 months postoperatively)
PRIMARY ADENOID CYSTIC CARCINOMA OF THE LUNG
Risk factors for this uncommon bronchial tumor have not been clearly identified, given the paucity of cases reported in the literature over the past 60 years. Interestingly, occasional reports14,27 have described patients with a prior history of pulmonary tuberculosis and suspicious multinodular lungs, which on final surgical pathology contained both tuberculomas and tumor. No formal causal or correlative relationship, however, has been established between these two diseases. Smoking is not felt to be a risk factor for this disease, unlike the more common bronchogenic carcinomas.2 This observation is consistent with reported data on risk factors for adenoid cystic carcinomas of other primary sites,5 – 8 including the upper aerodigestive tract. There are no data regarding environmental exposures or known genetic abnormalities that may predispose patients to these bronchial lesions. Most patients present with symptoms of bronchial irritation, including cough, hemoptysis, or wheezing, typically of subacute or even chronic duration. Kanematsu et al.31 found that 80% of his 16 patients were symptomatic. A Mayo Clinic series that reported a 50-year and 20-patient experience found an average duration of such symptoms before diagnosis to be 4.5 years, underscoring the symptomatic chronicity of this entity.15 Occasionally, a history of recurrent pulmonary infections can be elicited, particularly affecting very young children,32,33 probably from subtle progressive airway obstruction. Given the more common diseases responsible for these generic pulmonary symptoms, misdiagnosis is frequent. For example, Wright et al.34 reported the case of a 23-yearold woman found to have an adenoid cystic carcinoma occluding the left main bronchus, after presenting with subacute exertional dyspnea and chest tightness. This woman had a 10-year history of recurrent pulmonary infections, and was originally diagnosed with MacLeod’s (or Swyer-James) syndrome on the basis of an abnormally hyperlucent left lung on chest films at the age of 15. Stalpaert et al.35 described a 23-year-old woman, who suffered from recurrent bronchitis, dyspnea, and intermittent stridor for at least 3 years before diagnosis. Toole et al.36 reported the case of a 31-year-old woman with recurrent pneumonitis for 8 years, which for unclear reasons seemed to flare during three of her four pregnancies; she was later found to have adenoid cystic carcinoma in her lower trachea and right mainstem bronchus. Older patients are often erroneously treated empirically for chronic obstructive pulmonary disease or asthma, but their symptomatology persists.37 Other presentations have been described. Cases of acute pneumonitis and even life-threatening stridor have been reported as a result of frank airway obstruction, occasionally necessitating emergent intervention.38 A single report exists of a patient presenting with progressively worsening right scapular pain radiating down the right arm, ipsilateral Horner’s syndrome, and classic physical examination findings for a Pancoast tumor, who was later found to have an adenoid cystic carcinoma of the right lung apex.39 In one review of 50 patients with various tracheobronchial tumors, including 10 with adenoid cystic carcinomas of the trachea, signs and symptoms included stridor, respiratory
323
distress, and cyanosis.29 These patients were typically misdiagnosed with asthma, as often seen in other studies.29,38 In the same study, patients with primary bronchus tumors presented with atelectasis (with complete obstruction) or recurrent suppurative pulmonary infections (with incomplete obstruction).29 In another review of 14 patients with predominantly tracheal adenoid cystic carcinomas, dyspnea was the presenting symptom in 10 patients, cough occurred in 8 patients, and 7 patients were found to have stridor or wheezing. Although patients are typically symptomatic, these tumors have also been found incidentally on chest imaging in a minority of patients.40 – 42
RADIOGRAPHIC APPEARANCE AND DIAGNOSIS Diagnosis is usually made at the time of bronchoscopy; the gross appearance is often very similar to that of conventional bronchogenic carcinoma. Conlan et al.15 described the usual appearance of this tumor as a polypoid mass with a broad base and superficial necrosis, partially or even totally obstructing the airway lumen; in one patient bronchoscopic debulking was actually required to restore the airway function. Albers et al.9 have described multiple lobulated whitish-yellow masses with increased vascularity, covered by intact bronchial mucosa. In earlier-stage disease, the overlying mucosa is often uninvolved, though fine needle aspiration of suspiciously prominent submucosa has been reported to concur with definitive surgical pathologic examination in one study.43 This tumor classically arises in the proximal tracheobronchial tree, where the usual extensive submucosal spreading is often observed.16,17,44 Prommegger et al.44 reported a review of 16 patients with primary adenoid cystic carcinomas of the trachea and bronchus: 9 occurred in the tracheal bifurcation, 5 in the trachea itself, 1 in the middle-lobe bronchus, and only 1 was parenchymal. Occasional peripheral cases have been reported; however, the incidence of peripheral tumors is felt to be only 10%.27,41,45 Peripheral lesions form an intraluminal polypoid mass more readily, and perhaps at an earlier stage, though at least one case of submucosal extension to the proximal bronchus has been reported.27 Additionally, the tumor may rarely assume the form of multiple peripheral pulmonary nodules, as reported by Conlan et al.15 Radiologic chest evaluation, in the case of bronchial tumors, most commonly finds opacification of a portion of the affected lung, suggesting atelectasis as a direct result of airway obstruction.15 However, small tracheal tumors may be missed on standard chest radiography.29 CT scanning typically confirms a central mass, postobstructive pneumonitis, and may or may not show localized mediastinal invasion in advanced cases. Pleural effusion, pneumothorax, solitary lung masses, multifocal metastatic disease, and other patterns are rare. Typical positron emission tomography (PET) scan and magnetic resonance imaging (MRI) findings have not yet been described. Pulmonary function tests (PFTs) may provide evidence of an obstructive defect that does not improve with bronchodilators.
324
THORACIC TUMORS
Laboratory evaluation is not generally helpful. Interestingly, one report of a CA19-9–producing tumor exists,46 describing a trend in serum levels of CA19-9 that seemed to correlate with clinical events. Serum levels of CA125 and other tumor markers were less helpful.46,47 Nonetheless, no tumor marker is considered reliable enough for routine use in the management of this malignancy.
TREATMENT AND PROGNOSIS Surgical Resection Surgical resection is the most appropriate treatment option for tracheobronchial adenoid cystic carcinoma. The goals of surgery are complete resection of the tumor (if feasible), relief of obstruction, and restoration of ventilation.29 While occasionally curative, the main goal of surgery is palliative. Early authors advocated exploratory surgery to evaluate candidacy for resection,48 even when palliative excision was all that could be accomplished. Near-unanimous agreement exists in the literature, however, that carinal involvement with simultaneous bilateral mainstem bronchus invasion precludes an attempt at optimal curative resection.49 Given the typical submucosal location and pattern of local spread, surgical excision, as complete as possible, is mandatory in an effort to prevent recurrence.15 Most authors also recommend frozen section of margins intraoperatively to assure the highest chance of complete surgical resection.15 Unfortunately, complete resection is often not possible, due to the submucosal and central nature of this tumor, and the advanced stage at diagnosis; consequently, resectability rates were reported to be rather low in initial studies.15,16 Presentation at an earlier stage of disease has become more common, perhaps as a result of improved radiologic technology, which allows consideration of resection to occur more frequently.15 Additionally, since tobacco-associated comorbidity is often absent in these patients (unlike the usual lung cancer patient), it is possible that more patients are medically operable. Surgical approaches have been evaluated and described extensively,50 – 53 particularly in the case of tracheal tumors.50 – 52 Transverse cervical incision is the preferred approach for upper tracheal tumors, and a right posterolateral thoracotomy provides access to the lower half. Median sternotomy and a cervical incision allow for visualization and manipulation of the entire trachea. Ideally, the extent of tracheal resection is determined preoperatively, and the tracheal anastomosis, done with absorbable suture, is kept under as little tension as possible postoperatively to avoid complications.49 Given the central location of most tumors, unilateral sleeve pneumonectomy is the most common procedure reported, although small peripheral tumors have been successfully excised by lesser resections such as lobectomy.10 Tracheal and carinal resections have been reported with mixed results. A bad outcome is more likely to occur when greater than 6.6 cm of trachea is resected.29,35,49 – 52 Some authors recommend a two-stage procedure when bilateral thoracotomy is needed, to minimize peri-operative morbidity.49
Even with adequate resection, this tumor often recurs, either locally or distantly, after some time. Recurrence of primary tracheobronchial adenoid cystic carcinomas may occur even decades after the initial resection.10,54,55 In one report, 11 of 16 patients were available for follow-up assessment.44 Of the 11 patients available for follow-up, 3 had suffered a local recurrence at a median time of over 14 years after surgery (range 10–16 years), and 6 patients had experienced a distant recurrence at a median time of 8 years (range 2–16.5 years) postoperatively.44 However, recurrence is often amenable to resection, and surgery in this instance may provide patients with a long-term disease-free interval. For example, of the 10 patients with tracheobronchial adenoid cystic carcinomas followed up on average for 5.5 years (range: 1–10 years) by Xu and colleagues, 1 patient experienced a local recurrence 2 years after a lateral tracheal wall resection, which was again successfully resected.29 Long-term survival has occasionally been reported, even in cases that were initially resected suboptimally, perhaps due to the rather indolent growth rate of the tumor.12,14,29,54,55 For example, Schoenfeld et al.14 reported on a middleaged man who survived for 22 years after an incomplete resection of a T4N0M0 proximal left mainstem bronchus tumor. The patient had initially presented with a tubular variant adenoid cystic carcinoma and near-total atelectasis of the left lung. Pneumonectomy failed to remove all gross tumor; the aortic wall and bronchial margin were noted to contain residual carcinoma. Twenty-two years later, the patient was readmitted with a 2-week history of moderately severe dyspnea at rest, ultimately attributed to occlusion of the bronchus intermedius and right mainstem bronchus by recurrent tumor, which on biopsy resembled the original adenoid cystic carcinoma. Endobronchial laser and radiation treatments were employed, with some success, though a new right lower lobe metastasis appeared soon after. Interestingly, this patient eventually succumbed to unrelated ileus and pancreatitis, and not to respiratory difficulties or tumor, though no autopsy was performed.14 Though debate exists about the actual chance of cure, surgery usually affords most patients a long disease-free interval, which often ranges from years to decades. A patient reported by Xu et al.29 died 8 years after his initial carinal resection, and 3 years after he was found to have bilateral pulmonary metastases, though “unrelated causes” were reportedly the cause of his death. Prommegger et al.,44 in a study of 11 patients treated with surgical resection alone, reported survival rates of 79% at 5 years, and 57% at 10 years. Gaissert et al.56 reported a retrospective experience with 270 patients (135 of whom had tracheal adenoid cystic carcinoma) treated surgically; most also received additional radiation. A statistically significant association with longterm survival was observed for completeness of resection, negative airway margins, or adenoid cystic histology.56 Interestingly, nodal status, tumor length, and type of resection were not significantly associated with survival.56 In the rare studies that have reported survival rates, 5-year and 10-year survival rates with surgical resection appear to be in the range of 50–80%, and 30–60%, respectively.
PRIMARY ADENOID CYSTIC CARCINOMA OF THE LUNG
325
Interventional Bronchoscopy
Chemotherapy
For patients with symptomatic partial or near-total airway obstruction, who are not surgical candidates, palliative endobronchial laser resection has become a standard measure. Diaz-Jimenez et al.57 reported a symptom-free survival of 4–6 months in one patient, after laser resection, endobronchial radiotherapy, and stent placement. One patient with this rare tumor reported by Rau et al.58 achieved a temporary, but not otherwise defined, palliation of airway symptoms after neodymium:YAG laser therapy. Albers et al.9 reported on 14 patients (9 women and 5 men), 12 of whom were primarily treated with palliative laser resection and radiation therapy; the other 2 patients were treated with radiation alone, and with tracheal resection followed by radiation, respectively. Recurrence occurred in 11 of the 14, with an average time to recurrence or metastasis of 4.6 years. Half of all recurrences were localized to the trachea or bronchi; 5 patients had multiple recurrences in different sites. Metastases occurred in four patients, primarily to the lungs, and five patients died of the disease. Three of the 12 patients initially treated with laser resection later required tracheal resection with anastomosis.
There have been no trials evaluating chemotherapy exclusively in bronchial adenoid cystic carcinomas. Extrapolation from upper aerodigestive tract variants, however, may provide insight. Chemotherapy is generally reserved for locoregionally recurrent disease without further surgical or radiation options, or for frank metastatic disease. Since patients may live for years despite recurrence and/or metastases, it is difficult to define the exact time to begin such therapy, and reliance on clinical judgment is essential (i.e. patients with symptoms requiring palliation or with evidence of rapid progression may benefit the most). Though often responsive initially, these tumors tend not to be controlled by chemotherapy. Dreyfuss et al.61 reported a response rate of 40–50% and an average response duration of 3–7 months, with cyclophosphamide, doxorubicin, and cisplatin (CAP). Adding 5-fluorouracil to CAP increased the duration of response to 8 months, albeit with increased toxicity.62 A recent review by Laurie et al., however, has cited a somewhat lower response rate of 30%.63 Vinorelbine, mitoxantrone, and 5-fluorouracil each have shown roughly a 10% response rate.63 Taxanes have rarely been reported to provide benefit.64 Overall combination chemotherapy has not clearly improved survival, although responses may translate into symptom palliation in some cases.63
Radiation Therapy Radiation therapy may provide additional survival benefit. In one 10-year review of salivary gland variants,59 surgical resection followed by radiation therapy provided excellent local control, but was not found to affect survival. In a retrospective experience of 16 patients with tracheobronchial adenoid cystic carcinomas, 11 of whom underwent resection, microscopic residual disease was evident in 6 patients, who were then treated with palliative radiotherapy.31 Five additional patients were medically inoperable, had unresectable tumors, and/or had refused surgery, and were treated with radiation therapy alone. Five-year (91 vs 40%) and 10-year survivals (76 vs 0%) were superior for the patients who underwent surgery followed by radiation versus those who had radiation alone. Refaely et al.49 reported the results of 13 patients with resected adenoid cystic carcinomas of the tracheobronchial tree, 10 of whom received adjuvant radiation therapy; of the remaining 3, 1 died intraoperatively, and 2 did not receive radiation. One patient who had received radiation died of myocardial infarction 3 years postoperatively (with no sign of recurrence), and the other 9 patients were still alive without recurrence as of publication date (range 4 years to 12 years-2 months postoperative). The two patients who did not receive radiation suffered fatal “metastatic spread” at 5 and 7 years after the surgery, respectively. Sites of recurrence were not anatomically defined. Andou et al.60 reported a 5-year survival rate of 68.6% in 7 patients treated surgically; 6 had also received either adjuvant or neoadjuvant radiation. In the study by Gaissert et al.,56 most patients had received radiation (as reviewed above), and fared quite well. Thus, current opinion suggests that adjuvant radiation therapy may decrease the rates of local recurrence, and that this tumor is radiosensitive. The sparse data seem to suggest a possible survival benefit, particularly in combination with surgical resection, though conclusions are limited because of the small numbers of patients studied.
Targeted Therapy Trials of targeted therapy in adenoid cystic carcinoma have evaluated the use of imatinib mesylate (Gleevec) in salivary gland tumors;65,66 there are no exclusive trials in bronchial variants. Imatinib mesylate inhibits platelet-derived growth factor receptor beta (PDGFR-β), the Philadelphia chromosome protein bcr-abl, and the c-KIT protein (CD117), which is often overexpressed in adenoid cystic tumors.9,65,66 One study reported objective responses to imatinib mesylate in two patients with salivary gland adenoid cystic carcinomas, allowing one patient to undergo a potentially curative resection.65 However, a subsequent study of 16 patients with unresectable or metastatic adenoid cystic carcinomas of the salivary glands found no objective responses to imatinib mesylate in any of the 15 evaluable patients, leading to early termination of the study.66 Further investigation of targeted therapies will no doubt continue.
PROGNOSTIC FACTORS There have been efforts to assign values to various prognostic factors. Some authors have advocated that overall survival may be influenced by pathologic factors, such as tumor grade or level of differentiation.12 Data from upper aerodigestive tract adenoid cystic carcinoma studies13,67 – 70 had already suggested this correlation, most notably from Spiro et al.67 : although patients can survive for years to decades despite recurrence and metastases, long-term survival is rare in patients with grade III tumors. Additionally, patients with tubular tumors tended to have a better prognosis, which was also observed by Ishida et al.68 in primary adenoid cystic carcinomas of the lung.13,67 – 70 Nomori et al.12
326
THORACIC TUMORS
Table 2 Prognosis based on clinical stage at diagnosis for 16 patients with adenoid cystic carcinoma of the lung and tracheobronchial tree (Moran et al.10 )
Clinical stage (by TNM)
I (T1 – 2N0M0)
Prognosis
5 of 8 patients alive 5 – 12 years postoperatively; 3 patients deceased 3 – 9 years postoperatively One alive patient has metastasis to contralateral lung; 3 deceased patients died of other causes
Follow-up information
III (T1 – 2N2 – 3M0, II (T1 – 2N1M0 T3N1 – 3M0, or T3N0M0) or T4N1 – 3M0) N/a
N/a
Only a single patient, alive with “massive” local recurrence 2 years postoperatively Patient had regional lymph node metastasis at diagnosis
IV (any T, any N, M1)
Unknown
2 of 2 patients deceased, Of 4 patients, 1 2 – 12 months after diagnosis alive (3 lost to follow-up) 1 patient with liver metastasis died 2 months later of ACC; 1 patient with rib and lymph node metastasis died 12 months later of ACC
The single patient with follow-up is alive but experienced a stump recurrence
TNM, Tumor-Node-Metastasis; ACC, adenoid cystic carcinoma.
described 12 patients with adenoid cystic carcinomas of the tracheobronchial tree, all of whom were treated surgically, 8 of whom also received radiation. In addition to observing that grade I/tubular tumors tended to grow intraluminally, and that grade III/solid tumors tended to invade extraluminally, the authors also observed that the two patients who died of distant metastasis had grade III tumors (see Table 1). A preliminary correlation between prognosis and tumor grade was thus hypothesized. Other studies, though obviously handicapped by the small number of cases, have suggested similar associations between histology and prognosis. Interestingly, the single patient overexpressing Her-2-neu (erb2) reported in the study by Lin et al.18 was the only one to develop distant metastases, roughly 4 years after surgery. Albers et al.9 reported that perineural invasion could not be correlated with worse survival in the three patients in whom it was observed. Some authors have disputed these assumptions:9,10 most notably, Moran et al.10 assigned clinical stage at diagnosis as the most important prognostic factor in their study of 16 patients with predominantly (n = 14) tracheal adenoid cystic carcinomas (see Table 2). Two patients who presented with liver and/or bone metastasis died in 2 and 12 months, respectively. In contrast, three patients who had well-circumscribed endobronchial lesions were alive at 5, 10, and 12 years after surgery, respectively. Thus, earlier-stage disease portended a better prognosis, despite tumor grade.
SUMMARY AND RECOMMENDATIONS Primary adenoid cystic carcinoma of the tracheobronchial tree and lung is a rare tumor. Its recognition as a distinct entity from the more common lung cancers is essential given its dramatically different natural history. Its indolent growth rate explains both its insidious symptoms and its prolonged natural history; higher-grade tumors behave more aggressively, but are relatively rare. Long-term survival has been reported, most often with surgical resection, which remains the preferred treatment whenever possible. Radiotherapy provides a reduction in the rates of local recurrence, although trials have not demonstrated a clear-cut survival benefit to date. Systemic chemotherapy and targeted therapy have shown some promise, though small numbers of
patients have precluded efforts to develop definitive recommendations. Eventual recurrence and/or metastasis are the rule rather than the exception. Nonetheless, patients may live for years to decades despite these events. Future areas for research include the identification of genetic predisposition and possible environmental risk factors, and the development of optimal treatment strategies based on the stage and/or grade of the cancer at diagnosis.
REFERENCES 1. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55: 10 – 30. 2. Sessions RB, Harrison LB, Forastiere AA. Tumors of the salivary glands and paragangliomas. In DeVita VT, Hellman S, Rosenberg SA (eds) Cancer. Principles and Practice of Oncology, 6th ed. Philadelphia, Pennsylvania: Lippincott Williams & Wilkins, 2001: 886 – 900. 3. American Joint Committee on Cancer. Lung. In Greene FL, et al. (eds) AJCC Cancer Staging Handbook, 6th ed. New York: Springer-Verlag, 2002: 189 – 203. 4. Billroth T. Die cylindergeschwalst. In Untersuchungen ueber die Entwicklung der Blutgefasse. Berlin, Germany: G Reimer, 1856: 55 – 69. 5. Cavanzo FJ, Taylor HB. Adenoid cystic carcinoma of the breast: an analysis of 21 cases. Cancer 1969; 24: 740 – 6. 6. Fowler WC, et al. Adenoid cystic carcinoma of the cervix: report of 9 cases and a reappraisal. Obstet Gynecol 1978; 52: 337 – 9. 7. Olaffsson J, VanNostrand A. Adenoid cystic carcinoma of the larynx: a report of four cases and a review of the literature. Cancer 1977; 40: 1307 – 12. 8. Lawrens JB, Mazur MT. Adenoid cystic carcinoma: a comparative pathologic study of tumors in the salivary glands, breast, lung, and cervix. Hum Pathol 1982; 13: 916 – 24. 9. Albers E, et al. Tracheobronchial adenoid cystic carcinoma: a clinicopathologic study of 14 cases. Chest 2004; 125: 1160 – 5. 10. Moran CA, Suster S, Koss MN. Primary adenoid cystic carcinoma of the lung: a clinical and immunohistochemical study of 16 cases. Cancer 1994; 73: 1390 – 7. 11. Cotran RS, et al. Head and neck. In Schoen FJ (ed) Robbins Pathologic Basis of Disease, 5th ed. Philadelphia, Pennsylvania: W. B. Saunders, 1994: 752 – 753. 12. Nomori H, et al. Adenoid cystic carcinoma of the trachea and mainstem bronchus: a clinical, histopathologic, and immunohistochemical study. J Thorac Cardiovasc Surg 1988; 96: 271 – 7. 13. Szanto PA, et al. Histologic grading of adenoid cystic carcinoma of the salivary glands. Cancer 1984; 54: 1062 – 9. 14. Schoenfeld N, Rahn W, Loddenkemper R. Twenty-two year survival after incomplete resection of advanced adenoid cystic bronchogenic carcinoma. Eur Respir J 1996; 9: 1560 – 1.
PRIMARY ADENOID CYSTIC CARCINOMA OF THE LUNG 15. Conlan AA, et al. Adenoid cystic carcinoma (cylindroma) and mucoepidermoid carcinoma of the bronchus: factors affecting survival. J Thorac Cardiovasc Surg 1978; 76: 369 – 77. 16. Payne WS, et al. The surgical treatment of cylindroma (adenoid cystic carcinoma) and muco-epidermoid tumors of the bronchus. J Thorac Cardiovasc Surg 1959; 38: 709 – 26. 17. Reid JD. Adenoid cystic carcinoma (cylindroma) of the bronchial tree. Cancer 1952; 5: 685 – 94. 18. Lin CM, et al. Adenoid cystic carcinoma of the trachea and bronchus – a clinicopathologic study with DNA flow cytometric analysis and oncogene expression. Eur J Cardiothorac Surg 2002; 22: 621 – 5. 19. Nagao T, et al. Dedifferentiated adenoid cystic carcinoma: a clinicopathologic study of 6 cases. Mod Pathol 2003; 16: 1265 – 72. 20. Daa T, et al. Mutations in components of the Wnt signaling pathway in adenoid cystic carcinoma. Mod Pathol 2004; 17: 1475 – 82. 21. Emanuel P, et al. p63 Immunohistochemistry in the distinction of adenoid cystic carcinoma from basaloid squamous carcinoma. Mod Pathol 2005; 18: 645 – 50. 22. Hewan-Lowe K, Dardick I. Ultrastructural distinction of basaloid squamous carcinoma and adenoid cystic carcinoma. Ultrastruct Pathol 1995; 19: 371 – 81. 23. Wain SL, et al. Basaloid-squamous carcinoma of the tongue, hypopharynx and larynx: report of 10 cases. Hum Pathol 1986; 17: 1158 – 66. 24. Tandler B. Ultrastructure of adenoid cystic carcinoma of salivary gland origin. Lab Invest 1971; 24: 504 – 12. 25. Epstein JI, et al. Carcinoma of the esophagus with adenoid cystic differentiation. Cancer 1984; 53: 1131 – 6. 26. Azzopardi JG, Menzies T. Primary oesophageal adenocarcinoma. Br J Surg 1962; 49: 497 – 506. 27. Inoue H, et al. Peripheral pulmonary adenoid cystic carcinoma with substantial submucosal extension to the proximal bronchus. Thorax 1991; 46: 147 – 8. 28. deLima R. Bronchial adenoma: clinicopathologic study and results of treatment. Chest 1980; 77: 81 – 4. 29. Xu LT, Sun ZF, Li ZJ. Clinical and pathologic characteristics in patients with tracheobronchial tumor: report of 50 patients. Ann Thorac Surg 1987; 43: 276 – 8. 30. Goldstraw P, et al. The malignancy of bronchial adenomas. J Thorac Cardiovasc Surg 1976; 72: 309 – 14. 31. Kanematsu T, et al. Treatment outcome of resected and nonresected primary adenoid cystic carcinoma of the lung. Ann Thorac Cardiovasc Surg 2002; 8: 74 – 7. 32. Scott BF. Cylindromatous adenoma of the bronchus in a 4-year old child. Dis Chest 1963; 44: 547. 33. Ahel V, Zubovic I, Rozmanic V. Bronchial adenoid cystic carcinoma with saccular bronchiectasis as a cause of recurrent pneumonia in children. Pediatr Pulmonol 1992; 12: 260 – 2. 34. Wright CL, Gandhi M, Mitchell CA. Adenoid cystic carcinoma of the left main bronchus mimicking MacLeod’s syndrome. Thorax 1996; 51: 451 – 2. 35. Stalpaert G, Deneffe G, van Maele R. Surgical treatment of adenoid cystic carcinoma of the left main bronchus and trachea by left pneumonectomy, resection of 7.5 cm of trachea, and direct reanastomosis of right lung. Thorax 1979; 34: 554 – 6. 36. Toole AL, Stern H. Carcinoid and adenoid cystic carcinoma of the bronchus. Ann Thorac Surg 1972; 13: 63 – 81. 37. Nakayama M, et al. A case of adenoid cystic carcinoma of the left main bronchus, which was performed carinal resection and reconstruction while the aortic arch is pulled down. [Article in Japanese] Kyobu Geka 2001; 54: 31 – 5. 38. Takenaka H, et al. A case of adenoid cystic carcinoma presenting with stridor and which was treated by reversed gamma type stent placement. [Article in Japanese] Nihon Kokyuki Gakkai Zasshi 1998; 36: 106 – 10. 39. Hatton MQ, Allen MB, Cooke NJ. Pancoast syndrome: an unusual presentation of adenoid cystic carcinoma. Eur Respir J 1993; 6: 271 – 2. 40. Azukari K, et al. Adenoid cystic carcinoma arising in the intrapulmonary bronchus. Intern Med 1996; 35: 407 – 9. 41. Okura T, et al. A case of peripheral adenoid cystic carcinoma. [Article in Japanese] Nihon Kyobu Shikkan Gakkai Zasshi 1990; 28: 773 – 6.
327
42. Dohba S, et al. Adenoid cystic carcinoma of the lower lobe of right lung: report of a case. [Article in Japanese] Kyobu Geka 2003; 56: 977 – 80. 43. Qiu S, et al. Primary pulmonary adenoid cystic carcinoma: report of a case diagnosed by fine-needle aspiration cytology. Diagn Cytopathol 2004; 30: 51 – 6. 44. Prommegger R, Salzer GM. Long-term results of surgery for adenoid cystic carcinoma of the trachea and bronchi. Eur J Surg Oncol 1998; 24: 440 – 4. 45. Tolis GA, et al. Bronchial adenomas. Surg Gynecol Obstet 1972; 134: 605 – 10. 46. Tamura S, et al. A case of adenoid cystic carcinoma of the bronchus producing cancer-associated antigen, CA19-9. Intern Med 1992; 31: 363 – 7. 47. Tamura S, et al. Immunohistochemical analysis of CA19-9, SLX, and CA125 in adenoid cystic carcinoma of the trachea and bronchus. [Article in Japanese] Nihon Kyobu Shikkan Gakkai Zasshi 1992; 30: 407 – 11. 48. Thompson DT, Doyle JA, Roncoroni AJ. Carinal resection, left pneumonectomy, and right lung anastomosis for adenocystic basal cell carcinoma (cylindroma). Thorax 1969; 24: 752 – 5. 49. Refaely Y, Weissberg D. Surgical management of tracheal tumors. Ann Thorac Surg 1997; 64: 1429 – 32. 50. Grillo HC. Reconstruction of the trachea. Thorax 1973; 28: 667 – 79. 51. Grillo HC. Circumferential resection and reconstruction of the mediastinal and cervical trachea. Annals of Surgery 1965; 162: 374 – 88. 52. Grillo HC, et al. Extensive resection and reconstruction of mediastinal trachea without prosthesis or graft: an anatomical study in man. J Thorac Cardiovasc Surg 1964; 48: 741 – 9. 53. Pearson FG, Todd TRJ, Cooper JD. Experience with primary neoplasms of the trachea and carina. J Thorac Cardiovasc Surg 1984; 88: 511 – 8. 54. Houston HE, et al. Primary cancers of the trachea. Arch Surg 1969; 99: 132 – 40. 55. Wilkins EW, et al. A continuing clinical survey of adenomas of the trachea and bronchus in a general hospital. J Thorac Cardiovasc Surg 1963; 46: 279 – 91. 56. Gaissert HA, et al. Long-term survival after resection of primary adenoid cystic and squamous cell carcinoma of the trachea and carina. Ann Thorac Surg 2004; 78: 1889 – 96. 57. Diaz-Jimenez JP, Canela-Cordona M, Maestre-Alcacer J. Nd:YAG laser photoresection of low-grade malignant tumors of the tracheobronchial tree. Chest 1990; 97: 20 – 2. 58. Rau BK, Harikrishnan KM, Krishna S. Neodymium:YAG laser therapy for obstructing tracheobronchial tumors. Ann Acad Med Singapore 1994; 23: 29 – 31. 59. Sur RK, et al. Adenoid cystic carcinoma of the salivary glands: a review of 10 years. Laryngoscope 1997; 107: 1276 – 80. 60. Andou A, et al. A clinical study of resected adenoid cystic carcinoma of the tracheobronchial tree. [Article in Japanese] Kyobu Geka 1993; 46: 134 – 9. 61. Dreyfuss AI, et al. Cyclophosphamide, doxorubicin and cisplatin combination chemotherapy for advanced carcinomas of salivary gland origin. Cancer 1987; 60: 2869 – 72. 62. Dimery IW, et al. Fluorouracil, doxorubicin, cyclophosphamide and cisplatin combination chemotherapy for advanced or recurrent salivary gland carcinoma. J Clin Oncol 1990; 8: 1056 – 62. 63. Laurie SJ, Su YB, Pfister DG. Chemotherapy in the management of metastatic adenoid cystic carcinoma: a systematic review. J Clin Oncol 2005; 23(16S): 5581. 64. Perrot E, et al. Chemotherapy with paclitaxel for lung metastases of cystic adenoid carcinoma. A case report and review of the literature. Rev Pneumol Clin 2003; 59: 371 – 4. 65. Alcedo JC, et al. Imatinib mesylate as treatment for adenoid cystic carcinoma of the salivary glands: report of two successfully treated cases. Head Neck 2004; 26: 829 – 31. 66. Hotte SJ, et al. Imatinib mesylate in patients with adenoid cystic cancers of the salivary glands expressing c-kit: a Princess Margaret Hospital Phase II Consortium Study. J Clin Oncol 2005; 23: 585 – 90. 67. Spiro RH, Huvos AG, Strong EW. Adenoid cystic carcinoma: factors influencing survival. Am J Surg 1979; 138: 579 – 83.
328
THORACIC TUMORS
68. Ishida T, et al. Clinical applications of the pathologic properties of small cell carcinoma, large cell carcinoma, and adenoid cystic carcinoma of the lung. Semin Surg Oncol 1990; 6: 53 – 63. 69. Fordice J, et al. Adenoid cystic carcinoma of the head and neck: predictors of morbidity and mortality. Arch Otolaryngol Head Neck Surg 1999; 125: 149 – 52.
70. Perzin KH, Gullane P, Clairmont AC. Adenoid cystic carcinoma arising in salivary glands: a correlation of histological features and clinical outcome. Cancer 1978; 42: 265 – 82.
Section 5 : Thoracic Tumors
29
Mucoepidermoid Tumors of the Lung Tracey L. Evans and Thomas J. Lynch
INTRODUCTION The vast majority of primary lung tumors are either small cell carcinoma or non-small cell carcinoma, the latter encompassing the adeno-, squamous cell, large cell, and adenosquamous histologic subtypes.1 The remainder of primary pulmonary tumors have been called bronchial adenomas. Mucoepidermoid tumors (METs) make up only about 2.5% of all bronchial adenomas,2 and approximately 0.2% of all primary lung tumors (see Table 1).3 Adenoid cystic carcinomas, carcinoid tumors, and mixed tumors comprise the remainder of bronchial adenomas and are considered separately. While the tumor-types classified as bronchial adenomas are indeed usually indolent, they are in fact true malignancies, and therefore the term “bronchial adenoma” is a misnomer.
HISTORICAL BACKGROUND The MET histology was first observed in the major salivary glands. Stewart et al. initially reported on METs as a distinct group with a 45-patient case series published in 1945.4 It was originally believed that METs of the salivary glands occurred in both benign and malignant forms. The two forms were distinguishable based on the uniformity of the cells and the number of observed mitoses. When Foote et al. reported that three tumors classified as “benign” did indeed recur or metastasize,5 the classification schema changed. The well-differentiated tumors became “low-grade” malignancies and the less well-differentiated became “high-grade”. Tumors that histologically appeared to be low grade and yet had certain atypical features and a more aggressive clinical course were later categorized as “intermediate grade.” MET of the lung was first reported by Smetana in 1952.6 An early review from Massachusetts General Hospital of five cases of MET of the lung published in 1958 emphasized a benign clinical course for this disease.7 Later, Wilkins et al. in 1963 reported that bronchial adenomas in general were low-grade malignancies with favorable long-term survival when treated with conservative surgery resection.8 Other
reports, however, emphasized that some METs of the lung do have the potential for malignant, aggressive behavior.9 – 13 Optimal management of MET requires an understanding of their unique biology and clinical behavior. While surgery will remain the mainstay of treatment, modalities including radiation and chemotherapy may increasingly play a role in the future.
ANATOMY Tumors of mucoepidermoid histology most commonly arise from the major salivary glands, and, in fact, MET is the most common histology observed in malignant salivary gland tumors. Adenoid cystic carcinoma is the second most common malignant histology in the salivary glands. In a series of 85 patients with malignant salivary gland tumors, 47% had mucoepidermoid histology and 18% had adenoid cystic histology.14 The trachea and the large bronchi contain mixed seromucinous glands and ducts within their submucosa. These are believed to be the origin of salivary glandlike tumors of the lung that histologically appear identical to those tumors arising from the salivary gland.15 The most common salivary glandlike tumor of the lung and trachea is adenoid cystic carcinoma.14 Other salivary glandlike tumors of the lung include acinic cell carcinoma, oncocytoma, pleomorphic adenoma, and mucoepidermoid carcinoma.16 METs have also been reported in the esophagus,17 the breast,18 the skin,19 the colon,20 and the conjunctiva.21 METs of the lung occur within the proximal bronchial tree or trachea. Some would consider a proximal location, or occurrence within the trachea or cartilage-containing bronchi, to be a necessary feature for the diagnosis of high-grade METs of the lung.3,16 Case reports of METs occurring in the more distal lung do exist.22 Histologically, the highgrade variant of MET of the lung appears almost identical to that of adenosquamous carcinoma, and the latter diagnosis is generally favored for more distal tumors. Some believe that in fact high-grade METs do not exist and rather all are truly of adenosquamous histology.8,23 Others believe that while
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
330
THORACIC TUMORS
Table 1 Summary of mucoepidermoid tumors of the lung.
• Comprise less than 0.5% of primary lung tumors and can be seen in younger patients • Present as endobronchial lesions covered by intact epithelium • Most are low-grade tumors that rarely metastasize, however, high-grade tumors can spread to lymph nodes and distant organs • High-grade tumors can be confused with adenosquamous carcinoma of the lung • Low-grade tumors are treated with surgical resection alone • High-grade tumors are treated with surgery, if resection is feasible, and adjuvant therapy with radiation or chemotherapy should be considered
high-grade MET of the lung and adenosquamous carcinoma of the lung are indeed the same entity, the term “high-grade MET” applies to proximal tumors and “adenosquamous” to distal tumors.3 Still others believe that adenosquamous carcinomas of the lung and high-grade METs of the lung are, indeed, distinct with differential characteristics as will be discussed below.16
BIOLOGY AND EPIDEMIOLOGY Unlike the more common small cell and non-small cell lung tumor histologies, METs are not associated with smoking or any other known carcinogen. METs occur in a wide range of age-groups16 from 3 months to 78 years and with approximately half of the cases occurring in patients under the age of 30 years.24 In adults, MET histology accounts for a mere 0.2% of primary lung tumors. However, in children and adolescents, mucoepidermoid carcinomas account for 9% of all malignant primary lung neoplasms, and MET the third most common malignant lung primary occurring in children and adolescents, after carcinoid and bronchogenic carcinoma.25 It should be noted that all primary lung tumors in the pediatric population are exceedingly rare, and over 80% of pediatric lung tumors are actually metastatic disease from nonpulmonary primaries.25 Low-grade MET is more common than the high-grade variety in both adults and children. METs of the lung occur as well-circumscribed endobronchial, exophytic, polypoid masses covered with normal respiratory endothelium.26 They occur more commonly in the segmental bronchi than in the trachea or main bronchi,27 and can cause varying degrees of bronchial lumen occlusion. The biologic behavior correlates with the histologic grade. Lowgrade tumors frequently invade the bronchial submucosa but rarely invade the pulmonary parenchyma.26 Metastases to lymph nodes or extrathoracic sites is exceedingly rare in the case of low-grade MET. High-grade METs, on the other hand, invariably demonstrate invasion into the bronchial submucosa, and invade the pulmonary parenchyma approximately half of the time.26 Nodal and distant metastases are more common in high-grade MET than with low-grade MET.
PATHOLOGY In the largest series to date,26 58 patients with MET of the lung were reported between 1960 and 1986 from the Pulmonary and Mediastinal Pathology Registry of the Armed
Forces Institute of Pathology. Tumors were distinguished as being low or high grade, and this histology predicted the clinical outcome. METs range in size from 0.63 to 6 cm.26 They consist of a combination of cystic and solid elements. The cystic areas are made up of mucus-producing glandular elements. The lumen-forming cells of METs consist predominantly of mucus-filled goblet cells as well as nonmucus-secreting columnar cells.26 The solid areas are made up of sheets of epithelial cells that are either squamoid cells or transitional (intermediate) cells.15,26 The squamoid cells are polygonal with occasional intercellular bridging but without frank keratinization while the transitional cells are oval with round nuclei.15,26 It is this area of solid growth that differentiates METs from bronchial mucous gland adenomas, their even-more-rare benign counterpart.15 Similar to their designations in the case of salivary glands, METs of the lung are divided into low-grade and high-grade tumors, and these designations predict the tumor biology and patient prognosis. Unlike their salivary gland counterpart, there is no “intermediate grade” designation for METs of the lung.15 While it is likely that there also exists the same spectrum within the METs of the lung, given the rarity of these tumors, the convention has been to categorize these tumors as either low grade or high grade.16 The high-versus low-grade distinction (see Table 2) is based primarily on nuclear pleomorphism in addition to mitotic activity, cellular necrosis, and the predominance of transitional cells.3,15 Mitotic figures are usually <1 per 20 high-power field (hpf) in low-grade METs and necrosis is usually absent.26 In contrast, mitotic activity in highgrade METs averages 4 figures per 10 hpf and necrosis is frequent.26 The gland-forming units are but a minor component of the high-grade METs relative to the sheets of epidermoid and transitional cells, which can have the appearance of a nonkeratinizing squamous cell carcinoma.26 Abrupt transitions between the transitional/squamoid cells and the lumen-forming cells are also a hallmark of the highgrade tumors.26 Distinguishing high-grade MET of the lung from adenosquamous carcinoma with its glandular and squamous components can be particularly challenging, and, as discussed previously, there are those who do believe that these are in fact the same entity. On the basis of ultrastructural analysis of high-grade MET performed by Klacsmann,16 Yousem and Hochholzer developed the following criteria for high-grade MET, which they used in their published series:26 1. The tumor must occur within the proximal bronchial tree and have an exophytic endobronchial component. Table 2 Pathologic criteria for differentiating high- and low-grade METs.
Nuclear pleomorphism Mitoses Necrosis Gland-forming units Transitional cells hpf, high-power field.
Low grade
High grade
Minimal Rare (<1/20 hpf) Absent Predominant Minor component
Prominent Frequent (>4/10 hpf) Present Minor component Predominant
MUCOEPIDERMOID TUMORS OF THE LUNG
2. The surface epithelium must not demonstrate changes of in situ carcinoma. 3. The tumor must contain a random mixture of solid sheetlike and glandular areas with heterogeneous cellular components but without keratinization or squamous pearl formation. 4. The tumor must contain transition areas to low-grade MET. There can be cases where it is impossible to distinguish whether a tumor is an adenosquamous bronchogenic carcinoma or a high-grade mucoepidermoid tumor.16 On cytogenetics, a translocation between chromosomes 11 and 19 has been noted in four cases of MET’s, three of which were from tumors originating in the minor salivary glands, and one from an MET of the lung.28,29 This may prove to be a useful marker if it holds up in future studies.
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS Because METs of the lung occur within the central airways, they frequently present with symptoms of proximal airway irritation. Patients may develop hemoptysis, wheezing, postobstructive pneumonia, and cough. Younger patients are frequently presumed to be asthmatic. Because of the indolent nature of the low-grade form of this tumor, symptoms can persist for years or even decades prior to diagnosis. Patients may also present with symptoms of metastatic disease; one patient with a biologically aggressive but histologically lowgrade MET presented with an enlarging skin nodule that was indeed a metastasis from a lung tumor,30 and another patient presented with intracranial metastases, and the diagnosis of MET of the lung was made at autopsy.31 Another unusual presentation of MET that occurred in a 19-year-old initially presented with deep venous thrombosis and recurrent pulmonary embolism.32 On chest X ray, MET of the lung may appear as a pulmonary nodule or mass. Given the central location of these tumors, chest X rays may frequently appear as entirely normal.3 Because METs frequently cause bronchial obstruction, lobar or segmental atelectasis, consolidation, or bronchial “cut-off” sign may also be apparent.33 On computed tomography imaging, mucoepidermoid carcinomas of the tracheobronchial tree appear homogeneous with mild enhancement following administration of contrast and attenuation similar to that of chest wall muscle.27 They have an ovoid shape with the long axis parallel to the branching of the involved airway. Distal bronchial dilation with mucoid impaction, subsegmental atelectasis, postobstructive pneumonia, and air trapping may also be apparent secondary to bronchial obstruction.27 Radiologically, carcinoid tumors can be differentiated from METs by their increased enhancement secondary to their increased vascularity. Adenoid cystic carcinomas, unlike METs, frequently demonstrate extraluminal extension.27 Punctate calcification occurs in about 50% of METs.27 On bronchoscopy, METs of the lung appear as wellcircumscribed, polyploid endobronchial masses covered with intact mucosa.16 Because METs are frequently covered
331
with normal respiratory epithelium, bronchial washings and brushings have a low diagnostic yield,34 and forceps biopsy is frequently required.
IMPORTANCE OF HISTOLOGIC GRADE The Yousem and Hochholzer series is the largest to date and provides perhaps the best window on epidemiology, presentation, treatment, and importance of histologic grade as a predictor of prognosis.26 In this series, 58 patients were classified as having MET of the lung by the Armed Forces Institute of Pathology between the years 1960 and 1986. Follow-up on patient outcomes was available in 93% of the cases (54/58). From this large series, one can see that tumor grade truly matters. Forty-five patients (78% of the entire cohort) were determined to have low-grade MET; 60% of these low-grade patients were female. Ages ranged from 9 to 78 with a median age of 29 years. Seven patients were under 20 years of age. The most common presenting symptoms were cough, pneumonia, fever, and hemoptysis. Only 9 out of 22 respondents reported a history of cigarette smoking (41%). All but one patient had an abnormal chest X ray: 29 patients reported a solitary mass or nodule, 15 demonstrated focal consolidation. One remaining patient was found to have MET at bronchoscopy. Forty of forty-five low-grade cases did show evidence of invasion of the bronchial submucosa on microscopy. Twenty-three cases demonstrated a postobstructive pneumonia. Hilar lymph node metastasis was found in only one case. None of these patients had clinical evidence of extrathoracic disease. Thirty-four patients underwent lobectomy, five pneumonectomy, three bilobectomy, and three sleeve resection. No adjuvant chemotherapy or radiation was given to any patient. Follow-up information was available on 41 of the low-grade patients with median follow-up of 67 months. None of the low-grade patients were known to have developed recurrent disease. Thirty-nine patients were alive and without evidence of disease at follow-up, and two died of unrelated causes at 60 and 252 months after surgery. The one patient who had a lymph node metastasis was lost to follow-up. Thirteen patients (22%) in this series were classified as high grade. Seven of these (54%) were men, with an age range of 13 to 67 years and an average age of 44.5 years. Four patients were under 30. Ten patients were symptomatic with hemoptysis and cough being the most common complaints, and two patients had systemic complaints of fatigue and weight loss. Four of five respondents were cigarette smokers. Twelve patients had nodules on chest X ray and a pneumonic infiltrate was seen in one patient. The high-grade patients were treated with lobectomy in seven cases, bilobectomy in four cases, and sleeve resection in one. One patient with a subcutaneous chest wall metastasis underwent biopsy but with no surgical resection. Hilar nodal metastases were noted in two cases. Three patients received postoperative radiation therapy. Follow-up was available in all 13 patients. Eight were alive without evidence of disease at a median of 31 months (range 1–132 months). One patient died of unknown causes
332
THORACIC TUMORS
at 42 months. Three patients (23%) died of disease. The patient with the chest wall metastasis died 1 month following diagnosis. Another patient developed widespread metastatic disease 12 months following lobectomy, and another developed osseous and pulmonary metastases 60 months following surgery and died a year later. Another patient developed a recurrence on the same side of the original tumor and underwent lobectomy 4 years after diagnosis and remained disease-free 132 months later. No patient under the age of 30 developed recurrent disease. The two patients with lymph node metastases did indeed die of metastatic disease. Overall 4-year survival in this group was 66%.
TREATMENT All patients with METs of the lung should undergo complete surgical resection if possible. Conservative lung-sparing procedures are frequently required given the central location of these tumors. Successful outcomes have been reported even in cases of carinal involvement.35 Therefore, it is essential that appropriate patients be referred to institutions specializing in bronchoplastic techniques as well as carinal resection and reconstruction. Low-grade malignancies most likely do not require mediastinal lymph node dissection given the low risk of nodal involvement.36 Adjuvant chemotherapy and radiation following surgery is not recommended in cases of low-grade METs because of the favorable prognosis with surgery alone. Even in the setting of positive surgical margins, patients with low-grade METs of the lung can be followed up closely without thoracic radiation and be successfully managed with reoperation at the time of recurrence.3 Treatment recommendations are more difficult to make for patients with high-grade METs whose outcomes are not so favorable. Postoperative radiation is typically used following resection of salivary gland tumors in cases of positive surgical margins, perineural or vascular invasion, extracapsular spread, neck node metastasis, and recurrent tumors. In these circumstances, postoperative radiation has
been shown to improve locoregional control,37 – 39 although there are no prospective studies proving an overall or disease-free survival benefit for postoperative irradiation.14 Extrapolating the experience to bronchial METs is often done and postoperative radiation is offered to some patients. It should be remembered, in light of the fact that high-grade METs do resemble adenosquamous carcinoma histologically, biologically, and prognostically, that postoperative radiation has not been shown to improve outcomes in bronchogenic carcinoma.40 There are no randomized controlled trials studying the benefit of adjuvant chemotherapy in either METs of the lung or the salivary glands, and given the rarity of these tumors, such studies will likely never be performed. Recurrence rates are higher for patients with high-grade histology and lymph node metastases, and recurrences most often occur in distant sites. Therefore, a reasonable approach in such patients would be to consider adjuvant chemotherapy despite the absence of data. Data is accumulating for the use of platinum-based chemotherapy following surgical resection of non-small cell lung cancer.41 – 43 Again, to the extent that high-grade MET does resemble adenosquamous carcinoma, the use of adjuvant chemotherapy following resection of larger tumors or node-positive tumors may be reasonable. Patients with evidence of distant metastases cannot be treated with the intent to cure. Chemotherapy may offer palliative benefit. Which chemotherapy to use? Clinical data for any specific chemotherapy regimen utilized in METs of the lung is lacking. There is more information about chemotherapeutic regimens in the treatment of salivary gland tumors. However, most of these studies combined several histologic subtypes. A partial list of chemotherapeutic regimens utilized in salivary gland MET tumors is provided in Table 3. Partial responses have been noted in salivary gland tumors treated with single-agent cisplatin, cyclophosphamide, daunorubicin, doxorubicin, paclitaxel, and vinorelbine.48,49,56 As in many tumors, response rates appear higher with combination chemotherapy than with single agents.44,46 In one
Table 3 Chemotherapy in salivary gland tumors.
Regimen Cyclophosphamide and doxorubicin Cisplatin, bleomycin, methotrexate Cyclophosphamide, doxorubicin, cisplatin Cisplatin, doxorubicin, 5-fluorouracil Cisplatin, doxorubicin, 5-fluorouracil, cyclophosphamide Cisplatin Epirubicin and 5-fluorouracil Cisplatin Cyclophosphamide, doxorubicin, cisplatin Vinorelbine Carboplatin and paclitaxel Vinorelbine Vinorelbine and cisplatin Paclitaxel Gefitinib NR, not reported.
No. of patients
Response rate (%)
No. of mucoepidermoid
Response rate mucoepidermoid
13 3 11 17 16 25 7 9 22 20 14 20 16 50 29
38 67 46 35 50 16 0 11 27 20 14 20 44 18 0
3 3 0 4 1 NR 1 2
0 67%
0 1 0 1 12 2
25% 100% NR 0 NR
0 NR 25% 0
Reference 44 44 45 46 47 48 49 49 50 51 52 53 53 54 55
MUCOEPIDERMOID TUMORS OF THE LUNG
report of chemotherapy in 18 patients with salivary gland tumors,44 three patients with mucoepidermoid carcinoma of the salivary gland (two with intermediate-grade tumors and one with low-grade) were treated with cisplatin, bleomycin, and methotrexate (PBM). The two patients with intermediategrade tumors achieved partial responses and went on to definitive surgery. Both patients ultimately had a recurrence. Two of the mucoepidermoid patients (one of the intermediate-grade patients and the low-grade patient) were also treated with second-line cyclophosphamide and doxorubicin chemotherapy, and neither responded. In another study of chemotherapy in patients with recurrent salivary gland malignancies, three patients with mucoepidermoid carcinoma were treated with cisplatin and 5-fluorouracil and one patient achieved a partial response.57 In a study of 5-fluorouracil, doxorubicin, cyclophosphamide, and cisplatin in 17 patients, the overall response rate was 50%.47 The one patient in the study with mucoepidermoid carcinoma had a partial response that lasted 13 weeks, and the patient’s overall survival was 66 weeks. Newer chemotherapeutic regimens have also been utilized in salivary gland tumors. Single-agent paclitaxel (200 mg m−2 every 3 weeks) administered as first-line chemotherapy as part of an Eastern Cooperative Oncology Group study produced an overall response rate of 18% in 44 patients with advanced salivary gland tumors; the response rate in the 12 patients with mucoepidermoid carcinoma was 25%.54 Carboplatin and paclitaxel in a phase II study of 14 patients with recurrent salivary gland tumors (8 of whom had had prior cisplatin) produced a response rate of 14%, but only one patient had MET.52 There is some in vitro data of chemotherapeutic regimens in MET of the lung generated through the utilization of NCI-H292, a cell line derived from a cervical lymph node metastasis of a pulmonary mucoepidermoid carcinoma in a 32-year-old woman. Gemcitabine has been shown to decrease the viability, inhibit colony formation, and trigger apoptosis of NCI-H292 in vitro.58 The National Cancer Institute of Canada is currently conducting a phase II evaluation of carboplatin or cisplatin with gemcitabine in patients with locally advanced, recurrent, or metastatic salivary gland tumors. Gefitinib, an oral epidermal growth factor receptor inhibitor initially approved for use in non-small cell lung cancer, has been shown to suppress the proliferation of the NCI-H292 mucoepidermoid cell line in a dose-dependent manner and to inhibit intracellular synthesis of the major mucin MUC5AC mRNA in H292 cells.59 Epidermal growth factor regulates mucin secretion in normal goblet cells of the respiratory tract. In a phase II trial of gefitinib in 29 patients with incurable salivary gland tumors, gefitinib was associated with stable disease in 53% of the patients with adenoid cystic histology, but neither of the two enrolled patients with mucoepidermoid carcinoma achieved a response or stable disease.55 Additionally, gefitinib failed to improve survival in a placebo-controlled study in patients with non-small cell lung cancer and has therefore lost its approval for the treatment of this patient population.
333
PROGNOSIS The most important prognosticators in MET of the lung are histologic grade, ability to achieve a complete surgical resection, and lymph node metastasis.3,26 The prognosis for patients with low-grade MET of the lung is excellent, and the overwhelming majority of patients are cured with surgical resection alone.3,26,60 – 62 However, there are reported cases of histologically low-grade appearing METs of the lung behaving in an aggressive fashion, metastasizing widely, and leading to death.30,63 Pediatric patients with METs of the lung also fare quite well with no recurrences following surgical resection in 31 reported cases.25 The prognosis for patients with high-grade MET of the lung is significantly worse than for patients with low-grade MET, though reports of exactly how bad high-grade MET of the lung can be, vary. In the Turnbull series of 12 patients with high-grade MET of the lung, none of the patients survived longer than 15 months, and the average survival was 5.3 months.9 Only four of these patients were surgically resectable, and only two of these had complete resections. However, these two patients did not survive any longer than the patients with residual disease.9 In the Yousem series, the 4-year survival for patients with high-grade METs of the lung was 66%, and both patients who had positive hilar lymph nodes at the time of surgical resection died of metastatic disease. The Heitmiller series included three patients with high-grade MET of the lung, two of whom were unresectable at the time of their initial surgery. None of these patients had evidence of distant disease at the time of diagnosis, and yet all three died of disease within 16 months.3 In a series published by Yang, the 5-year survival of eight patients with high-grade MET of the lung following surgical resection was 25%.61 Because studies evaluating chemotherapy in metastatic MET are not randomized, because there are no studies of chemotherapy specifically in MET of the lung, and because even in the studies of chemotherapy in salivary gland tumors the MET histology makes up a minority of the tumor-types, it is impossible to know for sure whether chemotherapy improves overall survival for these patients. It is clear that chemotherapy cannot cure metastatic disease. However, the response rates observed for chemotherapy in MET of the lung are similar to those obtained in metastatic non-small cell lung cancer where a survival benefit for chemotherapy has been established.64 Therefore, administration of palliative chemotherapy is indeed reasonable.
RECOMMENDED TREATMENT APPROACH Our approach for all patients with METs of the lung is to achieve a complete surgical resection whenever possible. No further treatment is required for patients with lowgrade disease, and the long-term prognosis is excellent though recurrences can occur. Consideration should be given to adjuvant chemotherapy for high-grade METs that are large or that have spread to regional lymph nodes. Postoperative radiation may be appropriate for high-grade tumors with high risk for local recurrence. While metastatic
334
THORACIC TUMORS
mucoepidermoid carcinoma cannot be cured, chemotherapy can offer palliative benefit. We would recommend either platinum-based chemotherapy or enrollment in a clinical trial.
REFERENCES 1. Sekine I, et al. Rare pulmonary tumors – a review of 32 cases. Oncology 1998; 55(5): 431 – 4. 2. Payne WS, Schier J, Woolner LB. Mixed tumors of the bronchus (Salivary gland type). J Thorac Cardiovasc Surg 1965; 49: 663 – 8. 3. Heitmiller RF, et al. Mucoepidermoid lung tumors. Ann Thorac Surg 1989; 47(3): 394 – 9. 4. Stewart FW, Foote JFW, Becker WF. Mucoepidermoid tumors of the salivary gland. Ann Surg 1945; 122: 820 – 44. 5. Foote FW, Frazell EL Jr. Tumors of the major salivary glands. Cancer 1953; 6(6): 1065 – 133. 6. Smetana HF, Iverson L, Swan LL. Bronchogenic carcinoma, an analysis of 100 autopsy cases. Mil Surg 1952; 111: 335 – 51. 7. Sniffen RC, Soutter L, Robbins LL. Muco-epidermoid tumors of the bronchus arising from surface epithelium. Am J Pathol 1958; 34(4): 671 – 83. 8. Wilkins EW Jr, et al. A continuing clinical survey of adenomas of the trachea and bronchus in a General Hospital. J Thorac Cardiovasc Surg 1963; 46: 279 – 91. 9. Turnbull AD, et al. Mucoepidermoid tumors of bronchial glands. Cancer 1971; 28(3): 539 – 44. 10. Axelsson C, Burcharth F, Johansen A. Mucoepidermoid lung tumors. J Thorac Cardiovasc Surg 1973; 65(6): 902 – 8. 11. Goldstraw P, et al. The malignancy of bronchial adenoma. J Thorac Cardiovasc Surg 1976; 72(2): 309 – 14. 12. Ozlu C, Christopherson WM, Allen JD Jr. Mucoepidermoid tumors of the bronchus. J Thorac Cardiovasc Surg 1961; 42: 24 – 31. 13. Leonardi HK, et al. Tracheobronchial mucoepidermoid carcinoma. Clinicopathological features and results of treatment. J Thorac Cardiovasc Surg 1978; 76(4): 431 – 8. 14. Bell RB, et al. Management and outcome of patients with malignant salivary gland tumors. J Oral Maxillofac Surg 2005; 63(7): 917 – 28. 15. Litzky L. Epithelial and soft tissue tumors of the tracheobronchial tree. Chest Surg Clin N Am 2003; 13(1): 1 – 40. 16. Klacsmann PG, Olson JL, Eggleston JC. Mucoepidermoid carcinoma of the bronchus: an electron microscopic study of the low grade and the high grade variants. Cancer 1979; 43(5): 1720 – 33. 17. Hagiwara N, et al. Biological behavior of mucoepidermoid carcinoma of the esophagus. J Nippon Med Sch 2003; 70(5): 401 – 7. 18. Hastrup N, Sehested M. High-grade mucoepidermoid carcinoma of the breast. Histopathology 1985; 9(8): 887 – 92. 19. Gallager HS, Miller GV, Grampa G. Primary mucoepidermoid carcinoma of the skin; report of a case. Cancer 1959; 12(2): 286 – 8. 20. Sato H, et al. Mucoepidermoid carcinoma of the ascending colon: report of a case. Surg Today 2002; 32(11): 1004 – 7. 21. Hwang IP, et al. Mucoepidermoid carcinoma of the conjunctiva: a series of three cases. Ophthalmologica 2000; 107(4): 801 – 5. 22. Green LK, Gallion TL, Gyorkey F. Peripheral mucoepidermoid tumour of the lung. Thorax 1991; 46(1): 65 – 6. 23. Reichle FA, Rosemond GP. Mucoepidermoid tumors of the bronchus. J Thorac Cardiovasc Surg 1966; 51(3): 443 – 8. 24. Granata C, et al. Mucoepidermoid carcinoma of the bronchus: a case report and review of the literature. Pediatr Pulmonol 1997; 23(3): 226 – 32. 25. Welsh JH, et al. Tracheobronchial mucoepidermoid carcinoma in childhood and adolescence: case report and review of the literature. Int J Pediatr Otorhinolaryngol 1998; 45(3): 265 – 73. 26. Yousem SA, Hochholzer L. Mucoepidermoid tumors of the lung. Cancer 1987; 60(6): 1346 – 52. 27. Kim TS, et al. Mucoepidermoid carcinoma of the tracheobronchial tree: radiographic and CT findings in 12 patients. Radiol 1999; 212(3): 643 – 8. 28. Nordkvist A, et al. Recurrent rearrangements of 11q14-22 in mucoepidermoid carcinoma. Cancer Genet Cytogenet 1994; 74(2): 77 – 83.
29. Johansson M, et al. Translocation 11;19 in a mucoepidermoid tumor of the lung. Cancer Genet Cytogenet 1995; 80(1): 85 – 6. 30. Metcalf JS, Maize JC, Shaw EB. Bronchial mucoepidermoid carcinoma metastatic to skin. Report of a case and review of the literature. Cancer 1986; 58(11): 2556 – 9. 31. Wolf KM, Mehta D, Claypool WD. Mucoepidermoid carcinoma of the lung with intracranial metastases. Chest 1988; 94(2): 435 – 8. 32. Devbhandari M, et al. Unusual presentation of mucoepidermoid carcinoma with recurrent pulmonary embolism. Eur J Cardiothorac Surg 2002; 22(3): 482 – 4. 33. Fisher DA, et al. Mucoepidermoid tumor of the lung: CT appearance. Comput Med Imaging Graph 1995; 19(4): 339 – 42. 34. Pandya H, Matthews S. Case report: Mucoepidermoid carcinoma in a patient with congenital agenesis of the left upper lobe. Br J Radiol 2003; 76(905): 339 – 42. 35. Chen F, Tatsumi A, Miyamoto Y. Successful treatment of mucoepidermoid carcinoma of the carina. Ann Thorac Surg 2001; 71(1): 366 – 8. 36. Kamiyoshihara M, et al. Low-grade malignant tumors of the lung: is lymph node dissection necessary? Oncol Rep 1998; 5(4): 841 – 3. 37. Armstrong JG, et al. Malignant tumors of major salivary gland origin. A matched-pair analysis of the role of combined surgery and postoperative radiotherapy. Arch Otolaryngol Head Neck Surg 1990; 116(3): 290 – 3. 38. North CA, et al. Carcinoma of the major salivary glands treated by surgery or surgery plus postoperative radiotherapy. Int J Radiat Oncol, Biol, Phys 1990; 18(6): 1319 – 26. 39. Garden AS, et al. Postoperative radiotherapy for malignant tumors of the parotid gland. Int J Radiat Oncol, Biol, Phys 1997; 37(1): 79 – 85. 40. PORT Meta-analysis Trialists Group. Postoperative radiotherapy for non-small cell lung cancer. Cochrane Database Syst Rev 2000; (2): CD002142. 41. Winton T, et al. Vinorelbine plus cisplatin vs. observation in resected non-small-cell lung cancer. N Engl J Med 2005; 352(25): 2589 – 97. 42. Arriagada R, et al. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med 2004; 350(4): 351 – 60. 43. Pisters KM. Adjuvant chemotherapy for non-small-cell lung cancer – the smoke clears. N Engl J Med 2005; 352(25): 2640 – 2. 44. Posner MR, et al. Chemotherapy of advanced salivary gland neoplasms. Cancer 1982; 50(11): 2261 – 4. 45. Dreyfuss AI, et al. Cyclophosphamide, doxorubicin, and cisplatin combination chemotherapy for advanced carcinomas of salivary gland origin. Cancer 1987; 60(12): 2869 – 72. 46. Venook AP, et al. Cisplatin, doxorubicin, and 5-fluorouracil chemotherapy for salivary gland malignancies: a pilot study of the Northern California Oncology Group. J Clin Oncol 1987; 5(6): 951 – 5. 47. Dimery IW, et al. Fluorouracil, doxorubicin, cyclophosphamide, and cisplatin combination chemotherapy in advanced or recurrent salivary gland carcinoma. J Clin Oncol 1990; 8(6): 1056 – 62. 48. Licitra L, et al. Cisplatin in advanced salivary gland carcinoma. A phase II study of 25 patients. Cancer 1991; 68(9): 1874 – 7. 49. Jones AS, et al. A randomised phase II trial of epirubicin and 5-fluorouracil versus cisplatinum in the palliation of advanced and recurrent malignant tumour of the salivary glands. Br J Cancer 1993; 67(1): 112 – 4. 50. Licitra L, et al. Cisplatin, doxorubicin and cyclophosphamide in advanced salivary gland carcinoma. A phase II trial of 22 patients. Ann Oncol 1996; 7(6): 640 – 2. 51. Airoldi M, et al. Vinorelbine treatment of recurrent salivary gland carcinomas. Bull Cancer 1998; 85(10): 892 – 4. 52. Airoldi M, et al. Paclitaxel and carboplatin for recurrent salivary gland malignancies. Anticancer Res 2000; 20(5C): 3781 – 3. 53. Airoldi M, et al. Phase II randomized trial comparing vinorelbine versus vinorelbine plus cisplatin in patients with recurrent salivary gland malignancies. Cancer 2001; 91(3): 541 – 7. 54. Jennings T, Li Y, Pinto H. Phase II trial of paclitaxel in advanced or metastatic salivary gland malignancies: an Eastern Cooperative Oncology Group study. Proc Am Soc Clin Oncol 2001; 20: 236a.
MUCOEPIDERMOID TUMORS OF THE LUNG 55. Glisson BS, et al. Phase II trial of gefitinib in patients with incurable salivary gland cancer. Proc Am Soc Clin Oncol 2005; 23(16S): 508 (Abstract 5532). 56. Rentschler R, Burgess MA, Byers R. Chemotherapy of malignant major salivary gland neoplasms: a 25-year review of M. D. Anderson Hospital experience. Cancer 1977; 40(2): 619 – 24. 57. Airoldi M, et al. Chemotherapy for recurrent salivary gland malignancies: experience of the ENT Department of Turin University. ORL J Otorhinolaryngol Relat Spec 1994; 56(2): 105 – 11. 58. Pace E, et al. Effects of gemcitabine on cell proliferation and apoptosis in non-small-cell lung cancer (NSCLC) cell lines. Cancer Chemother Pharmacol 2000; 46(6): 467 – 76. 59. Kitazaki T, et al. Gefitinib inhibits MUC5AC synthesis in mucinsecreting non-small cell lung cancer cells. Lung Cancer 2005; 50(1): 19 – 24.
335
60. Shimizu J, et al. Clinicopathologic study of mucoepidermoid carcinoma of the lung. Int Surg 1998; 83(1): 1 – 3. 61. Yang CS, et al. Mucoepidermoid tumors of the lung: analysis of 11 cases. J Chin Med Assoc 2004; 67(11): 565 – 70. 62. El Mezni F, et al. Mucoepidermoid carcinoma of the lung. A series of 10 cases. Rev Pneumol Clin 2005; 61(2): 78 – 82. 63. Barsky SH, et al. “Low grade” mucoepidermoid carcinoma of the bronchus with “high grade” biological behavior. Cancer 1983; 51(8): 1505 – 9. 64. Non-small Cell Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trials. Br Med J 1995; 311(7010): 899 – 909.
Section 6 : Gastrointestinal Tumors
30
Uncommon Cancers of the Esophagus John G. Devlin, Robert D. Odze and Jonathan D. Cheng
INTRODUCTION In the United States, 14 520 new esophageal cancer cases and 13 570 esophageal cancer deaths were estimated for 2005.1 There is tremendous geographic variation in the incidence of esophageal cancer, with the highest mortality rates in China, Puerto Rico, and Singapore. In China, the incidence of this disease, by county, ranges from 1 per 100 000 persons to 100 per 100 000 persons, which suggests a significant environmental influence on the development of this disease.2 The two most common histologic types, squamous cell carcinoma (SCC) and adenocarcinoma, together represent 92–95% of all esophageal cancers.3 SCCs have historically represented the predominant histologic subtype, with an incidence that has remained relatively stable. However, the incidence of adenocarcinomas has increased considerably over the past three decades4 and has surpassed SCC as the most frequent type of cancer in the esophagus. Although esophageal cancer does not show an apparent gender predilection in high-incidence areas, it occurs more frequently in men than in women (3 : 1) in relatively lowincidence areas, such as the United States.2 Some studies have associated poor socioeconomic status and residence in urban areas with an increased risk of this disease.2 Esophageal cancer has a higher incidence in people of African descent compared to Caucasians (3 : 1),3 but there are significant differences in the histologic pattern observed. Demographic studies have demonstrated that the incidence of esophageal adenocarcinomas is highest in Caucasian men, whereas SCCs are most common in men of African descent.5 Traditional risk factors for the SCCs include alcohol intake and tobacco use,2 although exposure to various subtypes of human papilloma virus (HPV) has also been proposed as a contributing factor in the development of SCC.6 Established risk factors for esophageal adenocarcinoma include obesity, chronic gastroesophageal reflux disease (GERD), and Barrett’s esophagus. Other potential contributing factors include vitamin deficiencies (zinc, vitamin A, riboflavin), low intake
of fruits and vegetables, ingestion of N -nitroso compounds, and chronic thermal injury.2 Patients with SCC or adenocarcinoma of the esophagus typically present with weight loss and dysphagia, at a mean age of 55 years.2 Diagnosis is usually made by endoscopic biopsy, and the tumor-node-metastasis (TNM) system provides valuable treatment-related staging information (see Table 1).7 SCC and adenocarcinoma are normally indistinguishable by endoscopic or radiologic means. SCC arises from the squamous epithelial lining of the esophagus, and occurs most frequently in the upper-to-middle third of the esophagus.8 Adenocarcinomas develop from metaplastic columnar epithelium (Barrett’s esophagus), and are most commonly found in the distal esophagus.8 Treatment of these two most common histologic types is similar, and frequently involves a multimodality approach, in the form of chemotherapy, radiation therapy, and surgery. Uncommon tumors of the esophagus represent approximately 5–8% of all primary esophageal tumors. These include verrucous carcinomas, carcinosarcomas, basaloid carcinomas, adenoid cystic carcinomas, mixed squamous and adenocarcinomas, neuroendocrine tumors, gastrointestinal stromal tumors (GISTs), and other sarcomas, choriocarcinomas, melanomas, and metastases to the esophagus. This chapter focuses on the clinical characteristics, etiology, pathologic features, natural history, treatment, and prognosis of these uncommon esophageal tumors.
CARCINOMAS Squamous Cell Variants Verrucous Carcinoma Background Friedell and Rosenthal initially described verrucous carcinomas in 1941;9 they reported eight cases of a peculiar type of oral carcinoma with a “papillary verrucoid character”. These tumors were strongly associated with chewing tobacco, and tended to grow slowly, with metastasis occurring rather late in the course of the illness. In
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
338
GASTROINTESTINAL TUMORS
Table 1 Esophageal cancer staging.
Primary tumor (T) TX – Primary tumor cannot be assessed T0 – No evidence of primary tumor Tis – Carcinoma in situ T1 – Tumor invades lamina propria or submucosa T2 – Tumor invades muscularis propria T3 – Tumor invades adventitia T4 – Tumor invades adjacent structures Regional lymph nodes (N) NX – Nodes cannot be assessed N0 – No nodal involvement N1 – Regional nodal metastasis Distant metastasis (M) MX – Distant metastases cannot be assessed M0 – No distant metastases M1 – Distant metastases present Lower esophagus M1a – Metastases in celiac lymph nodes only M1b – Other distant metastases Middle esophagus M1a – Not applicable M1b – Nonregional lymph nodes and/or distant metastases Upper esophagus M1a – Metastases in cervical nodes M1b – Other distant metastases Stage grouping Stage 0 Stage 1 Stage 2A Stage 2B Stage 3 Stage 4 Stage 4A Stage 4B
Tis T1 T2 T3 T1 T2 T3 T4 Any T Any T Any T
N0 N0 N0 N0 N1 N1 N1 Any Any Any Any
N N N N
M0 M0 M0 M0 M0 M0 M0 M0 M1 M1a M1b
1948, Ackerman10 identified “verrucous squamous cell carcinoma” as a distinct morphologic subtype of oropharyngeal epidermoid carcinoma, and reported the pathologic findings of 31 patients with this type of tumor. In contrast to SCC, patients with verrucous carcinoma shared a good prognosis when treated with radiation and/or surgery. Verrucous carcinoma has also been reported in the nasal fossa, larynx, urinary bladder, scrotum, glans penis, cervix, endometrium, vulva, vagina, and ano-rectal region.11 – 13 Primary verrucous carcinoma of the esophagus is a rare tumor, and was initially reported by Minielly et al. in a small series of five patients.14 Since then, fewer than 25 case reports have been published in the English medical literature over the previous 30 years.14 – 17 Unfortunately, some of the previously published cases lack typical diagnostic features, and thus may not represent true verrucous cancers. Clinical Characteristics Patients with verrucous carcinoma are typically men with a mean age of 64 years (range 36–79).18 Subacute dysphagia to solid foods and weight loss are the most common presenting complaints. Unfortunately, there was often a substantial delay in the diagnosis of this tumor, ranging from 18 months to 7 years in previously reported cases.
Etiology and Pathology The etiology of verrucous carcinoma is unclear. One hypothesis suggests that chronic caustic ingestion (e.g. lye, kerosene, or acid exposure), together with retention of dietary carcinogens, may contribute to its development.15 Napalkov and Pozharisski19 induced esophageal verrucous carcinoma in rats after prolonged administration of N -methyl-N -nitrosoaniline. Over time, the esophagus of the rats revealed progressive stages of leukoplakia, epithelial hypertrophy, and hyperkeratotic papillary proliferation. The clinical association of verrucous carcinoma with lye ingestion supports the concept of chronic inflammation contributing to verrucous carcinogenesis. Associations have also been reported between verrucous carcinoma and tobacco use, acid ingestion, and other factors, such as achalasia and esophageal diverticula.17,18 Alternatively, exposure to HPV has been associated with the development of verrucous carcinoma in other body sites (vulva, vagina, and penis).20 In fact, several investigators have reported the presence of HPV in esophageal SCC, with a frequency ranging from 5.7 to 60%.21,22 The impact of viral exposure, and the role of other environmental factors, remains to be fully elucidated in the pathogenesis of verrucous carcinoma. Unlike pure SCCs of the esophagus, roughly onethird of verrucous carcinomas arise in the upper third of the esophagus. Grossly,8 verrucous carcinomas are warty, exophytic, cauliflower-like lesions that often involve the entire circumference of the esophagus. They may be ulcerated and reveal abundant keratosis. Since this tumor shares some superficial features with benign squamous papilloma, inadequate sampling may lead to diagnostic confusion.15 However, benign squamous papillomas are typically well-circumscribed nodules or polyps, without necrosis or ulceration. Histologically,8 verrucous carcinomas are well-differentiated, extremely low-grade SCCs with minimal cytologic atypia (see Figure 1). Epithelial hyperplasia, hyperkeratosis, and swollen rete ridges are prominent features. Fingerlike papillary projections may extend into the lamina propria and show a prominent lymphocytic response.
Figure 1 Verrucous carcinoma of the esophagus. This medium power photograph shows the superficial portion of a verrucous carcinoma characterized by markedly acanthotic and hyperplastic squamous epithelium with minimal cytologic atypia, hyperkeratosis, and parakeratosis.
UNCOMMON CANCERS OF THE ESOPHAGUS
Natural History and Treatment Verrucous carcinomas are typically slow-growing tumors that usually invade locally, with rare metastasis.8 The prognosis for patients with verrucous carcinoma is poor, despite the fact that it is considered to be a low-grade tumor. Local invasion, fistula formation, esophageal perforation, and pneumonia account for high morbidity and mortality rates.14,23 Complete surgical resection is potentially curative, and thus remains the primary treatment of choice. In previous studies of eight patients who underwent esophagectomy, five were without evidence of disease recurrence after 9 months to 3 years of follow-up.23,24 Most patients, however, are considered unsuitable for surgery and require a palliative approach. In verrucous carcinomas located outside the esophagus, radiation therapy is not typically recommended10,12 due to its poor response rate and possibility of anaplastic transformation. Experience with chemotherapy is limited. Bleomycin has been attempted in one case report25 with a partial response. Two patients with nonesophageal verrucous carcinoma (of the oral cavity and larynx) responded to oral capecitabine.26 However, nonesophageal verrucous carcinomas generally do not respond well to chemotherapy.10 The response to combined-modality therapy, or to the newer targeted therapies, requires further study. Carcinosarcoma Background Carcinosarcoma was first described in 1957 by Stout and Lattes27 (and simultaneously by Lane28 ) as a polypoid epithelial tumor containing a spindle cell component. Carcinosarcoma of the esophagus has also been referred to as pseudosarcomatous squamous cell carcinoma, spindle cell squamous carcinoma, polypoid carcinoma, and pseudosarcoma. The World Health Organization (WHO) classifies this tumor as “spindle cell squamous carcinoma”, underscoring the more traditional view that this tumor is an epithelial malignancy with sarcomatous transformation. Overall, carcinosarcoma represents 0.529 to 2.5%30 of all esophageal cancers. However, one study from Japan, in which 7 “polypoid carcinomas” were detected among 101 SCCs of the esophagus, suggests that the incidence may be higher in some populations.29 Clinical Characteristics There are no significant differences between carcinosarcomas and SCC of the esophagus, with regard to age of onset, gender, or malignant potential.30 – 33 Affected patients are more likely to be male (5 : 1 ratio) in the fifth or sixth decade of life (mean age 64). Dysphagia is the presenting symptom in over 95% of cases, although weight loss is also present in over 80%. Retrosternal pain and regurgitation may also occur. The duration of symptoms is typically less than 3 months but can be somewhat variable, ranging from 2 weeks to 1 year. Diagnosis by endoscopic biopsy has been reported to have an accuracy of 95.8%,34 although in rare cases, exploratory thoracotomy may be necessary for the establishment of an accurate diagnosis. Widened mediastinum may be seen on routine chest imaging, since these bulky, polypoid tumors may distend the esophagus and impinge upon surrounding soft tissues. Barium swallow may illustrate a filling defect, with the contrast
339
material forming a characteristic domelike “cupola” sign as it fills the space between the tumor and the esophageal wall. In 16–25% of cases, tumor infiltration of the esophageal wall may cause ulceration or constriction,30,33 although luminal dilation is more commonly seen. Etiology and Pathology Immunohistochemical staining has shown keratin in both the carcinomatous and sarcomatous components, suggesting an epithelial origin to both morphologic elements. These findings led Osamura et al.35 and Kuhajda et al.36 to suggest that “polypoid carcinoma of the esophagus” represents a carcinoma with sarcomatous transformation of the epithelial element. Epithelial features such as tonofilaments and desmosomes have been detected in the sarcomatous elements by ultrastructural studies.36 Additionally, electron microscopy and immunohistochemical studies have shown that these tumors often contain transitional cell forms, suggesting that specific epithelial features, such as cytokeratins and intercellular junctions, may be lost during the process of sarcomatous transformation. Roughly 60% of carcinosarcomas are located in the midesophagus, with less than 10% located more proximally. Histologically,8 these tumors are characterized by large, pleomorphic stellate cells with atypical mitotic figures intermingled with spindle-shaped cells (see Figure 2). The surrounding stroma often displays benign or malignant-appearing bone, cartilage, or muscle elements. The carcinomatous component is almost always in the form of a moderately to well-differentiated, either in situ or invasive, SCC, but a few tumors with “adenomatous differentiation” of the epithelial areas have also been reported.8 An association has been made with TP53 gene mutations in some cases.37 Natural History and Treatment Hughes and Cruickshank38 published a report in 1969 of a completely resected esophageal “pseudosarcoma” which later recurred locally. The invasive sarcomatous component was felt to be responsible for the patient’s demise. Subsequently, cases of “pseudosarcoma” with nodal39 and distant35 metastasis of the
Figure 2 Carcinosarcoma of the esophagus. This high-power photograph shows undifferentiated epithelioid and spindle-shaped cells with pleomorphic nuclei, increased mitoses, and markedly increased N/C ratio. This area represents the sarcomatous component of an overlying squamous cell carcinoma (not shown in this photograph).
340
GASTROINTESTINAL TUMORS
sarcomatous element were reported, confirming the aggressive nature of the sarcomatous component of this tumor. Standard surgical treatment involves radical resection with lymphadenectomy. Less extensive treatment potentially increase the likelihood of local recurrence at the anastomotic site.40 The superficial location, intraluminal exophytic growth pattern, and infrequent distant metastatic rate of these tumors make them amenable to potentially curative resection. In one study of 63 patients with carcinosarcoma, the depth of invasion was found to be limited to the mucosa or submucosa in 80% of cases.41 However, lymph node metastasis occurs in about 65% of patients, which is similar to that of pure SCCs.30 Thus, the majority of patients have lymph node metastases at the time of diagnosis. Carcinosarcoma may have a slightly better prognosis than SCC, although 5-year survival rates with various treatment modalities have been reported to range from 8 to 50%.29,31 – 33,36,42 When adjusted for stage, there appears to be no significant difference in the 5-year survival rate between carcinosarcomas and SCCs. Thus, treatment approaches mirror that of pure SCCs of the esophagus. Basaloid Squamous Carcinoma Background These tumors have been described in the breast, skin, palate, maxillary gums, upper respiratory tract, tongue, uterine cervix, and the floor of the mouth.43,44 Roughly 100 cases of basaloid squamous carcinomas of the esophagus have been described. The incidence of basaloid squamous carcinomas in Japan45 was found to be 0.1% in a review of nearly 12 000 esophageal neoplasms. Despite clear differences in their clinical and pathologic features, some reports have confused this tumor with adenoid cystic carcinomas, making the interpretation of the literature challenging. Clinical Characteristics A female predominance was initially reported, but later series showed a 3 : 1 male-to-female ratio.46 The clinical presentation of esophageal basaloid squamous carcinomas is similar to that of the more common SCC.44 The mean age of onset is 65 years, with a range of 36 to 83 years. Presenting symptoms are subacute and include progressive dysphagia, obstruction, and weight loss. Etiology and Pathology Current evidence suggests that these tumors are derived from the basal cells of the esophageal squamous epithelium. The middle third of the esophagus is most commonly affected (63%), whereas the lower third (30%) and upper third (7%) are less commonly involved.8 Basaloid squamous carcinomas histologically resemble other basaloid carcinomas of the upper aerodigestive tract.47 They are typically composed of uniform basaloid cells arranged in cords, trabeculae, solid nests, or cribriform glandular structures (see Figure 3).48 They characteristically show palisading of basaloid cells. Compared to pure adenoid cystic carcinomas, the tumor cells of basaloid squamous carcinomas are more pleomorphic, are larger in size, show higher nuclear-cytoplasmic ratio, and display more prominent mitotic activity. Foci of typical squamous differentiation are often present.8 Dysplasia is often found in the overlying or
Figure 3 Basaloid squamous cell carcinoma. Medium power photograph of a basaloid squamous cell carcinoma of the esophagus showing nests of tumor cells with small, hyperchromatic, nuclei. Peripheral palisading of tumor cells at the edges of the tumor nodules is a characteristic feature of this tumor.
adjacent esophageal mucosa.44 Direct continuity can often be demonstrated with the overlying squamous mucosa, which often shows carcinoma in situ.49 Similar to adenoid cystic carcinomas, basaloid squamous carcinomas also have an abundant hyaline stroma that are present both around and within the tumor cell nests and glands.8 These tumors normally show jagged, deeply invasive borders and often invade nerves. Rare cases have been reported to arise in association with Barrett’s esophagus.50 Natural History and Treatment For reasons that are poorly understood, esophageal basaloid tumors appear to behave in a more aggressive fashion than their nonesophageal counterparts.51 The prognosis of patients with basaloid squamous carcinoma is poor. Basaloid squamous carcinomas are locally aggressive tumors,52,53 with metastases being detected in the majority of patients at the time of initial diagnosis.46 Metastasis most frequently involves the regional lymph nodes,48 lung, mediastinum, liver, bones, cerebellum, and soft tissues.46 In patients without distant metastasis, surgical resection is the treatment of choice. Treatment with radiotherapy or chemotherapy has generally shown poor results in the few cases where it has been reported. However, in one study of five patients, radiation therapy provided one patient with long-term survival.54 Chemotherapy regimens with cisplatin, cytoxan, doxorubicin, and vincristine have provided only an occasional clinical response in patients with metastatic disease.46 Overall, 1-year survival rates range from 23 to 60%,46,55 with median survivals of 7–16 months.44,55
Adenocarcinoma Variants Adenoid Cystic Carcinoma Background Historically, two different types of esophageal carcinomas have been referred to as adenoid cystic carcinoma.56 One type is histologically and biologically identical to the adenoid cystic carcinomas of the salivary
UNCOMMON CANCERS OF THE ESOPHAGUS
gland, and the other is the basaloid squamous carcinoma (reviewed above), which may have an “adenoid cystic” growth pattern.57 Although the microscopic appearance of these two tumor types overlaps in some respects, they differ markedly in their clinical behavior.2 “True” adenoid cystic carcinomas, which have also been called cylindromas, are exceedingly rare in the esophagus. Clinical Characteristics Pure adenoid cystic carcinomas are more common in women than in men, and are most often found in patients between the ages of 40 and 60 years. Pure adenoid cystic carcinomas clearly not associated with alcohol or tobacco, unlike most other esophageal malignancies.2 Etiology and Pathology True adenoid cystic carcinomas are intramural tumors believed to arise from the submucosal mucous glands of the esophagus,58 – 61 although some have suggested that they may originate from tracheobronchial rests.62 Grossly,8 adenoid cystic carcinomas of any location are small, infiltrative lesions with a poorly defined capsule, typically tan in color, but often pink or gray. Histologically,8 they are composed of small, uniform “basaloid” cells arranged in clusters, as well as cribiform or acinar structures in a characteristic hyalinized or myxoid stroma. Mitoses and necrosis are rare. Natural History and Treatment Similar to their salivary gland counterpart, these tumors are slow-growing, indolent lesions. Metastases preferentially involve the lungs, which impacts mortality rates substantially.2,63 These tumors are, in general, less aggressive than esophageal basaloid squamous carcinomas, and usually follow a relatively indolent course. Surgical resection of the primary tumor remains the mainstay of treatment. Although experience in the treatment of primary esophageal adenoid cystic carcinomas is limited, treatment recommendations are extrapolated from their salivary gland counterparts. Patients with adenoid cystic carcinomas of salivary gland origin are frequently treated with radiation therapy after surgical resection of both early stage lesions,64 or of more locally advanced tumors that invade adjacent structures; extensive nodal involvement also warrants this therapy. In localized tumors, combined treatment with radiation and surgery has shown to provide a superior local control rate and survival, when compared with surgery alone.65 This combined approach represents the best opportunity for potentially curative treatment of locally advanced disease. Chemotherapy is generally doxorubicin-based, and is reserved for symptomatic or rapidly progressive metastatic disease. Chemotherapy is also reasonable for patients with locoregional recurrent disease without a surgical or radiation therapy option. The (cyclophosphamide, doxorubicin, cisplatin) CAP regimen has been reported to provide a response rate of 40–50%, and an average response duration of 3–7 months.66 The addition of 5-fluorouracil may provide a more durable response duration, although with increased toxicity.67 Cisplatinanthracycline –based combinations provide a response rate of 30% in patients with metastatic disease.68 Vinorelbine and mitoxantrone, each has single-agent activity with a response
341
rate of 10–15%.68 Although combination chemotherapy may increase the response rate compared to single-agent chemotherapy, an improved survival has not been clearly seen.68 Thus, combination chemotherapy is typically reserved for patients with symptomatic bulky disease, who may benefit from an increased response rate for palliation of symptoms. Targeted therapy has been evaluated in the form of imatinib mesylate, which inhibits c-KIT, a proto-oncogene commonly overexpressed in adenoid cystic tumors. An initial report suggested clinical activity, as two patients treated with imatinib experienced an objective response, allowing one patient to receive a potentially curative resection.69 However, a phase II study of 16 patients with unresectable or metastatic adenoid cystic tumors of the salivary glands found no objective responses to imatinib, leading to the appropriate early termination of the study.70 Further investigation of targeted therapies for adenoid cystic carcinoma is ongoing.
Carcinomas of Mixed Differentiation Mixed Squamous and Adenocarcinomas Background SCCs that show a mucin-secreting component have been referred to by a variety of terms. These include mixed squamous and glandular carcinoma, collision tumor, adenoacanthoma, adenosquamous carcinoma, mucoepidermoid carcinoma, SCC with adenocarcinomatous component, and adenocarcinoma with squamous metaplasia. In general, these are all tumors that show a combination of SCC with a mucin-producing adenocarcinoma, but depending on the particular pattern of growth are referred to by different names. Therefore, it is difficult to ascertain the true incidence of this uncommon tumor. The first case was reported in 1935 and was termed an adenoacanthoma.71 Dodge first employed the term “mucoepidermoid carcinoma”72 in 1961, and hypothesized that these tumors probably arose from esophageal mucous glands. The incidence of mixed adenosquamous carcinomas has been estimated to range from 0.05 to 2.2%.73 – 75 In one large Japanese series, the incidence of adenosquamous carcinoma was estimated to be 1.0% of all esophageal cancers.76 Clinical Characteristics The demographic characteristics of patients with these tumors are similar to that of pure SCCs. The median age of patients is 61 years (range 43–81), and males predominate (ratio ranging from 2 : 1 to 9 : 1).77,78 Dysphagia is the presenting symptom in 95% of patients, although other constitutional symptoms such as weight loss may occur. Endoscopic and barium studies have been unable to visually differentiate these tumors from typical SCCs.77,79 Of note, a significant proportion of endoscopic biopsy specimens are either nondiagnostic or initially reported as SCC, with the vast majority being definitively diagnosed only after complete surgical resection. These tumors have rarely been reported to secrete serologic tumor markers, such as α-fetoprotein.80 Etiology and Pathology Mixed carcinomas of the esophagus were initially believed75,81 – 83 to arise from the ducts of
342
GASTROINTESTINAL TUMORS
neuroendocrine tumors arise commonly in the gastrointestinal tract, but only occasionally in the esophagus. Included in this heterogeneous group of neoplasms is a spectrum of esophageal tumors, which range markedly in clinical behavior, and response to treatment, from the indolent, classic carcinoid tumor to the highly lethal small cell carcinomas (oat cell carcinomas), and the intermediate “atypical” carcinoids.88 Brenner et al.89 described the first esophageal carcinoid tumor in 1969. A recent German review90 found that primary esophageal neuroendocrine tumors composed roughly 1% of the 511 382 cases reviewed. Of the 8305 carcinoid tumors of various primary sites reviewed by Modlin et al.,91 only 3 primary esophageal cases were found; this represents 0.05% of all gastrointestinal carcinoids and only 0.02% of all esophageal cancers. Figure 4 Adenosquamous carcinoma of the esophagus. High-power view showing a combination of moderately differentiated squamous cell carcinoma (a) and well-differentiated adenocarcinoma (b).
the esophageal submucosal glands. However, most investigators now believe that they are derived from the overlying squamous epithelium. Although Japanese series suggest that up to 73% of these tumors involve the midesophagus, Western series have shown a higher percentage involving the distal esophagus.84 The WHO classification separates these tumors as two histologic types. The first type, termed adenosquamous carcinoma, shows an “intermingling” of squamous and adenocarcinomatous components (see Figure 4). The second type, termed mucoepidermoid carcinoma, displays a more “intimate” admixture of squamous and mucous-producing cells, and may also show cells of intermediate differentiation. Mixed carcinomas of the esophagus are morphologically similar to those that occur in the salivary glands. Most cases show carcinoma in situ in association with the invasive component.73,85 Nearly half of the 18 cases reviewed by Yachida et al.76 displayed carcinoma in situ associated with the invasive tumor. In the same study,76 nearly 80% of patients with adenosquamous tumors had lymphatic invasion, and roughly 40% had histologic evidence of angioinvasion. Fegelman et al.77 reported that 79% of patients had nodal metastases. Natural History and Treatment The prognosis of patients with this tumor type is poor, with a median survival of 9–15 months.77,84,86 However long-term survival has been reported after curative resection. Esophagectomy can be performed in a significant proportion77,79 of patients presenting with mixed esophageal carcinoma, but with a perioperative mortality approaching 10%. These cancers are reported to be refractory to radiation and chemotherapy,86 and trials of newer targeted therapies have not been studied.
Neuroendocrine Tumors Carcinoid Background Oberndofer87 first described low-grade “carcinoma-like” tumors in the small bowel in 1907. Primary
Clinical Characteristics Males are most commonly affected by esophageal neuroendocrine tumors, at a median age of 60 years.92 Symptomatic patients often present with subacute dysphagia, chest discomfort, and weight loss, in a manner similar to that of more-common esophageal cancers.93 In asymptomatic patients, these tumors are found incidentally, often in association with Barrett’s esophagus.94 Only once, carcinoid syndrome has95 been described in a patient with a primary esophageal neuroendocrine tumor, in contrast to carcinoid tumors of other primary sites in the gastrointestinal tract, where it occurs more frequently. Etiology and Pathology In contrast to other parts of the gastrointestinal tract, no well-defined neuroendocrine system is present in the esophagus; thus, a primary neuroendocrine esophageal tumor is a relatively rare event. However, argyrophilic (i.e. silver-stain-avid) neuroendocrine cells are known to populate esophageal mucosal glands in small numbers, and perhaps represent the origin of these tumors.96 Although formation of a primary polypoid tumor is possible, these tumors are also frequently encountered incidentally, in association with either Barrett’s esophagus or adenocarcinoma, leading Cary et al.97 to speculate that a multipotent stem cell may simultaneously give rise to both adenocarcinomatous and neuroendocrine elements. Grossly, the lower third of the esophagus is most commonly affected by neuroendocrine tumors, especially the gastroesophageal junction. Histologically, typical carcinoid tumors tend to be low-grade mucosal or submucosal neoplasms, typically with a mixed cellular pattern of nests and cords, intracellular neurosecretory granules,98 and abundant fibrous stroma. Immunohistochemical testing99 displays staining for synaptophysin, chromogranin, and neuronspecific enolase; rarely, staining for glucagon or serotonin is seen as well. Epithelial markers, such as keratins, are also typically positive. Arrigoni et al.100 (in a review of bronchial carcinoid tumors) described pathologic criteria that have proven useful in describing typical versus atypical carcinoids. Tumors possessing significant nuclear pleomorphism, increased mitoses, hypercellularity, disorganized architecture, and tumor necrosis are more consistent with atypical carcinoids, and may follow a more aggressive natural history than classic carcinoids.
UNCOMMON CANCERS OF THE ESOPHAGUS
Natural History and Treatment Low-grade classic esophageal carcinoid tumors are indolent, with patients often living for years, even in the absence of treatment. When treatment is needed for control of symptoms, palliative stenting or debulking surgery are options. Partensky et al.101 reported encouraging 5-year survival rates with transhiatal resection of locally advanced esophageal carcinoid tumor. Atypical carcinoids have a somewhat more aggressive behavior, and metastases preferentially involve the liver. Octreotide can be considered for symptom relief, particularly with symptoms suggestive of the carcinoid syndrome. Targeted therapies have not yet been evaluated in exclusively esophageal neuroendocrine tumors. For more information on gastrointestinal carcinoid tumors, see chapters 35 and 36. Poorly Differentiated Small Cell Carcinoma Background and Clinical Characteristics McKeown et al.102 reported the first esophageal small cell carcinoma in 1952. Primary esophageal small cell carcinoma currently represents approximately 1–3% of all esophageal cancers.103,104 Smoking is strongly linked to small cell carcinoma, but not to carcinoid. Ectopic gastrin and calcitonin production has been reported to occur with esophageal small cell carcinomas.105 Etiology and Pathology Poorly differentiated neuroendocrine tumors, such as small cell carcinoma, are grossly ulcerated and bulky. They vary in histologic appearance, showing either small or large “blue” cells, with evidence of neurosecretory differentiation. Cytoplasm is typically scant, and nuclei are prominent. Cell-to-cell molding, absence of nucleoli, and individual cell necrosis are the characteristic features. Natural History and Treatment Primary esophageal small cell carcinomas respond initially to chemotherapy, most commonly to a combination of a platinum agent and a topoisomerase inhibitor. These tumors are also radiosensitive, and patients can often achieve a temporary clinical remission with these modalities. Like their primary lung counterpart, subsequent proliferation of residual clones that are resistant to such therapy often leads to a rapidly progressive decline. Metastases are frequently found in up to 70% of presenting patients. Stage is the major determinant of survival; bulky mediastinal involvement or distant metastatic disease portends a worse prognosis. Median survival with modern treatment has been reported to be 11 months in extensive-stage disease, and up to 18 months in limited-stage disease by Bennouna et al.103 Given the systemic nature of small cell carcinomas of other body sites, chemotherapy should be the mainstay of treatment in these patients. Law et al.104 reported a median survival of 16.7 months for patients treated with systemic chemotherapy, while patients not receiving systemic treatment averaged 2.2 months. Tanabe et al.106 have described the possibility of a recurrence in the brain after a complete clinical remission, and thus advocate prophylactic cranial irradiation for those patients with responsive disease. This recommendation parallels the observed survival benefit from prophylactic cranial irradiation in small cell carcinomas of the lung.
343
More information on small cell carcinomas of the gastrointestinal tract may be found in chapter 38.
MESENCHYMAL TUMORS Background Mesenchymal tumors can be classified as either benign or malignant tumors. The most common benign mesenchymal tumors in the esophagus are leiomyomas.107,108 Leiomyomas often have an asymptomatic presentation. However, leiomyomas can also be of sufficient size to cause dysphagia and other obstructive symptoms. Grossly, these tumors may be large, firm, lobulated masses with focal areas of hemorrhage. Histologically, they are mildly to moderately cellular, with spindle cells composed of abundant eosinophilic cytoplasm (see Figure 5). Mitoses are rare. Immunohistochemical analyses typically demonstrate staining for smoothmuscle actin (SMA) and desmin, but not the hematopoietic stem cell marker CD34 or CD117 (c-KIT). This pattern distinguishes these tumors from GISTs. Treatment is surgical, and prognosis is excellent. In a study of 48 leiomyomas surgically treated by Miettinen et al.,107 local recurrences were rare, and were not associated with disease-specific mortality.
Gastrointestinal Stromal Tumors (GISTs) Background
GISTs represent approximately 0.5–1.0% of all malignant esophageal tumors.109 – 122 Primary esophageal GISTs are much less common (1–5% of all GISTs) than GISTs of the stomach (70%), small bowel (20%), or the colon and rectum (5–10%).108 Because some GISTs may resemble leiomyosarcomas on light microscopy, they were originally considered to be smooth muscle tumors, although a lower response rate to anthracycline-based chemotherapy was noted.123 Recent advances in targeted therapy, particularly with imatinib
Figure 5 Leiomyoma of the esophagus. In this high-power photograph, normal muscularis mucosa is present in the top portion of the photograph and leiomyoma is present in the bottom portion. The leiomyoma shows a hypocellular proliferation of well-differentiated spindle cells containing elongated nuclei and eosinophilic cytoplasm. The cells show virtually no cytologic atypia, and there are no mitoses.
344
GASTROINTESTINAL TUMORS
mesylate, have made reports of this tumor more frequent in the world literature, although these tumors are still exceedingly rare in the esophagus.107,123 – 125 Miettinen et al.107 reported the first case series of primary esophageal GISTs. Clinical Characteristics
Patients with esophageal GISTs107,111,116,120,126 have a mean age of 59–63, and males account for just over half of all reported cases. Dysphagia is the most common presenting complaint (75–85% of cases), due to GIST infiltration of the esophageal wall. Weight loss, chest pain, regurgitation, and shortness of breath (due to tracheobronchial compression) may also occur. Symptoms may be present for many years, since some GISTs may have an indolent growth rate.127 Patients with a predominantly polypoid, exophytic tumor may be asymptomatic if esophageal wall compliance is preserved. Chest radiograph110 shows a mediastinal mass in roughly 50% of patients, which may represent mediastinal invasion. Barium swallow110 evaluation may show a polypoid (60% of esophageal GISTs), lobulated, intraluminal mass with a broad base of attachment. Infiltrative lesions (40%) typically show irregular areas of luminal narrowing with ulceration, and are often indistinguishable from SCC. Computerized tomography (CT) scan110,115 frequently shows a thickened esophageal wall, a posterior mediastinal mass, necrotic soft tissue masses, or a large exophytic mass with extraluminal gas or contrast material within the tumor.110,117,121 On magnetic resonance imaging (MRI),110,128 these tumors are isointense to skeletal muscle on T1-weighted images, and hyperintense on T2weighted images. Endoscopic ultrasonography115,129 shows a well-defined, hyperechoic, homogeneous mass with scattered, strong echoes originating from the muscularis layer, which may represent calcifications. Angiography110,115,120 typically reveals a well-circumscribed, hypervascular mass with dilated vascular channels, early venous drainage, and occasional central avascularity.
surrounding tissues, nuclear pleomorphism, hypercellularity, a disorganized appearance, and mitotic count. Recently, however, tumor size and mitotic rate have evolved as two important prognostic features,131 despite earlier questions regarding the reproducibility and prognostic value of mitotic counts.118,122,132 Generally, GISTs with less than 5 mitoses per high-power field (hpf), and size less than 5 cm are considered to be of low risk, while tumors with more than 5 mitoses per hpf and size more than 5 cm are considered to be of high risk.131 Miettinen et al.,107 in a series of esophageal GISTs, found that no patients with a tumor smaller than 5 cm succumbed to their disease, whereas most patients with tumors exceeding 10 cm died of disease. In the esophagus, GISTs occur in the lower third in 45% of cases; the remainder 55% are equally distributed between the middle and upper thirds. GISTs in this location range in size from 2 to 22 cm.108,121 Grossly,107 GISTs are soft, with a fleshy, variegated cut surface. Central coagulative necrosis and residual “ghost tumor cells” are often found, particularly in larger tumors that tend to outgrow their vascular supply. Histologically,108 GISTs are often hypercellular, showing irregular whorls of pleomorphic spindle cells (see Figure 6). Mitoses are variable.108 Growth patterns have been described as solid, myxoid, perivascular, or pseudo-organoid.107 GISTs stain for CD34 by immunohistochemistry in roughly 70% of cases.108 Furthermore, GISTs uniformly stain for cytoplasmic CD117, thus differentiating them from leiomyomas and leiomyosarcomas.107 Endoscopic biopsy may be helpful if the overlying mucosa is ulcerated, but if the mucosa is intact, nonspecific findings and false-negative reports are common. Gaede et al.122 found that a histologic diagnosis was established in only 40% of biopsies. Submucosal, infiltrative GISTs are often initially misdiagnosed as leiomyomas, and intraluminal polypoid GISTs may be confused with verrucous squamous cell carcinomas. Other tumors included in the differential diagnosis include schwannomas (which, unlike GISTs, are
Etiology and Pathology
GISTs are believed to develop from the interstitial cells of Cajal (i.e. the pacemaker cells of the GI tract). Both GISTs and these suspected progenitor cells uniformly stain for CD117, which marks the presence of the proto-oncogene c-KIT.107,108,123 However, some investigators believe that primitive multipotent mesenchymal cells may be the cells of origin.123 Genetic mutations in c-KIT result in constitutive ligand-independent receptor activation, which in turn induces aberrant downstream signaling and tumorigenesis.123,130 Evidence of the essential role of c-KIT was demonstrated by Hirota et al.,130 who showed that transfection of mutant c-KIT into murine lymphoid cells leads to malignant transformation. Miettinen et al.107 reported c-KIT mutations on exon 11 in all of the 10 evaluable esophageal GISTs. Prediction of the clinical behavior of GISTs is best made on the basis of traditional pathologic features,108 such as tumor size, presence or absence of hemorrhage and necrosis, ulceration of the overlying mucosa, invasion into
Figure 6 Gastrointestinal stromal tumor (GIST). This high-power photograph shows interlacing bundles of cigar-shaped cells with elongated nuclei and spindle-shaped cytoplasm. The lesion is hypercellular and shows increased numbers of mitotic figures. This tumor was CD34 and c-kKIT positive by immunohistochemistry, which helps confirm the diagnosis.
UNCOMMON CANCERS OF THE ESOPHAGUS
schwann protein 100 (S-100) positive on immunohistochemistry), Kaposi sarcoma (which are CD117 negative), and carcinosarcoma.107 Natural History and Treatment
Primary esophageal GISTs tend to be low-grade malignant tumors with local invasion to the lungs, pleura, pericardium, diaphragm, or stomach. Metastases are relatively uncommon even with large primary tumors, and typically occur late in the course of the illness. Only 16% of GISTs have evidence of distant metastases at the time of diagnosis, which commonly involves the liver, lung, mediastinum, and pericardium.114,133 The treatment of choice is wide surgical resection, which even in the presence of distant metastases may provide palliation and long-term survival.134 Radiation therapy alone or with chemotherapy has been shown to provide symptom palliation and objective regression of tumor.112,114,135 Prolonged survival has also been achieved with radiation therapy in patients who were deemed unsuitable for surgery.113,135 Chemotherapy is ineffective, since there have not been any reports of responses to chemotherapy when used as the primary method of treatment. Imatinib mesylate, which specifically targets c-KIT, is reasonable given its utility in the treatment of nonesophageal GISTs.123 – 125,136 Esophageal GISTs have a much more optimistic prognosis compared to SCCs, with a median survival of 29 months,107 and 1-, 3-, and 5-year survival rates of 60.3, 42.8, and 29.1%, respectively.120 Polypoid tumor type, female gender, and lower esophageal location are all associated with improved survival. A more extensive discussion of GISTs may be found in Chapter 37, Gastrointestinal Stromal Tumors.
Other Sarcomas Non-GIST primary esophageal sarcomas are exceedingly rare, and their occurrence as true primary tumors in the esophagus is debatable. Most tumors contain both epithelial and stromal components, but pure sarcomas have been described very rarely. Case series of liposarcomas,137 – 144 malignant fibrous histiocytomas,145 – 148 synovial sarcomas,149 – 151 and rhabdomyosarcomas,152,153 and individual cases of fibrosarcoma, osteosarcoma, chondrosarcoma, and malignant schwannoma have been reported.154 – 157 These sarcomas typically arise in the soft tissue of the legs or retroperitoneum,142 but they have also been reported in the soft tissue of the abdominal wall, chest wall, cheek, soft palate, and the gastrointestinal tract. In addition to their infrequent occurrence in the esophagus, these tumors share other common features. Clinically, patients typically present with symptoms from infiltration or compression of adjacent structures. Patients may occasionally present with fever. Radiologic and endoscopic evaluation typically reveals a bulky mass. The etiology of sarcomas probably involves a primitive mesenchymal precursor cell. Grossly, sarcomas are often large, pedunculated masses. Histological appearances vary by tumor type, and the diagnosis is often aided by special studies. For example, in malignant fibrous histiocytoma spindle-shaped
345
cells admixed with giant cells are seen without signs of epithelial or myogenic features.148 Electron microscopy or immunohistochemical studies of α1-antichymotrypsin or α1antitrypsin may be required to adequately differentiate them from leiomyosarcomas.146 In synovial cell sarcoma, the presence of myosin or actin filaments may similarly be useful. The natural history of many of these sarcomas is relatively indolent. Metastases occur infrequently, but a high incidence of local recurrence is often seen, sometimes many years after initial resection.142 An exception is rhabdomyosarcoma, which appears to be a more aggressive tumor type than other esophageal sarcomas,152 and patients are more likely to present with metastases.153 Treatment of these rare sarcomas is primarily surgical. One notable exception is synovial cell sarcoma, in which radiation therapy may have a clinical impact.149
MISCELLANEOUS TUMOR TYPES Choriocarcinoma Background
Roughly 5% of primary germ cell tumors occur in extragonadal sites,158 such as the mediastinum and retroperitoneum.159 These tumors have also been reported to arise in the brain, jaw, lung, breast, prostate, thymus, pineal gland, nose, and bladder, as well as most parts of the hepatobiliary system and gastrointestinal tracts. The esophagus as a primary site of growth for a germ cell tumor has been reported only seven times in the world literature,158,160 – 165 and all cases were choriocarcinomas. Clinical Characteristics
Of the seven reported cases, four patients were male and three were female. Age at diagnosis ranged from 40 to 80 years old. The four cases diagnosed antemortem had elevated tumor markers with serum levels of α-human chorionic gonadotropin (α-hCG) as high as 17 000 mIU mL−1 . Three male patients were symptomatic and showed darkening of the areolae and genitalia, and gynecomastia. Etiology and Pathology
The origin of extragonadal choriocarcinomas is postulated to be from midline germ cells present during embryogenesis, which are later displaced during development.166 Choriocarcinomatous elements have also been observed in adenocarcinomas arising in Barrett’s esophagus.158 Choriocarcinomas of the esophagus have also been noted in association with adenocarcinomas,162 SCCs,160 pagetoid squamous cell carcinomas in situ,167 and small cell carcinomas.161 This suggests a second possible origin, namely, retrodifferentiation of neoplastic gastrointestinal epithelial cells within tumors during tumorigenesis. Six of the seven reported cases of esophageal choriocarcinoma were initially misdiagnosed as poorly differentiated adenocarcinoma, and one as a SCC.160 Three of the seven cases were diagnosed at autopsy. Natural History and Treatment
Patients are typically treated with combination chemotherapy. One patient158 obtained a clinical response with
346
GASTROINTESTINAL TUMORS
bleomycin, etoposide, and carboplatin. Other patients have received chemotherapies consisting of bleomycin, cisplatin/5-fluorouracil, or chlorambucil/methotrexate/actinomycin D.160 The mean survival for all patients was approximately 4 months. Recognition of this rare tumor type on initial biopsy is crucial to the timely use of appropriate, potentially effective chemotherapies that have shown efficacy against germ cell tumors at other sites.
Melanoma Background
Baur168 is credited with the first description of this esophageal tumor in 1906, describing a “melanosarcoma” associated with a tracheoesophageal fistula. Because of a varied histologic appearance, the terms melanocarcinoma, melanotic epithelioma, and melanosarcoma have all been used to describe this tumor. Primary esophageal melanoma represents approximately 0.1% of all esophageal cancers.169,170 Less than 200 cases have been described in the world literature since 1906. Clinical Characteristics 171
A review by DiCostanzo et al. in 1987 of all patients with primary esophageal melanoma at Memorial SloanKettering from 1949 to 1985 yielded only six cases. Half were men and half were women, and the mean age at presentation was 60 years. Other series suggest a 2 : 1 maleto-female ratio.172,173 The most common presenting symptom is dysphagia, which is present in 75% of cases, but weight loss, occult bleeding, and retrosternal pain may occur. Patients, typically, do not have a history of cutaneous melanoma. Radiographic tests are usually unable to differentiate this tumor from other esophageal cancers.174,175 Endoscopic biopsy is diagnostic in about half of cases, since some are misinterpreted as poorly differentiated SCC. Endoscopic ultrasonography may provide additional diagnostic benefits.176 Etiology and Pathology
The existence of primary esophageal melanoma was debated until 1963, when the term “melanosis” was introduced by de la Pava et al.177 to describe the presence of benign melanocytes adjacent to malignant melanoma. Melanocytes are thought to represent the precursor cell of esophageal melanomas, which may have migrated from the embryonic neural crest. Allen and Spitz178 defined primary esophageal melanomas as tumors that contain melanin, with melanocytes in the adjacent epithelium. Grossly, melanomas are friable, fungating and polypoid, and grow to large sizes. Given their typical submucosal location, the overlying mucosa may be intact. “Melanosis”, described177 as grey-black mucosal discoloration of mucosa surrounding the tumor, occurs in roughly 25% of cases, and sometimes involves the entire esophagus.179 Nearly 90% of esophageal melanomas involve the middle or lower esophagus,180,181 and this tumor has also been described in association with Barrett’s esophagus.182
Histologically,180,181 these tumors show nests and sheets of epithelioid or spindled, round-to-polygonal cells with abundant clear cytoplasm, oval nuclei, mitoses, and prominent eosinophilic nucleoli. Multinucleated giant cells are often seen. Benign melanocytes may be found adjacent to the tumor. Extensive submucosal and lateral lentiginous spreading is commonly seen. The deep surface of the tumor is often compressive rather than invasive, although infiltration of muscularis or periesophageal soft tissue may be seen as well. Melanin granules may be absent in tumors, but present in benign melanocytes. Interestingly, the histologic appearance of “melanosis” is that of benign-appearing, scattered, small hyperpigmented cells, with small nuclei, in stark contrast to malignant melanoma.177,179 Immunohistochemical examination180,181 typically shows strong staining of tumor cells with S-100 and anti-HMB45, and lack of cytokeratin expression. Overexpression of p53, estrogen receptors, and progesterone receptors is variable, which may have implications for hormonal therapy.183 Natural History and Treatment
The prognosis of melanomas is poor, and is similar to that of other esophageal malignancies. Average survival is 9.8–13.4 months, with 5-year survival rates of 2.69–4.2%.172,173 Interestingly, clinicopathologic criteria that predict prognosis in nonesophageal cutaneous melanomas, such as visible size, thickness, and histologic degree of lymphatic infiltration, are not as predictive in esophageal malignant melanomas.184 Metastases, either through lymphatic or hematogenous routes, are typically present at the time of diagnosis in 50% of patients. In a French review172 of 46 patients with autopsy findings available, only 20% had no evidence of metastatic disease. The most common sites of distant metastases were the liver (31%), lungs (17%), brain (13%), mediastinum, and pleura, with occasional deposits in the supraclavicular lymph nodes. Chello et al.185 reported a left atrial metastasis from an esophageal melanoma. Surgery has been attempted in just over half of the 110 patients reported in the literature.170 – 194 Although perioperative mortality rates which range from 0 to 10% are typically due to postoperative anastomotic leaks,186 surgery remains the best therapeutic option.172,173,180,181,187 Because of extensive lateral spreading in the submucosa, an extensive radical resection and en bloc lymphadenectomy is normally required, with large margins in excess of what is typically performed for SCCs. Ludwig et al.188 reported a median survival of only 9 months in patients who underwent resections with smaller margins. Mukaiya et al.189 reported successful resection of a local recurrence with prolonged survival, suggesting that repeat surgical excision in properly selected patients may be of benefit. In patients deemed ineligible for surgery, endoscopic Nd:YAG laser therapy in conjunction with radiation has been reported to offer palliation.190 Radiation has been shown to reduce rates of local recurrence, and to offer adequate short-term palliation, but does not necessarily extend survival.188 Chemotherapy and
UNCOMMON CANCERS OF THE ESOPHAGUS
347
immunotherapy have not provided benefit in the adjuvant setting, although there are isolated reports of benefit from combination chemotherapy/antiestrogen regimens.191,192 There have been reports of long-term survival with dacarbazine treatment.186
Secondary Neoplasms Background
Apart from direct esophageal invasion, primary tumors from distant sites may uncommonly reach the esophagus via mediastinal lymph nodes or hematogenous routes. Cancers of contiguous body sites such as the stomach, hypopharynx, or larynx, can directly invade the esophagus, or gain access by lymphatic spread through the subepithelial lymphatic plexus. Lung and breast carcinomas may also spread to the esophagus through subcarinal, paratracheal, and paraesophageal lymph node chains (see Figure 7). Although the esophagus may be the most common gastrointestinal site for distant metastases, hematogenous metastases to the esophagus are rare.193 The most common cancers that spread to the esophagus via hematogenous routes are breast, pharynx, stomach, and lung cancers. Arteriolar branching in the esophageal submucosa allows distant metastases to this location. The incidence of esophageal metastases has been estimated to be 3–6%.194 – 197 Clinical Characteristics
Metastases to the esophagus do not always produce symptoms, although dysphagia may be present with extensive esophageal involvement. In a review of breast cancer patients with esophageal metastases, patients developed dysphagia at a mean postmastectomy interval of 7.1 years.197 Weight loss, anorexia, aspiration, dysphonia, ulceration with bleeding, perforation, and tracheoesophageal fistula may also occur rarely. Metastases to the myenteric plexus of Auerbach may cause aperistalsis or achalasia, even without gross evidence of disease.198 The majority of esophageal metastases are found incidentally at autopsy. Esophageal symptoms do not typically herald the onset of metastatic disease, as most patients have widespread disease in other sites, which precedes the diagnosis of esophageal metastases. The lower or middle portion of the esophagus is most commonly affected. Barium swallow evaluation may show short strictures or narrowing, extrinsic indentation or compression, or mucosal changes indistinguishable from primary esophageal carcinomas. Breast metastases may show smooth concentric strictures of varying length, whereas melanomas or Kaposi sarcomas may show intraluminal polypoid lesions. Endoscopic biopsies of submucosal metastases are frequently nondiagnostic; thus thoracotomy may be needed to establish the correct diagnosis. Natural History and Treatment
The prognosis of patients with esophageal metastases is typically poor due to the advanced nature of the underlying disease. The choice of treatment is dependent on the
Figure 7 Metastatic lobular carcinoma of the breast to the esophagus. In this high-power photograph, there are malignant epithelioid cells, some in a linear-file arrangement, present beneath the squamous epithelium. The cells show high N/C ratio, pleomorphic hyperchromatic nuclei, and occasional signet ring cell forms. Mitoses are easily identified.
primary tumor, and the goals of treatment are typically palliative in nature. Long-term survival of a patient with an isolated esophageal metastatic resection has been reported in hypernephroma199 and ovarian200 carcinoma. Esophageal dilatation of malignant strictures in this setting should be performed with caution201 because of a high incidence of esophageal perforation, since weakening of the muscular layers of the esophageal wall and disruption of esophageal circulation by tumor cells may lead to esophageal fragility. Most patients receive tumor-specific, nonsurgical palliative therapy. Endoscopic procedures, such as esophageal stents, may also provide short-term palliation for patients with significant symptoms.
REFERENCES 1. Parker SL, et al. Cancer statistics, 1997. CA Cancer J Clin 1997; 47: 5 – 27. 2. Schrump DS, et al. Cancer of the esophagus. In DeVita VT, Hellman S, Rosenberg SA (eds) Cancer. Principles and Practice of Oncology, 5th ed. Philadelphia, Pennsylvania: Lippincott, 2001: 1051 – 1091. 3. Yang PC, Davis S. Incidence of cancer of the esophagus in the US by histologic type. Cancer 1988; 61: 612 – 7. 4. Blot WJ, et al. Rising incidence of adenocarcinoma of the esophagus and gastric cardia. JAMA 1991; 265: 1287 – 9. 5. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55: 10 – 30. 6. Sur M, Cooper K. The role of the human papilloma virus in esophageal cancer. Pathology 1998; 30: 348. 7. American Joint Committee on Cancer. Esophagus. In Greene FL, et al. (eds) AJCC Cancer Staging Handbook, 6th ed. New York: SpringerVerlag, 2002: 101. 8. Glickman JN, Odze RD. Epithelial neoplasms of the esophagus. In Odze R, Goldblum J, Crawford J (eds) Surgical Pathology of the GI Tract, Liver, Biliary Tract and Pancreas. Philadelphia, Pennsylvania: Saunders, 2003: 381 – 408. 9. Friedell HL, Rosenthal LM. The etiologic role of chewing tobacco in cancer of the mouth. JAMA 1941; 116: 2130 – 5. 10. Ackerman LV. Verrucous carcinoma of the oral cavity. Surgery 1948; 23: 670 – 8.
348
GASTROINTESTINAL TUMORS
11. Kraus FT, Perezmesa C. Verrucous carcinoma. Clinical and pathologic study of 105 cases involving oral cavity, larynx and genitalia. Cancer 1966; 19: 26 – 38. 12. el-Sebai I, et al. Verrucose squamous carcinoma of bladder. Urology 1974; 4: 407 – 10. 13. Gingrass PJ, et al. Anorectal verrucose squamous carcinoma: report of two cases. Dis Colon Rectum 1978; 21: 120 – 2. 14. Minielly JA, et al. Verrucous squamous cell carcinoma of the esophagus. Cancer 1967; 20: 2078 – 87. 15. Kavin H, et al. Chronic esophagitis evolving to verrucous squamous cell carcinoma: possible role of exogenous chemical carcinogens. Gastroenterology 1996; 110: 904 – 14. 16. Sridhar C, Zeskind HJ, Rising JA. Verrucous squamous-cell carcinoma: an unusual tumor of the esophagus. Radiology 1980; 136: 614. 17. Jasim KA, Bateson MC. Verrucous carcinoma of the oesophagus – a diagnostic problem. Histopathology 1990; 17: 473 – 5. 18. Agha FP, Weatherbee L, Sams JS. Verrucous carcinoma of the esophagus. Am J Gastroenterol 1984; 79: 844 – 9. 19. Napalkov NP, Pozharisski KM. Morphogenesis of experimental tumors of the esophagus. J Natl Cancer Inst 1969; 42: 927 – 40. 20. Ferenczy A. Pathology of malignant tumors of vulva and vagina. In Gynecologic Oncology, 2nd ed, Churchill Livingstone, 1992: Vol. 1: 419 – 442. 21. Poljak M, Cerar A. Human papillomavirus type 16 DNA in oesophageal squamous cell carcinoma. Anticancer Res 1993; 13: 2113 – 6. 22. Chen B, Yin H, Dhurandhar N. Detection of human papillomavirus DNA in esophageal squamous cell carcinomas by the polymerase chain reaction using general consensus primers. Hum Pathol 1994; 5: 920 – 3. 23. Garrard CL, et al. Verrucous carcinoma of the esophagus: surgical treatment for an often fatal disease. Am Surg 1994; 60: 613 – 6. 24. Malik AB, et al. Long-term survival in a patient with verrucous carcinoma of the esophagus. Am J Gastroenterol 1996; 91: 1031 – 3. 25. Sakurai T, et al. Bleomycin in verrucous squamous cell carcinoma of the oesophagus. Postgrad Med J 1983; 59: 578 – 80. 26. Salesiotis A, et al. Capecitabine induces rapid, sustained response in two patients with extensive oral verrucous carcinoma. Clin Cancer Res 2003; 9: 580 – 5. 27. Stout AP, Lattes R. Tumors of the esophagus. Atlas of the Tumor Pathology. First Series, Fasc. 20, Washington, District of Columbia, Armed Forces Institute of Pathology, 1957. 28. Lane N. Pseudosarcoma (polypoid sarcoma-like masses) associated with squamous-cell carcinoma of the mouth, faces, and larynx. Report of ten cases. Cancer 1957; 10: 19 – 41. 29. Xu LT, et al. Clinical and pathological characteristics of carcinosarcoma of the esophagus: report of four cases. Ann Thorac Surg 1984; 37: 197 – 203. 30. Iyomasa S, et al. Carcinosarcoma of the esophagus: a twenty-case study. Jpn J Clin Oncol 1990; 20: 99 – 106. 31. Talbert JL, Cantrell JR. Clinical and pathologic characteristics of carcinosarcoma of the esophagus. J Thorac Cardiovasc Surg 1963; 45: 1 – 10. 32. McCort JJ. Esophageal carcinosarcoma and pseudosarcoma. Radiology 1972; 102: 519 – 24. 33. Hinderleider CD, Aguam AS, Wilder JR. Carcinosarcoma of the esophagus: a case report and review of the literature. Int Surg 1979; 64: 13 – 9. 34. Bruni HC, Nelson RS. Carcinoma of the esophagus and cardia: diagnostic evaluation in 113 cases. J Thorac Cardiovasc Surg 1975; 70: 367 – 70. 35. Osamura RY, et al. Polypoid carcinoma of the esophagus. A unifying term for “carcinosarcoma” and “pseudosarcoma”. Am J Surg Pathol 1978; 2: 201 – 8. 36. Kuhajda FP, Sun TT, Mendelsohn G. Polypoid squamous carcinoma of the esophagus. A case report with immunostaining for keratin. Am J Surg Pathol 1983; 7: 495 – 9. 37. Amatya VJ, et al. Esophageal carcinosarcoma with basaloid squamous carcinoma and rhabdomyosarcoma components with TP53 mutation. Pathol Int 2004; 54: 803 – 9.
38. Hughes JH, Cruickshank AH. Pseudosarcoma of the oesophagus. Br J Surg 1969; 56: 72 – 6. 39. Martin MR, Kahn LB. So-called pseudosarcoma of the esophagus: nodal metastases of the spindle cell element. Arch Pathol Lab Med 1977; 101: 604 – 9. 40. Hay-Roe V, Hill RL, Civin WH. An unclassifiable tumor of the esophagus. A case report. J Thorac Cardiovasc Surg 1960; 40: 107 – 13. 41. Hamabe Y, et al. (Carcinoma of the esophagus with sarcomatous elements) (Japanese). Geka Chiryo 1985; 52: 255 – 64. 42. Sasajima K, et al. Polypoid squamous cell carcinoma of the esophagus. Cancer 1989; 64: 94 – 7. 43. Pourzand A, et al. Primary adenoid cystic carcinoma of the esophagus. Report of a case and review of the literature. J Thorac Cardiovasc Surg 1975; 69: 785 – 9. 44. Epstein JI, et al. Carcinoma of the esophagus with adenoid cystic differentiation. Cancer 1984; 53: 1131 – 6. 45. Suzuki H, Nagayo T. Primary tumors of the esophagus other than squamous cell carcinoma – histologic classification and statistics in the surgical and autopsied materials in Japan. Int Adv Surg Oncol 1980; 3: 73 – 109. 46. Petursson SR. Adenoid cystic carcinoma of the esophagus. Complete response to combination chemotherapy. Cancer 1986; 57: 1464 – 7. 47. Tsang WY, et al. Basaloid-squamous carcinoma of the upper aerodigestive tract and so-called adenoid cystic carcinoma of the oesophagus: the same tumour type? Histopathology 1991; 19: 35 – 46. 48. Rotmensch J, Herbst A. Neoplasms of vulva and vagina. In Holland JF, Frei E, Bast R (eds) Cancer Medicine, 3rd ed, Lea and Febiger, 1993: Vol. 2: 1620 – 1630. 49. Azzopardi JG, Menzies T. Primary oesophageal adenocarcinoma. Br J Surg 1962; 49: 497 – 506. 50. Kaushik N, et al. Basaloid squamous cell cancer arising in Barrett’s esophagus. Int J Gastrointest Cancer 2003; 34: 95 – 9. 51. Sweeney EC, Cooney T. Adenoid cystic carcinoma of the esophagus: a light and electron microscopic study. Cancer 1980; 45: 1516 – 25. 52. Eby LS, Johnson DS, Baker HW. Adenoid cystic carcinoma of the head and neck. Cancer 1972; 29: 1160 – 8. 53. Spiro RH, Huvos AG, Strong EW. Adenoid cystic carcinoma of salivary origin. A clinicopathologic study of 242 cases. Am J Surg 1974; 128: 512 – 20. 54. Raven RW. Rare tumors of the pharynx and esophagus. Ann N Y Acad Sci 1964; 114: 1061 – 79. 55. Zhang XH, Sun GQ, Zhou XJ. Basaloid squamous carcinoma of esophagus: a clinicopathological, immunohistochemical and electron microscopic study of sixteen cases. World J Gastroenterol 1998; 4: 397 – 403. 56. Lewin KJ, Appelman HD. Tumors of the esophagus and stomach. Atlas of Tumor Pathology. Third Series, Fasc. 18, Washington, District of Columbia: Armed Forces Institute of Pathology, 1996. 57. Li TJ, et al. Basaloid squamous cell carcinoma of the esophagus with or without adenoid cystic features. Arch Pathol Lab Med 2004; 128: 1124 – 30. 58. Mukada T, Sato E. Adenoid cystic carcinoma of the esophagus. Tohoku J Exp Med 1974; 113: 257 – 67. 59. Nelms DC, Luna MA. Primary adenocystic carcinoma (cylindromatous carcinoma) of the esophagus. Cancer 1972; 29: 440 – 3. 60. O’Sullivan JP, Cockburn JS, Drew CE. Adenoid cystic carcinoma of the oesophagus. Thorax 1975; 30: 476 – 80. 61. Raphael HA, Ellis FH, Dockerty MB Jr. Primary adenocarcinoma of the esophagus: 18-year review and review of literature. Ann Surg 1966; 164: 785 – 96. 62. Bergmann M, Charnas RM. Tracheobronchial rests in the esophagus: their relation to some benign structures and certain types of cancer of the esophagus. J Thorac Surg 1958; 35: 97 – 104. 63. Eveson JW, Cawson RA. Salivary gland tumours. A review of 2410 cases with particular reference to histologic types, site, age and sex distribution. J Pathol 1985; 146: 51. 64. Storey MR, et al. Postoperative radiotherapy for malignant tumors of the submandibular gland. Int J Radiat Oncol Biol Phys 2001; 51: 952 – 8.
UNCOMMON CANCERS OF THE ESOPHAGUS 65. Armstrong JG, et al. A matched-pair analysis of the role of combined surgery and postoperative radiotherapy. Arch Otolaryngol Head Neck Surg 1990; 116: 290 – 3. 66. Dreyfuss AI, et al. Cyclophosphamide, doxorubicin and cisplatin combination chemotherapy for advanced carcinomas of salivary gland origin. Cancer 1987; 60: 2869 – 72. 67. Dimery IW, et al. Fluorouracil, doxorubicin, cyclophosphamide and cisplatin combination chemotherapy for advanced or recurrent salivary gland carcinoma. J Clin Oncol 1990; 8: 1056 – 62. 68. Laurie SA, et al. Chemotherapy in the management of metastatic adenoid cystic carcinoma: a systematic review. J Clin Oncol 2005; 23(16S): 5581. 69. Alcedo JC, et al. Imatinib mesylate as treatment for adenoid cystic carcinoma of the salivary glands: report of two successfully treated cases. Head Neck 2004; 26: 829 – 31. 70. Hotte SJ, et al. Imatinib mesylate in patients with adenoid cystic cancers of the salivary glands expressing c-kit: a Princess Margaret Hospital phase II consortium study. J Clin Oncol 2005; 23: 585 – 90. 71. Mallory TB. Case records of the Massachusetts General Hospital Case 21521. N Engl J Med 1935; 213: 1311 – 5. 72. Dodge OG. Gastro-esophageal carcinoma of mixed histological type. J Pathol Bacteriol 1961; 81: 459 – 71. 73. Mafune K. (Histochemical study of adenocarcinomatous components in carcinoma of the esophagus) (Japanese). Nippon Geka Gakkai Zasshi 1988; 89: 162 – 72. 74. Lortat-Jacob JL, et al. Primary esophageal adenocarcinoma: report of 16 cases. Surgery 1968; 64: 535 – 43. 75. Kay S. Mucoepidermoid carcinoma of the esophagus. Report of two cases. Cancer 1968; 22: 1053 – 9. 76. Yachida S, et al. Adenosquamous carcinoma of the esophagus: clinicopathologic study of 18 cases. Oncology 2004; 66: 218 – 25. 77. Fegelman E, et al. Squamous cell carcinoma of the esophagus with mucin-secreting component. Mucoepidermoid carcinoma. J Thorac Cardiovasc Surg 1994; 107: 62 – 7. 78. Matsufuji H, et al. Mucoepidermoid carcinoma of the esophagus – a case report. Jpn J Surg 1985; 15: 55 – 9. 79. Sasajima K, et al. Mucoepidermoid carcinoma of the esophagus: report of two cases and review of the literature. Endoscopy 1990; 22: 140 – 3. 80. Kawai H, et al. Alpha-fetoprotein producing esophageal carcinoma: a case report. Anticancer Res 2003; 23: 3837 – 40. 81. Osamura RY, et al. Mucoepidermoid carcinoma of the esophagus. Report of an unoperated autopsy case and review of literature. Am J Gastroenterol 1978; 69: 467 – 70. 82. Weitzner S. Mucoepidermoid carcinoma of esophagus. Report of a case. Arch Pathol 1970; 90: 271 – 3. 83. Woodard BH, et al. Mucoepidermoid carcinoma of the esophagus: a case report. Hum Pathol 1978; 9: 352 – 4. 84. Lam KY, et al. Squamous cell carcinoma of the oesophagus with mucin-secreting component (muco-epidermoid carcinoma and adenosquamous carcinoma): a clinicopathologic study and a review of literature. Eur J Surg Oncol 1994; 20: 25 – 31. 85. Pascal RR, Clearfield HR. Mucoepidermoid (adenosquamous) carcinoma arising in Barrett’s esophagus. Dig Dis Sci 1987; 32: 428 – 32. 86. Hagiwara N, et al. Biological behavior of mucoepidermoid carcinoma of the esophagus. J Nippon Med Sch 2003; 70: 401 – 7. 87. Oberndofer S. Karzinole tumoren des dunndarmes. Z Pathol 1907; 1: 426 – 32. 88. Ready AR, et al. Malignant carcinoid of the oesophagus. Thorax 1989; 44: 594 – 6. 89. Brenner S, et al. Carcinoid of esophagus. N Y State J Med 1969; 69: 1337 – 9. 90. Hauser H, et al. Neuroendocrine tumors in various organ systems in a ten-year period. Eur J Surg Oncol 1995; 21: 297 – 300. 91. Modlin IM, Sandor A. An analysis of 8305 cases of carcinoid tumors. Cancer 1997; 79: 813 – 29. 92. Hoang MP, et al. Carcinoid tumor of the esophagus : a clinicopathologic study of four cases. Am J Surg Pathol 2002; 26: 517 – 22.
349
93. Modlin IM, Shapiro MD, Kidd M. An analysis of rare carcinoid tumors: clarifying these clinical conundrums. World J Surg 2005; 29: 92 – 101. 94. Saw EC, et al. Synchronous primary neuroendocrine carcinoma and adenocarcinoma in Barrett’s esophagus. J Clin Gastroenterol 1997; 24: 116 – 9. 95. Lindberg G, et al. Atypical carcinoid of the esophagus: a case report and review of the literature. Cancer 1997; 79: 1476 – 81. 96. Tateishi R, et al. Argyrophil cells and melanocytes in esophageal mucosa. Arch Pathol Lab Med 1974; 98: 87 – 9. 97. Cary NR, et al. Combined oesophageal adenocarcinoma and carcinoid in Barrett’s oesophagitis: potential role of enterochromaffin-like cells in oesophageal malignancy. Thorax 1993; 48: 404 – 5. 98. Reyes CV, et al. Neuroendocrine carcinoma of the esophagus. Ultrastruct Pathol 1980; 1: 367 – 76. 99. Chefgec G, et al. Neuroendocrine tumors of the gastrointestinal tract. Pathol Res Pract 1988; 183: 143 – 54. 100. Arrigoni MG, et al. Atypical carcinoid tumors of the lung. J Thorac Cardiovasc Surg 1972; 64: 413 – 21. 101. Partensky C, et al. Five-year survival after transhiatal resection of esophageal carcinoid tumor with a lymph node metastasis. Cancer 1993; 72: 2320 – 2. 102. McKeown F. Oat cell carcinoma of the esophagus. J Pathol Bacteriol 1952; 64: 889 – 91. 103. Bennouna J, et al. Small cell carcinoma of the esophagus. Am J Clin Oncol 2000; 23: 455 – 9. 104. Law SYK, et al. Small cell carcinoma of the esophagus. Cancer 1994; 73: 2894 – 9. 105. Nishimaki T, et al. Primary small cell carcinoma of the esophagus with ectopic gastrin production. Report of a case and review of the literature. Dig Dis Sci 1993; 38: 767 – 71. 106. Tanabe G, et al. Effective chemotherapy for small cell carcinoma of the esophagus. Cancer 1987; 60: 2613 – 6. 107. Miettinen M, et al. Esophageal stromal tumors: a clinicopathologic, immunohistochemical, and molecular genetic study of 17 cases and comparison with esophageal leiomyomas and leiomyosarcomas. Am J Surg Pathol 2000; 24: 211 – 22. 108. Goldblum JR. Nonepithelial neoplasms of the GI tract. In Odze R, Goldblum J, Crawford J (eds) Surgical Pathology of the GI Tract, Liver, Biliary Tract and Pancreas. Philadelphia, Pennsylvania: Saunders, 2003: 505 – 522. 109. Rainer WG, Brus R. Leiomyosarcoma of the esophagus. Review of the literature and report of 3 cases. Surgery 1965; 58: 343 – 50. 110. Levine MS, et al. Leiomyosarcoma of the esophagus: radiographic findings in 10 patients. AJR Am J Roentgenol 1996; 167: 27 – 32. 111. Choh JH, Khazei AH, Ihm HJ. Leiomyosarcoma of the esophagus: report of a case and review of the literature. J Surg Oncol 1986; 32: 223 – 6. 112. Athanasoulis CA, Aral IM. Leiomyosarcoma of the oesophagus. Gastroenterology 1968; 54: 271 – 4. 113. Weinstein EC, et al. Leiomyosarcoma of the esophagus. Mil Med 1988; 4: 206 – 9. 114. Franklin GO, et al. Esophageal leiomyosarcoma. N Y State J Med 1982; 82: 1100 – 3. 115. Aimoto T, et al. Leiomyosarcoma of the esophagus: report of a case and preoperative evaluation by CT scan, endoscopic ultrasonography and angiography. Gastroenterol Jpn 1992; 27: 773 – 9. 116. Carnishion RC, Gibbon JH, Templeton JY. Leiomyosarcomas of the esophagus. Ann Surg 1961; 153: 951 – 6. 117. Patel SR, Anandarao N. Leiomyosarcoma of the esophagus. N Y State J Med 1990; 90: 371 – 2. 118. Grigg ER. Esophagogastrointestinal leiomyosarcomas. Am J Med 1961; 31: 591 – 618. 119. Stanley WM, Groshong LR. Leiomyosarcoma of the gastrointestinal tract. Am Surg 1969; 35: 809 – 16. 120. Koga H, et al. Rapidly growing esophageal leiomyosarcoma: case report and review of the literature. Abdom Imaging 1995; 20: 15 – 9. 121. Balthazar EJ. Gastrointestinal leiomyosarcoma – unusual sites: esophagus, colon and porta hepatis. Gastrointest Radiol 1981; 6: 295 – 301. 122. Gaede JT, et al. Leiomyosarcoma of the esophagus. Report of two cases, one with associated squamous cell carcinoma. J Thorac Cardiovasc Surg 1978; 75: 740 – 6.
350
GASTROINTESTINAL TUMORS
123. Savage DC, Antman KH. Imatinib mesylate – a new oral targeted therapy. N Engl J Med 2002; 346: 683 – 93. 124. Joensuu H, et al. Effect of the tyrosine kinase inhibitor STI571 in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med 2001; 344: 1052 – 6. 125. van Oosterom AT, et al. Safety and efficacy of imatinib (STI571) in metastatic gastrointestinal stromal tumors: a phase I study. Lancet 2001; 358: 1421 – 3. 126. Tran J, Davila JA, El-Serag HB. The epidemiology of malignant gastrointestinal stromal tumors: an analysis of 1,458 cases from 1992 – 2000. Am J Gastroenterol 2005; 100: 162 – 8. 127. Wada Y, et al. Esophageal gastrointestinal stromal tumor surrounding the middle esophagus with dysphagia for 8 years. Kyobu Geka 2004; 57: 1250 – 3. 128. Ohnishi T, Yoshioka H, Ishida O. MR imaging of gastrointestinal leiomyosarcoma. Radiat Med 1991; 9: 114 – 7. 129. Tio TL, Tytgat GN, den Hartog Jager FC. Endoscopic ultrasonography for the evaluation of smooth muscle tumors in the upper gastrointestinal tract: an experience with 42 cases. Gastrointest Endosc 1990; 36: 342 – 50. 130. Hirota S, et al. Gain-of-function mutations of c-kit in human gastrointestinal stromal tumors. Science 1998; 279: 577 – 80. 131. Fletcher CDM, et al. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Hum Pathol 2002; 33: 459 – 65. 132. Silverberg SG. Reproducibility of the mitosis count in the histologic diagnosis of smooth muscle tumors of the uterus. Hum Pathol 1976; 7: 451 – 4. 133. Matsumori M, et al. A two-stage operation successfully performed for giant leiomyosarcoma of the esophagus with hepatic metastasis. Surg Today 1992; 22: 543 – 7. 134. Pramesh CS, et al. Leiomyosarcoma of the esophagus. Dis Esophagus 2003; 16: 142 – 4. 135. Goodner JT, Miller TR, Watson WL. Sarcoma of the esophagus. Am J Roentgenol 1963; 89: 132 – 9. 136. Blanke CD, et al. Evaluation of the safety and efficacy of an oral molecularly-targeted therapy STI571 in patients with unresectable or metastatic gastrointestinal stromal tumors (GISTs) expressing c-kit (CD117). Proc Am Soc Clin Oncol 2001; 20: 1a; abstract. 137. Perch SJ, et al. Esophageal sarcomas. J Surg Oncol 1991; 48: 194 – 8. 138. Enzinger FM, Weiss SW. Liposarcoma. In Enzinger FM, Weiss SW (eds) Soft Tissue Tumors, 2nd ed. St. Louis, Missouri: CV Mosby, 1988: 346 – 382. 139. Garcia M, et al. Large esophageal liposarcoma : a case report and review of the literature. Arch Pathol Lab Med 2004; 128: 922 – 5. 140. Yates SP, Collins MC. Case report: recurrent liposarcoma of the oesophagus. Clin Radiol 1990; 42: 356 – 8. 141. Ruppert-Kohlmayr AJ, et al. Giant liposarcoma of the esophagus: radiologic findings. J Thorac Imaging 1999; 14: 316 – 9. 142. Beaudoin A, et al. Giant liposarcoma of the esophagus. Can J Gastroenterol 2002; 16: 377 – 9. 143. Chung JJ, et al. Imaging findings of giant liposarcoma of the esophagus. Yonsei Med J 2003; 44: 715 – 8. 144. Salis GB, et al. Pedunculated liposarcoma of the esophagus. Dis Esophagus 1998; 11: 68 – 71. 145. Geboes K, et al. Primary malignant fibrous histiocytoma of the esophagus. J Surg Oncol 1989; 40: 49 – 57. 146. Wright JR Jr, Kyriakos M, DeSchryver-Kecskemeti K. Malignant fibrous histiocytoma of the stomach. A report and review of malignant fibrohistiocytic tumors of the alimentary tract. Arch Pathol Lab Med 1988; 112: 251 – 8. 147. Fletcher CD. Malignant fibrous histiocytoma? Histopathology 1987; 11: 433 – 7. 148. Naganuma H, et al. Malignant fibrous histiocytoma of the esophagus. Pathol Int 1996; 46: 462 – 6. 149. Bloch MJ, et al. Polypoid synovial sarcoma of the esophagus. Gastroenterology 1987; 92: 229 – 33. 150. Cadman NL, Soule EH, Kelly PJ. Synovial sarcoma. An analysis of 134 tumors. Cancer 1965; 18: 613 – 27. 151. Bonavina L, et al. Synovial sarcoma of the esophagus simulating achalasia. Dis Esophagus 1998; 11: 268 – 71. 152. Shah R, et al. Rhabdomyosarcoma of the esophagus: a case report. J Cardiovasc Surg 1995; 36: 99 – 100.
153. Fox KR, et al. Clinical and pathologic features of primary gastric rhabdomyosarcoma. Cancer 1990; 66: 772 – 8. 154. Magovern CJ, et al. Primary inflammatory fibrosarcoma of the esophagus. Ann Thorac Surg 1996; 62: 1848 – 50. 155. McIntyre M, Webb JN, Browning GC. Osteosarcoma of the esophagus. Hum Pathol 1982; 13: 680 – 2. 156. Yaghmai I, Ghahremani GG. Chondrosarcoma of the esophagus. Am J Roentgenol 1976; 126: 1175 – 7. 157. Sanchez A, et al. Malignant nerve sheath tumor of the esophagus (malignant esophageal schwannoma). Gastroenterol Hepatol 2004; 27: 467 – 9. 158. Wasan HS, et al. Combined choriocarcinoma and yolk sac tumor arising in Barrett’s esophagus. Cancer 1994; 73: 514 – 7. 159. Toner GC, et al. Extragonadal and poor risk nonseminomatous germ cell tumors. Survival and prognostic features. Cancer 1991; 67: 2049 – 57. 160. Merimsky O, et al. Choriocarcinoma arising in a squamous cell carcinoma of the esophagus. Am J Clin Oncol 2000; 23: 203 – 6. 161. Motoyama T, Higuchi M, Taguchi J. Combined choriocarcinoma, hepatoid adenocarcinoma, small cell carcinoma and tubular adenocarcinoma in the oesophagus. Virchows Arch 1995; 427: 451 – 4. 162. Trillo AA, Accettullo LM, Yeiter TL. Choriocarcinoma of the esophagus: histologic and cytologic findings. A case report. Acta Cytol 1978; 23: 69 – 74. 163. McKechnie JC, Fechner RE. Choriocarcinoma and adenocarcinoma of the esophagus with gonadotrophin secretion. Cancer 1971; 27: 694 – 702. 164. Kikuchi Y, et al. Choriocarcinoma of the esophagus producing chorionic gonadotropin. Acta Pathol Jpn 1988; 38: 489 – 99. 165. Sasano N, et al. Choriocarcinoma mimicry of an esophageal carcinoma with urinary gonadotropic activities. Tohoku J Exp Med 1970; 100: 153 – 63. 166. Ulbright TM, Roth LM, Brodhecker CA. Yolk sac differentiation in germ cell tumors. A morphologic study of 50 cases with emphasis on hepatic, enteric, and parietal yolk sac features. Am J Surg Pathol 1986; 10: 151 – 64. 167. Ishihara A, Mori T, Koono M. Diffuse pagetoid squamous cell carcinoma of the esophagus combined with choriocarcinoma and mucoepidermoid carcinoma: an autopsy case report. Pathol Int 2002; 52: 147 – 52. 168. Baur EH. Ein fall von primaerem melanom des oesophagus. Arb Geb Pathol Anat Inst Tubingen 1906; 5: 343 – 54. 169. Turnbull AD, et al. Primary malignant tumors of the esophagus other than typical epidermoid carcinoma. Ann Thorac Surg 1973; 15: 463 – 73. 170. Suzuki H, Nagayo T. Primary tumors of the esophagus other than squamous cell carcinoma. Histologic classification and statistics in the surgical and autopsied materials in Japan. Adv Surg Oncol 1980; 3: 73 – 109. 171. DiCostanzo DP, Urmacher C. Primary malignant melanoma of the esophagus. Am J Surg Pathol 1987; 11: 46 – 52. 172. Chalkiadakis G, et al. Primary malignant melanoma of the esophagus. Ann Thorac Surg 1985; 39: 472 – 5. 173. Sabanathan S, Eng J, Pradhan GN. Primary malignant melanoma of the esophagus. Am J Gastroenterol 1989; 84: 1475 – 81. 174. Yoo CC, et al. Primary malignant melanoma of the esophagus: radiographic findings in seven patients. Radiology 1998; 209: 455 – 9. 175. Gollub MJ, Prowda JC. Primary melanoma of the esophagus: radiologic and clinical findings in six patients. Radiology 1999; 213: 97 – 100. 176. Namieno T, et al. Primary malignant melanoma of the esophagus: diagnostic value of endoscopic ultrasonography. Am Surg 1996; 62: 716 – 8. 177. De la Pava S, et al. “Melanosis” of the esophagus. Cancer 1963; 16: 48 – 50. 178. Allen AA, Spitz SL. Malignant melanoma: clinicopathological analysis of criteria for diagnosis and prognosis. Cancer 1953; 6: 1 – 13. 179. Piccone VA, et al. Primary malignant melanoma of the esophagus associated with melanosis of the entire esophagus: first case report. J Thorac Cardiovasc Surg 1970; 59: 864 – 70.
UNCOMMON CANCERS OF THE ESOPHAGUS 180. DeMatos P, et al. Primary malignant melanoma of the esophagus. J Surg Oncol 1997; 66: 201 – 6. 181. Joob AW, et al. Primary malignant melanoma of the esophagus. Ann Thorac Surg 1995; 60: 217 – 22. 182. Cuesta-Mejias T, et al. Primary melanoma of the esophagus: unusual case with Barrett’s esophagus. Rev Gastroenterol Mex 2001; 66(3): 146 – 9. 183. Lam KY, Law S, Wong J. Malignant melanoma of the oesophagus: clinicopathologic features, lack of p53 expression and steroid receptors and a review of the literature. Eur J Surg Oncol 1999; 25: 168 – 72. 184. Iverson K, Robins RE. Mucosal malignant melanomas. Am J Surg 1980; 139: 659 – 64. 185. Chello M, et al. Primary malignant melanoma of the oesophagus with a left atrial metastasis. Thorax 1993; 48: 185 – 6. 186. Caldwell CB, Bains MS, Burt M. Unusual malignant neoplasms of the esophagus: oat cell carcinoma, melanoma, and sarcoma. J Thorac Cardiovasc Surg 1991; 101: 100 – 7. 187. Keeley JL, et al. Primary malignant melanoma of the esophagus. Surgery 1957; 42: 607 – 14. 188. Ludwig ME, et al. Primary malignant melanoma of the esophagus. Cancer 1981; 48: 2528 – 34. 189. Mukaiya M, et al. Surgical treatment for recurrent tumors of primary malignant melanoma of the esophagus: a case report and review of the literature. Hepatogastroenterology 1999; 46: 295 – 8. 190. Woolery WA, Tripodis SP, Chacko DC. Endoscopic Nd:YAG laser and radiation therapy for primary malignant melanoma of the esophagus. J Am Osteopath Assoc 1990; 90: 543 – 6.
351
191. Naomoto Y, et al. Primary malignant melanoma of the esophagus: report of a case successfully treated with pre- and post-operative adjuvant hormone-chemotherapy. Jpn J Clin Oncol 1998; 28: 758 – 61. 192. Suzuki Y, et al. Amelanotic malignant melanoma of the esophagus: report of a patient with recurrence successfully treated with chemoendocrine therapy. Int J Clin Oncol 2005; 10: 204 – 7. 193. Antler AS, et al. Gastrointestinal metastases from malignant tumors of the lung. Cancer 1982; 49: 170 – 2. 194. Toreson WE. Secondary carcinoma of the esophagus as a cause of dysphagia. Arch Pathol 1944; 38: 82 – 4. 195. Luomanen RKJ, Watson WL. Autopsy findings. In Watson WL (ed) Lung Cancer: A Study of Five Thousand Memorial Hospital Cases. St Louis, Missouri: CV Mosby, 1968: 505 – 510. 196. Asch MJ, Wiedel PD, Habif DV. Gastrointestinal metastasis from carcinoma of the breast. Autopsy study and 18 cases requiring operative intervention. Arch Surg 1968; 96: 840 – 3. 197. Agha FP. Secondary neoplasms of the esophagus. Gastrointest Radiol 1987; 12: 187 – 93. 198. Simeone J, Burrell M, Toffler R. Esophageal aperistalsis secondary to metastatic invasion of the myenteric plexus. Am J Roentgenol 1976; 127: 862 – 4. 199. Middleton RG. Surgey for metastatic renal cell carcinoma. J Urol 1967; 97: 973 – 7. 200. Asamura H, et al. Esophageal and pulmonary metastases from ovarian carcinoma: a case report of long-term survival following metastatic resections. Jpn J Clin Oncol 1991; 21: 211 – 7. 201. Phadke M, Rao U, Takita H. Metastatic tumors of esophagus. N Y State J Med 1976; 76: 963 – 5.
Section 6 : Gastrointestinal Tumors
31
Uncommon Cancers of the Stomach Jordan D. Berlin and Mary K. Washington
INTRODUCTION Although gastric cancer is not common in the United States, it remains one of the leading causes of cancer death worldwide, accounting for nearly 650 000 deaths per year.1 Incidence of gastric cancer in the United States has declined from >45 per 100 000 in 1930 to <10 per 100 000 currently.2 Despite the overall decline in gastric cancer incidence, there has been a recent rise in proximal (gastric cardia) stomach cancers. In 2005, an anticipated 21 860 new cases of stomach cancer will occur in the United States with 11 550 deaths. There is a male predominance with a male : female ratio of 1.6 : 1. There is a geographic distribution to gastric cancer with areas of high incidence in South America and Asia. When people from areas of high incidence move to areas of low incidence, they retain their risk for gastric cancer.3 However, the second generation has the same risk as other people in the new low incidence region. This suggests an environmental cause to most gastric cancers. A variety of dietary factors, such as n-nitrosamines, have been implicated as potential causes for cancers; others, such as fruits and vegetables, have been considered protective. Helicobacter pylori infection, which can produce chronic inflammation of the stomach, has been associated with gastric adenocarcinomas and lymphomas. Over 90% of gastric cancers are adenocarcinoma. These are typically subclassified by the Lauren classification as either diffuse or intestinal type, on the basis of their histologic appearance. Diffuse tumors infiltrate deeply into the stomach wall, often exhibiting desmoplasia and sparing the mucosal layer.3 They infiltrate in a diffuse pattern and exhibit little gland formation. The intestinal type exhibits typical gland formation, often arising on a background of intestinal metaplasia. Intestinal-type adenocarcinomas are more often localized to the antrum while diffuse cancers are associated with a worse prognosis. Regardless of histology, gastric cancers most commonly present with nonspecific gastrointestinal (GI) symptoms such as dyspepsia, early satiety, nausea, and anorexia. Weight loss and abdominal pain are common. When localized to the stomach, gastric cancer produces few physical examination findings. It can spread to palpable lymph nodes such
as the left supraclavicular node (Virchow node) or periumbilical nodes (Sister Mary Joseph node) or left axillary node (Irish node). Gastric cancer is usually diagnosed by esophagogastroduodenoscopy (EGD) and biopsy. Staging is by computed tomography (CT) scan, endoscopic ultrasound (EUS) and laparoscopy. These three studies are complementary although they are not all available and/or standard at every institution. Treatment of localized gastric adenocarcinoma is surgery. Several studies have evaluated the appropriate extent of lymph node dissection.3 Currently, data support an extended lymphadenectomy. In western countries, the majority of patients are detected at advanced stages. Even when resectable, median survival is only 25 months and 30% are alive at 5 years.1 However, two recent studies have suggested that the addition of either postoperative chemoradiation or pre- and postoperative chemotherapy can improve upon the results obtained with surgery alone.4,5 When unresectable, the primary therapy of gastric adenocarcinoma is chemotherapy. Although several randomized trials exist, there is no single, standard chemotherapy regimen.1 Median survival for patients with advanced gastric adenocarcinomas is 8–10 months. In this chapter, we will consider gastric malignancies other than adenocarcinomas. These uncommon cancers of the stomach account for <10% of gastric cancers. Some of the rare tumor types such as undifferentiated carcinoma with lymphoid stroma and hepatoid adenocarcinoma and adenosquamous carcinoma are rare variants of adenocarcinoma. Primary germ cell tumors of the stomach and gastrointestinal stromal tumors (GISTs) are found in a variety of organ sites, including the stomach. Neuroendocrine tumors (endocrine cell proliferation) of the stomach and lymphomas are also found in a variety of organ sites, but those that occur in the stomach have some characteristic features unique to gastric primaries. Finally, parietal cell carcinoma appears to be extremely rare and may be unique to the stomach. In this chapter, we will review these uncommon gastric cancers, including epidemiology, pathology, clinical presentation, natural history, and treatment.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
UNCOMMON CANCERS OF THE STOMACH
UNDIFFERENTIATED CARCINOMA WITH LYMPHOID STROMA Background Knowledge of undifferentiated gastric carcinomas with lymphoid stroma is evolving with the development of molecular biology techniques. It has been alternatively called gastric carcinoma with lymphoid stroma (GCLS), and lymphoepithelioma-like carcinoma (LELC).6,7 With either name, this entity is most commonly associated with EpsteinBarr Virus (EBV) although there are cases with the morphologic features in whom EBV-encoded RNA was not detected.8 In addition, EBV has also been occasionally associated with gastric adenocarcinoma.9,10 Although LELC occurs in other organs, it has only been seen in organs of foregut embryologic origin.7 Historically, gastric carcinoma with lymphoid stroma accounts for 1–4% of gastric carcinomas.6,11
Pathology Grossly, they are usually bulky tumors with a homogeneous gray cut surface that resembles lymphoma.12 Lymph node metastases are less common than in typical gastric carcinomas. Microscopically, these tumors have a diffuse growth pattern with a pushing interface and are composed of sheets, syncytial aggregates, trabeculae, and rudimentary tubules of small polygonal cells (see Figure 1). The tumor cells are relatively monomorphic, with round vesicular nuclei and relatively scant eosinophilic to pale cytoplasm. Tumorinfiltrating lymphocytes are a prominent feature; most of the inflammatory cells are CD8 T lymphocytes, but plasma cells and germinal centers are common. The neoplastic cells may be widely separated by the intervening inflammatory infiltrate. Foreign body-type giant cell reaction may also be seen.13 Intranuclear accumulation of EBER-1 (EBV-encoded small RNA) may be demonstrated in most cases by in situ hybridization.14 In some cases, it may be necessary to use
353
immunohistochemistry to demonstrate keratin expression by the tumor cells to distinguish these carcinomas from gastric lymphoma. While H. Pylori infection is associated with gastric adenocarcinoma, it is found in only 22% of patients with GCLS.11
Clinical Presentation Most series of GCLS demonstrate a male predominance (2–3 : 1, male : female).6,8,11 Median age has been reported to be 55–58 years.7,10 GCLS has been seen in all areas of the stomach, although it appears to more commonly present in the proximal stomach.11 EBV-associated GCLS has also been reported to present as a multicentric tumor.8 GCLS appears to present with earlier stage and less lymph node involvement (33% at presentation) and a mean tumor size <3 cm.11
Treatment Most patients present at an early enough stage to undergo surgery. Results for surgery appear better than for adenocarcinoma patients.6,8,11 For metastatic disease, excellent response to multiagent chemotherapy has been reported for a single case.8 This report used a combination of systemic mitomycin C, doxorubicin, and cyclophosphamide in combination with hepatic arterial doxorubicin and 5-fluorouracil (5-FU), followed by intrarectal tegafur and two experimental agents given subcutaneously. It is unlikely this regimen will be studied further, and it is unknown what impact the individual components had.
Prognosis Most reports suggest that GCLS has a better prognosis than similarly staged gastric adenocarcinoma. Five-year survival for patients without invasion to the serosa was 97–100%, while it was 76–91% for gastric adenocarcinoma controls.6 With serosal involvement, survival for GCLS at 5 years is still 73.5% while it was only 35% for gastric adenocarcinoma patients (p < 0.001).
Recommendations Surgical resection results in an excellent prognosis. With such an excellent prognosis, it is unclear if there is a role for adjuvant therapy. For metastatic disease, there is not enough data to recommend a specific chemotherapy regimen.
HEPATOID ADENOCARCINOMA Background
Figure 1 Undifferentiated carcinoma with lymphoid stroma. The tumor cells grow in a diffuse pattern as solid nests interspersed among the prominent mononuclear cell inflammatory infiltrate.
Gastric carcinoma producing alpha fetoprotein (AFP) was first reported in 1970.15 Since that time, studies have demonstrated that up to 6% of gastric carcinomas produce AFP; however, not all exhibit hepatoid features. Nonhepatoid AFPproducing gastric carcinomas are well-differentiated tubular or papillary carcinomas with clear cytoplasm, and have a better clinical prognosis than their hepatoid counterpart.12 In 1985, Ishikura et al. proposed that AFP-producing tumors
354
GASTROINTESTINAL TUMORS
with hepatoid features constitute a distinct pathological entity.16
Pathology Although several morphologic subtypes of gastric adenocarcinoma are associated with AFP, hepatoid carcinomas have a distinctive microscopic appearance, with the tumor cells resembling hepatocytes to a greater or lesser extent, depending upon the degree of differentiation.12 The hepatic differentiation is thought to be due to the common embryologic foregut origin of the liver and the stomach. The hepatoid areas consist of large polygonal cells with abundant eosinophilic cytoplasm resembling nonneoplastic hepatocytes or hepatocellular carcinoma, and are usually admixed with areas of more typical gastric intestinaltype adenocarcinoma.17 In some cases, bile production is present. In approximately half of hepatoid gastric carcinomas, AFP may be demonstrated in the tumor cells by immunohistochemistry,18 similar to the percentage of hepatocellular carcinomas showing AFP accumulation. The tumor cells may also produce other factors synthesized by hepatocytes, such as albumin, prothrombin, α-1-chymotrypsin and α-1-antitrypsin, the latter may be visualized as periodic acid Schiff-positive cytoplasmic globules. Pronounced vascular invasion is a common feature.19
Clinical Presentation In a review of 59 cases reported from the literature in 1997, the majority of patients were male (2 : 1 ratio) with a mean age at presentation of 63.20 Data on AFP was available on 38 of the 59 cases, of whom 34 had elevated levels. AFP levels ranged from normal to 700 000 ng ml−1 , with a mean level over 50 000 ng ml−1 . Fifty-three of 59 patients presented with metastases as follows: 41 with lymph node metastases, 17 had liver metastases, 2 had omental metastases, and 1 each had involvement of the pancreas, colon, and peritoneum. Additionally, hepatoid adenocarcinoma can be locally aggressive, presenting with venous invasion and the capability of forming tumor thrombi.21 The antrum of the stomach is the most common site and is involved in up to 60% of cases.20,22 However, hepatoid adenocarcinoma may occur anywhere in the stomach and has even presented with the appearance of linitis plastica.20,22,23 The mean size of the primary tumor ranges from 6.5 to 8.7 cm.18,22 Clinical presentation does not appear to differ significantly from other forms of gastric cancer with epigastric pain, fatigue, and anemia among the most common symptoms and findings.22
Treatment The majority of cases have been treated with surgical resection. Most of the time the AFP decreases after surgery, but in one review, only one patient could be found with normalization of the AFP after surgery.20 Postoperative recurrence appears common and even early stage hepatoid adenocarcinoma patients appear to have a poor prognosis.18,22 One possible explanation for the aggressive behavior may be the extensive venous invasion including tumor thrombi of the portal venous system observed by some investigators.21
Radiation has been used in adjuvant therapy after surgical resection but no data is available on efficacy.22 Limited preclinical data suggests that AFP-producing variants of gastric cancers are resistant to chemotherapy.24 In the clinic, chemotherapy regimens commonly used for gastric adenocarcinoma have been tried with responses in some patients.22 Regimens combining 5-FU, doxorubicin, and mitomycin C or 5-FU and cisplatin have been most commonly used.
Prognosis Survival for hepatoid adenocarcinoma is poor, with the largest single series demonstrating a 5-year survival of 11.9%.18 It is unclear if this poor survival reflects a biologically aggressive tumor, the advanced stage at presentation, or a combination of these two factors.
Recommendations The only treatment of hepatoid adenocarcinoma that has resulted in long-term survival is surgical resection. Most of the case reports and series focus on pathologic features, and little data is available regarding adjuvant therapy or treatment of metastatic disease. In the absence of further information, it is reasonable to treat localized disease with neoadjuvant and adjuvant regimens used for adenocarcinoma of the stomach. Similarly, systemic regimens used for metastatic gastric adenocarcinoma are probably appropriate, although it is not known if hepatoid adenocarcinomas will respond as well to chemotherapy.
ADENOSQUAMOUS AND SQUAMOUS CELL CARCINOMA Background The first case of squamous cell carcinoma of the stomach was reported in 1905.25 To be classified as adenosquamous, at least 25% of the neoplastic component of a gastric cancer should show squamous differentiation; roughly 0.5% of gastric carcinomas will meet these criteria. Pure squamous cell carcinomas are even rarer, and in most cases tumors initially regarded as pure squamous in differentiation show small foci of glandular differentiation.12
Pathology In both subtypes, the squamous carcinoma resembles that arising in other organs, and is characterized by solid or nesting growth pattern of polygonal cells with eosinophilic cytoplasm, sharp cell borders, and intercellular bridges.12 Unequivocal features of squamous differentiation (keratin pearl formation and/or intercellular bridges) must be demonstrated by light microscopy to differentiate these tumors from poorly differentiated adenocarcinomas with minimal gland formation and a solid growth pattern. Origin from the distal esophagus and gastric metastases from other organs should be excluded. Vascular invasion is often prominent. Given that squamous mucosa does not normally occur in the stomach of humans, the histogenesis of adenosquamous and squamous gastric carcinomas has been a matter of
UNCOMMON CANCERS OF THE STOMACH
debate. Molecular studies indicate that in adenosquamous carcinoma, the squamous and adenocarcinoma components share the same alterations in p53, p16, and retinomblastoma (RB) proteins, suggesting that both components arise from the same or a genetically related clone.26 Most likely these carcinomas arise from primitive cells that differentiate along both glandular and squamous epithelial lines, although origin in squamous metaplasia has been suggested.12
Clinical Presentation Pure squamous cell cancer is a male predominant disease (18 of 22 patients in the literature).27 Squamous cell cancer has been reported in a patient as young as 17, and may occur in general at an earlier age than gastric adenocarcinoma but can occur at any age.27,28 Clinical presentation is not significantly different from other gastric cancers, although significant bleeding from the tumor may also occur.27
Treatment Most cases reported in the literature have been treated with surgical resection. The only long-term survivors reported have had surgery.27 There is not enough data on chemotherapy or radiation effects to determine their efficacy.
Prognosis Conflicting reports on prognosis state that squamous and adenosquamous carcinoma have either better or worse prognosis than adenocarcinoma.27,29,30 However, of the 22 patients with pure squamous cell carcinoma reported in the literature, 20 died within 1 year of surgical resection.27
Recommendations Surgical resection is the treatment of choice. There is no data to suggest options for adjuvant therapy or treatment of metastatic disease.
PRIMARY GERM CELL TUMORS OF THE STOMACH Background Choriocarcinomas, teratomas, and yolk sac tumors all occur as primary gastric tumors and histologically resemble their counterparts in more typical sites. Although these tumors are rare, the stomach may be the most common site of nongonadal, nongestational, nonmidline germ cell tumors.12 Choriocarcinoma is more common than yolk sac tumor in this site. Gastric teratomas are most commonly reported in infants and small children and behave in a benign fashion.
Gastric Choriocarcinoma Background
Primary choriocarcinoma of the stomach was first reported in the German literature in 1905.31 Since that time, fewer than 100 cases have been reported in the literature. Pure choriocarcinoma is extremely rare as many are associated with adenocarcinoma. Because of the association with adenocarcinoma, the origin of gastric choriocarcinoma has been
355
postulated to occur from dedifferentiation (opisthoplasia) of adenocarcinoma cells. However, researchers have identified β-HCG expressing cells within the normal gastric mucosa. The role of these β-HCG expressing cells in either normal gastric function and/or in the development of choriocarcinoma has not been elucidated. Pathology
Grossly, gastric choriocarcinoma is more hemorrhagic than typical gastric cancers.12 In roughly 70% of cases, the tumors contain adenocarcinoma elements, generally as admixed small glands,32 and comparative genomic hybridization also shows genomic imbalances commonly found in typical gastric adenocarcinomas.33 Approximately one-fourth are pure choriocarcinoma, and the possibility of gastric metastases from more typical sites for germ cell neoplasms should be excluded. Teratomatous elements are not present, although yolk sac components have rarely been reported.12 Microscopically, the tumors consist of a mixture of cytotrophoblastic and syncytiotrophoblastic elements, with syncytiotrophoblast dominating in some tumors. Metastases from the choriocarcinoma component are typically found in the lung and liver while the adenocarcinoma component, if present, tends to metastasize to regional lymph nodes. Immunohistochemical studies for β-HCG demonstrate immunoreactivity in the tumor cells in gastric choriocarcinoma. However, roughly 25% of typical gastric adenocarcinomas demonstrate HCG-containing cells that do not resemble syncytiotrophoblast. Normal antral neck cells and patchy areas of intestinal metaplasia are also reportedly positive for HCG.12 Clinical Presentation
Choriocarcinoma is more commonly seen in men than in women, although a smaller review of patients in the literature from 1980 to 2000 had equal numbers of males and females.34 The median age of gastric choriocarcinoma patients ranges from 52 to 57 years.34,35 Abdominal pain, anorexia, nausea, vomiting, and weight loss are among the most common presenting symptoms of choriocarcinoma as they are for most forms of gastric cancer.34 – 36 Choriocarcinoma presents with GI bleeding and melena more often than adenocarcinoma, consistent with the gross pathologic findings. Because of the β-HCG production, choriocarcinoma patients may develop gynecomastia, precocious puberty, and/or a morning emesis pattern similar to that seen in pregnancy. Choriocarcinomas are rapidly progressive and have frequently presented with palpable gastric masses.35 Endoscopically, the tumors appear large and may show signs of bleeding. Diagnosis requires biopsy. Gastric adenocarcinomas without choriocarcinoma have been reported to produce β-HCG, so this test cannot be considered diagnostic. Treatment
The primary therapy for gastric choriocarcinoma has been surgery.34 – 36 Many patients are diagnosed only after surgical resection has been performed and others require surgery even in the setting of metastases due to the GI hemorrhage. While one case of long-term survival has been reported in the
356
GASTROINTESTINAL TUMORS
English medical literature, this appears rare.34 That patient received surgery followed by postoperative mitomycin C, cisplatin, and 5-FU followed by an oral fluoropyrimidine. The patient was identified as having choriocarcinoma prior to surgery and the investigators surmised that the preoperative diagnosis played a role in long-term survival. Other reports have demonstrated that choriocarcinoma of the stomach can respond to chemotherapy, but the response is short-lived.35,36 Although β-HCG levels are not diagnostic or prognostic in this disease, they may be followed during and after therapy to gauge response and/or recurrence.34
patients with elevated AFP to evaluate for evidence of yolk sac tumor. In the rare event of metastatic disease, surgical resection should still be considered the primary therapy. At this time, there is no evidence for adjuvant therapy of teratoma, including gastric teratoma.
Yolk Sac Tumor Background
With limited exceptions, survival for gastric choriocarcinoma is measured in months, regardless of therapy.
Pure yolk sac tumor of the stomach has only been reported once in the medical literature.39 The remaining four cases identified in the English medical literature contained adenocarcinomatous elements.40,41 Yolk sac tumor may also be seen within immature teratomas which have been discussed in the previous section.
Recommendations
Pathology
Treatment for this disease should be similar to choriocarcinoma originating from other organ sites. The combination of local therapy (surgery) with systemic therapy is generally undertaken.
The yolk sac tumors of the stomach that have been reported have been large masses within the stomach. Four of the five have had distinct components of yolk sac tumor elements–adenocarcinoma, tubular adenocarcinoma, and/or choriocarcinoma.40,41 Like other yolk sac tumors, the gastric yolk sac tumors have a reticular pattern and express placental alkaline phosphatase. A variety of markers has been detected in these lesions including alpha fetoprotein (AFP), CEA, cytokeratin, and gastrin. None of the tumors has expressed β-HCG.
Prognosis
Teratomas Background
Gastric teratoma is rare, accounting for <1% of all teratomas.37 It occurs in childhood and more frequently in males than in females. It usually arises from the greater curvature or posterior gastric wall, although a few cases of lesser curvature teratoma have been reported. Pathology
Histologically, teratomas contain a mixture of mature tissue elements derived from the three germ cell layers, similar to teratomas originating from other sites.12 Teratoma is divided into mature and immature forms. Immature teratoma is graded on a scale of 1–3 on the basis of the presence of immature elements and on mitotic activity. Clinical Presentation
These tumors generally present with symptoms of vomiting and abdominal distension.37 Upon physical examination, a palpable mass is usually found. Elevated AFP levels should be considered as an indication that there is a yolk sac component to the tumor. Treatment
Resection is the treatment of choice for teratoma regardless of type and site of origin.37,38 While there is no trial specific to primary gastric teratoma, there is a study from the combined children’s cooperative groups that establishes surgery as the only modality needed in immature teratomas from a variety of sites.38 Even with immature teratomas of the stomach, a review of the literature did not find reports of recurrence after surgery alone.37 Recommendations
Patients with primary gastric teratoma should have resection of the primary tumor. Careful evaluation should be made in
Clinical Presentation
Of the five cases, four were male and one was female, with ages ranging from 36 to 88.40,41 Most case reports do not give information on clinical presentation, though all had masses located in either the body or antrum of the stomach.39 – 41 Patients have presented with both localized and metastatic disease. Prognosis
One of the five patients had long-term survival and one was an autopsy case.40 The remaining three patients had short survival times.40,41 Recommendations
Yolk sac tumors of the stomach should be treated with germ cell tumor regimens such as bleomycin–etoposide –cisplatin or vinblastine, ifosfamide, and cisplatin. However, the adenocarcinomatous components may not be as responsive as the germ cell tumors. There has been at least one report of a yolk sac tumor that did not respond at all to germ cell regimens.41 It is not known which component or components were not responsive.
PARIETAL CELL CARCINOMA AND ONCOCYTIC GASTRIC CARCINOMA Background Parietal cell carcinomas were first reported in the 1980s. The original reports of parietal cell carcinoma were diagnosed due to characteristics on hematoxylin and eosin (H and E)
UNCOMMON CANCERS OF THE STOMACH
staining and on electron microscopy similar to parietal cells.12 More recently, investigators have reported oncocytic gastric carcinoma in 1.8% of the gastric cancers seen; it is unclear if these truly represent a different entity.42
Pathology In the few reported cases, parietal cell carcinomas are bulky tumors involving the gastric body and antrum.12 The tumor cells exhibit histologic and ultrastructural features of parietal cells: solid nests of tumor composed of polygonal cells with abundant cytoplasm and abundant mitochondria, tubulovesicles, and intracellular canaliculi by electron microscopy. Like gastric parietal cells, the tumor cells contain cytoplasmic granules that stain with phosphotungstic acid hematoxylin. Oncocytic gastric carcinoma, also very rare, has a similar morphologic appearance but is negative for parietal cell markers by immunohistochemical studies using antibodies directed against H(+) – K(+)- adenosine triphosphatase.42 Focal parietal cell differentiation, as determined by ultrastructural examination, has been described in a well-differentiated intestinal-type early gastric cancer.43
Clinical Presentation The early reported cases of parietal cell cancers appeared to have presentations similar with other gastric cancers. A series of 10 patients of oncocytic gastric carcinoma was reported. The patients appeared to have a median age of 69 (age range: 58–81).42 All patients had early disease with eight stage I patients and two stage IIIA patients.
Treatment/Prognosis The prognosis for parietal cell cancer has been reported as better for the standard gastric adenocarcinomas, but the number of cases are limited.44 Similarly, of the 10 cases of oncocytic gastric adenocarcinoma reported by Takubo et al., all had resectable disease.42 Of the 10, 8 were alive at the time of report and only 1 of the 2 patients who died had evidence of cancer at the time of death.
Recommendations The majority of patients with parietal cell and/or oncocytic gastric carcinomas have presented with resectable disease and have had good prognoses, suggesting surgery is the primary modality of therapy. No recommendations can be made about adjuvant therapy since very few patients have had recurrences, suggesting that adjuvant therapy may not be necessary. Similarly, in metastatic disease there are no data about effects of systemic therapies.
ENDOCRINE CELL PROLIFERATIONS IN THE STOMACH Background Endocrine or neuroendocrine (NE) tumors of the stomach account for approximately 2% of neuroendocrine tumors and only 0.5–2% of gastric tumors.45,46 Carcinoid tumors, or well-differentiated NE tumors, make up the majority of
357
gastric NE tumors. In a series of 205 gastric NE tumors, only 12 were poorly differentiated.47 Gastric carcinoids are rising in incidence.46,48,49 The cause for this rise in incidence is not clear, but the development of gastric carcinoid tumors from enterochromaffin-like cells (ECLs) is thought to be stimulated by gastrin.48 Theoretically, with the rise in use of acid blocking agents, serum gastrin levels are also on the rise, but no definitive proof exists to support this hypothesis. Endocrine tumors of the stomach are most commonly nonfunctioning lesions composed of ECL cells, and may be divided into three types: type I tumors associated with autoimmune chronic atrophic gastritis; type II lesions arising in the setting of multiple endocrine neoplasia type I syndrome and Zollinger-Ellison (ZE) syndrome; and type III tumors, sporadic carcinoids not associated with hypergastrinemia or autoimmune gastritis.50 Most gastric endocrine tumors are type I.51 Rarely, sporadic non-ECL-cell gastric carcinoids occur and may produce gastrin or ACTH.
Pathology Macroscopically, types I and II lesions are usually multiple and produce small tan to yellow mucosa and submucosal nodules and plaques (see Figure 2). Type I tumors are usually smaller (mean diameter 0.5 cm) with over 95% less than 1.5 cm.50 Type III tumors are solitary, more likely to be greater than 2 cm in diameter, much more likely to invade the muscularis propria, and may show areas of necrosis. The microscopic appearance of types I and II lesions is similar; in most cases, the tumors are composed of small trabeculae and solid nests.12 The tumor cells are small, uniform, with round to oval nuclei with finely granular chromatin and inconspicuous nucleoli. Mitoses are rare, and vascular invasion and necrosis are not seen. Sporadic carcinoids (type III tumors) have a more solid growth pattern, and the tumor cells are more atypical, with coarsely clumped chromatin and prominent nucleoli. Mitoses are more common and vascular invasion is also seen. Small cell carcinomas, with histologic features identical to other extrapulmonary carcinomas, also occur in the stomach. Distinguishing between endocrine cell neoplasia and reversible endocrine cell hyperplasia due to hypergastrinemia may be difficult. In general, infiltrative or solid nodules of endocrine cells confined to the mucosa but reaching a size of >0.5 mm, and still confined to the mucosa, are considered microcarcinoids or intramucosal carcinoids.50,52 Any growth of endocrine cells that invades the submucosa or into the vascular structures is considered a true neoplasm. Features associated with aggressive behavior include invasion of muscularis propria or beyond, angioinvasion, tumor size >1 cm, and high mitotic activity. Sporadic (type III) carcinoids are more likely to behave in an aggressive fashion.
Clinical Presentation Gastric carcinoids are more common in females, with a male:female ratio of 0.57.46 The ratio of African-Americans to Caucasians is 0.17. Mean age at diagnosis is almost 64 years. Most commonly, gastric carcinoids present as localized (45.2–52.9%) or regional disease (10.3–28.6%). Distant
358
GASTROINTESTINAL TUMORS
(a)
(b) Figure 2 Gastric carcinoid tumor. (a) This type I gastric carcinoid is arising in the setting of chronic atrophic gastritis. This lesion is classified as an invasive neoplasms due to the presence of microscopic nests of tumor cells penetrating the muscularis mucosae and involving the submucosa. Hematoxylin and eosin, original magnification 100×. (b) The tumor cells have round to oval nuclei with finely granular chromatin and are arranged in nests and trabeculae. Hematoxylin and eosin, original magnification 400×.
metastases are present in 20.6–23.8% while 2.4–16% are unstaged. The antrum is rarely the only site of involvement
(<1%) and the rest involve the body of the stomach.47 Gastric carcinoids are also associated with other malignancies. Synchronous tumors are seen in 2–5% of cases and metachronous second primaries in up to 9% of cases.46,47 Other GI cancers appear to make up the majority of second primaries.47 Type I carcinoids comprise 70–80% of gastric NE tumors and usually arise in the body or fundus of the stomach.47,51,53 Women are affected more commonly than men with at least a 2 : 1 ratio. Although patients as young as 18 years old have been reported, the average age at diagnosis is >60 years.47,48 At the time of endoscopic diagnosis, the lesions are often multifocal (57%) and are usually asymptomatic.47 Endoscopy is more often done for unrelated symptoms or symptoms of atrophic gastritis. Seventy-seven percent of tumors are <1 cm. The carcinoids are confined to the mucosa or submucosa in 90% of patients and rarely present with metastases.48,51,53 However, over time, recurrence of carcinoids is not uncommon. At a mean follow-up of 58 months, of 199 patients, 65% had persistent tumor.47 On serum analysis, hypergastrinemia is common and serum chromogranin A and B levels are mildly elevated.47,54 Urinary 5-hydroxyindoleacetic acid (5-HIAA) is not elevated. These tumors rarely present with carcinoid syndrome.53 Type II carcinoids comprise 5–8% of gastric NE tumors.47,51,53 Mean age at diagnosis is 50 years.47 Like the type I gastric carcinoids, these lesions tend to present as small, multifocal tumors within the body and fundus of the stomach and less commonly in the antrum.47,51,53 By definition, the type II carcinoids develop in patients with ZE syndrome, although they are more likely to occur in patients with multiple endocrine neoplasia type I (MEN-I) than in patients with sporadic ZE syndrome. Up to 73% of tumors present at a size <1.5 cm.47 Despite this small tumor size at presentation, up to 30% will have lymph node metastases upon diagnosis and metastases in ∼10%.47,48 Serum gastrin and chromogranin levels can be elevated.48,54 These tumors rarely present with carcinoid syndrome. Type III carcinoids comprise 23% of all gastric carcinoids and are more common in males than females (3–4 : 1).47,53 Type III carcinoids appear more aggressive. Mean age at diagnosis is 55 years.47 Only 30% of patients present at a size <1 cm.47 Tumors are rarely confined to the mucosa and over half involve the serosa. The majority present with lymph node metastases and up to 75% have metastatic disease.47 The type III carcinoids can demonstrate lymphovascular invasion and can present as poorly differentiated tumors as well.55 The poorly differentiated type III carcinoids nearly always metastasize. Serum chromogranin levels are elevated, while gastrin levels usually are not.53,54 Tachykinin elevations have also been seen. Atypical or foregut carcinoid syndrome consisting of flushing, pruritis, bronchospasm, and lacrimation can be present.53 Poorly differentiated NE tumors are the least studied subset of NE tumors. They appear to be more common in men, with a mean age at presentation of 63 years.47 The mean size at presentation is >4 cm and they tend to be deeply invasive. Metastases to lymph nodes are common. Presenting
UNCOMMON CANCERS OF THE STOMACH
symptoms are more typical of other gastric cancers, including abdominal pain, weight loss, and fatigue. Foregut carcinoids have presented with carcinoid crisis.56 This usually develops after an event such as a surgical procedure, biopsy, induction of anesthesia and even chemotherapy. Symptoms include an intense flush of extended duration, severe diarrhea, CNS abnormalities, and hypotension sometimes requiring ICU support.
Treatment Localized Disease Surgical resection is the primary therapy for all subtypes of gastric NE tumors. In types I and II gastric carcinoids, the favored procedures are minimally invasive, usually endoscopic mucosal resection (EMR). The maximum tumor size for EMR is 0.5–1.0 cm.47,57 – 59 While EUS may be useful in selecting patients for EMR, it is not consistently reliable.60 Surveillance endoscopy should be performed routinely after EMR, although the exact frequency has not been determined. Repeat EMR may be performed for local recurrence although more aggressive surgery may be considered. Antrectomy is the treatment of choice for larger gastric carcinoids.61,62 Antrectomy can produce normalization of the gastrin levels which in turn removes the stimulus for carcinoid growth and is capable of producing regression of disease.63 If antrectomy does not result in regression, full gastrectomy may be considered.62 Laparoscopic resection has been reported as feasible in a series of four patients.57 For type III carcinoids and poorly differentiated NE tumors localized to the stomach, partial or total gastrectomy is the treatment of choice.61,62
Metastatic Disease Liver-directed Therapy
The most common site of distant metastasis is the liver. Data on liver-directed therapy for liver metastases exists for neuroendocrine tumors from all primary sites. Surgery for NE metastases has resulted in excellent long-term survival with 82% alive at 5 years, but with a slowly progressive disease such as carcinoid, these results may occur regardless of therapy.64 Similarly, radiofrequency ablation (RFA) can control disease, but the true benefit is uncertain.65 Encouragingly, RFA results in better local control of NE tumors than other tumor types such as adenocarcinoma or sarcoma, but this may be reflective of the natural history of the disease.65 Surgical ligation of the hepatic artery with or without systemic chemotherapy has produced responses in 60–80% of NE tumors metastatic to the liver, respectively.66 Duration of response also appears better with the addition of systemic therapy (18 months) than with ligation alone (4 months), but the data is retrospective and not randomized. This procedure has largely been replaced with transarterial chemoembolization (TACE), in which chemotherapy is instilled into the hepatic artery and a thick emulsion or small beads are injected into the hepatic artery to occlude blood flow. TACE has also produced objective responses in up to 77% of tumors embolized.67 The major impact of TACE is control of carcinoid symptoms and hormone levels in 67–73%, respectively.
359
Common side effects of this procedure include pain, fever, nausea, and vomiting. Elevations of liver function tests are common, but hepatic failure is uncommon. Techniques and choice of agents used for TACE vary significantly by institution. Systemic Therapy
Somatostatin is a small peptide capable of inhibiting the secretion of an array of hormones. At least 80% of carcinoid tumors express somatostatin receptors.68 An eight amino acid analog of somatostatin, octreotide, has been developed for both diagnosis and treatment of carcinoid tumor. Radioactive octreotide scintigraphy has a sensitivity of 89% for detecting carcinoid tumors.69 Injectable forms of octreotide have been used for therapeutic purposes. The half-life of octreotide is only 2 hours when injected subcutaneously.70 It primarily controls the symptoms of the disease with a reduction of symptoms in nearly 90% of patients and 50% decrease in 5-HIAA levels in 72% of patients.71 This agent requires subcutaneous administration three times daily. Longer-acting forms of octreotide that are injected once monthly have also been developed with similar effects on symptoms.70 It is not clear that octreotide plays a role in control of the tumor itself. However, it can suppress gastrin secretion and therefore should be able to control types I and II gastric carcinoids. Interferon therapy has been tried in multiple trials for NE tumors. In the largest series of 130 patients, the response rate to interferon was 15% and biochemical response in 42% of patients.72 However, response to interferon can be very shortlived.73,74 With newer radiographic techniques and RECIST criteria for assessing response, Yao et al. did not observe any radiographic responses in 21 patients. Systemic chemotherapy has had limited effect in lowgrade neuroendocrine tumors. Response rates up to 21% have been reported for single agent therapies and up to 33% for combination therapies.56 In a randomized trial comparing 5-FU + doxorubicin to 5-FU + streptozocin, both arms had response rates of 16% and stable disease rates of 15% with similar progression-free survivals of 4.5 and 5.3 months, respectively.75 However, survival was better for the patients treated with 5-FU and streptozocin over those treated with 5-FU and doxorubicin (24.3 months vs 15.7 months, p = 0.0267). Dacarbazine was tested as a second-line therapy in the same trial with a modest response rate of 8.2%. More recently, inhibitors of novel targets including imatinib, bevacizumab, and sunitinib (SU-011248), have been evaluated as therapy for carcinoid tumors with limited benefit.74,76,77 These studies mostly evaluated progressionfree survival. Because carcinoid tumors have variable rates of progression, it is difficult to assess the progression-free survival endpoint in small studies. However, bevacizumab produced a 15% response rate by RECIST criteria.74 In addition, when patients progressed on interferon, bevacizumab was added. This “crossover” resulted in one additional response. Sunitinib produced a modest 5.1% response rate in 39 carcinoid patients with over 90% experiencing stable disease.77 Systemic chemotherapy for poorly differentiated neuroendocrine tumors is platinum-based, usually cisplatin and etoposide.
360
GASTROINTESTINAL TUMORS
Prognosis The 5-year survival rates for gastric carcinoids (not poorly differentiated NE tumors) has been reported as 52% overall.45 For localized disease, Modlin et al. reported a 5-year survival of 64–93%, but for regional disease it was only 23–39%, and 0–10% for those with metastases.46 The 265 cases did occur over a 40-year period during which radiology techniques, therapy, and palliative care have all changed. By disease type, Rindi et al. noted that type I disease had an excellent prognosis.47 At 144 months mean follow-up, the majority of patients were still alive (119 of 144) with most deaths not related to the carcinoid. Of the 12 patients with type 2 disease, only 1 had died of disease at a mean followup of 84 months. Of the 27 patients with type 3 disease, 7 had died of disease with a mean follow-up of 28 months. For poorly differentiated NE tumors, Rindi et al. noted a mean survival of 6.6 months. In a retrospective study of 143 patients with poorly differentiated NE tumors of any GI site, median survival was 13 months with a 3-year survival rate of 14%.78
Recommendations For types I and II carcinoids, EMR is the treatment of choice for small tumors while antrectomy is the treatment of choice for larger tumors. Type III carcinoids should also be treated with resection if possible, although there is no data for EMR in this setting. For patients with gastric carcinoid, especially those with typical or atypical syndrome, octreotide should be at least available if the patient is undergoing anesthesia. For endocrine symptoms, octreotide should be considered. For patients with poorly differentiated NE tumors of the stomach, systemic chemotherapy with cisplatin and etoposide should be considered.
GASTRIC STROMAL TUMORS Background GIST is a mesenchymal tumor formerly known by a variety of names, most commonly gastric leiomyosarcoma, leiomyoma, and leiomyoblastoma. The stomach is the most frequent site of involvement for GIST, accounting for ∼50% of GIST tumors.79 The prognosis and treatment of GIST has significantly changed over the last 5 years. This disease is covered in more detail in Chapter 37. The current chapter will largely deal with the features unique to gastric GIST.
Pathology Grossly, small gastric GISTs are submucosal, subserosal, or intramural firm well circumscribed nodules.80 Most are single, but occasionally multiple tumors are found. With larger tumors, the tumor may have a dumb-bell shape, and the overlying gastric mucosa may be ulcerated. The cut surface is usually granular, and areas of cystic change, necrosis, and hemorrhage may be seen.81 Microscopically, gastric GISTs show diverse histologic features but may be divided into spindle cell and epithelioid types. In spindle cell GISTs, the tumor cells are arranged
in short fascicles, often with prominent nuclear palisading: Malignant spindle cell GISTs are more densely cellular and exhibit more nuclear pleomorphism than their benign counterparts. While degenerative changes may be seen in low-risk spindle cell GISTs, coagulative necrosis is often present in the malignant tumors. Epithelioid gastric GISTs are composed of rounded or polygonal cells with a clear zone surrounding the nucleus; this morphologic pattern is much more common in the stomach than in other GI sites. Morphologic predictors of behavior in GISTs include tumor size, cellularity, nuclear atypia, mitotic activity, mucosal invasion, and coagulative tumor cell necrosis. The single most important prognostic factor, not surprisingly, is the presence of metastases at the time of resection. While some investigators advocate a “risk assessment” approach82 on the basis of mitotic activity and tumor size, others argue that this algorithm is overly simplistic and neglects other important features of malignancy and the influence of anatomic site.81 However, there is agreement that one of the most important prognostic factors is the mitotic rate, with the majority of tumors with a high mitotic count (>5 per 50 high-power fields) behaving in a malignant fashion. Immunohistochemistry for the proliferation marker MIB-1 may be useful in selected cases, with tumors exhibiting labeling indices of >10% more likely to recur locally.83 Mucosal invasion is a specific but relatively nonspecific marker for malignant behavior, as it is uncommon and may be obscured by ulceration of the overlying mucosa. The most important histopathologic marker for GISTs is the expression of CD117, the c-kit gene product, by the majority of these tumors. While some pathologists require positivity for CD117 to establish the diagnosis of GIST, approximately 4% of otherwise typical GISTs will fail to express CD117 by immunohistochemistry. These CD117negative GISTs are more likely to occur in the stomach and often have epithelioid morphology; KIT or PDGFRA mutations have been demonstrated in some.84 Most GISTs (∼70%) express CD34, a vascular and hematopoietic marker; a minority express myloid markers, despite their histologic resemblance to smooth muscle tumors.81
Clinical Presentation The median age at presentation of gastric GIST is ∼60 years, with fewer than 10% of patients with gastric GIST present before the age of 40.79 In a review of 1765 cases of gastric GIST, Meittinen found that 54% presented with evidence of GI bleed as manifested most commonly by anemia and less often by hematemesis or melena. Abdominal pain is the presenting symptom in 16.8% and 17.7% are incidental, found at radiologic or endoscopic exam performed for other reasons. Rarely, acute abdomen or tumor rupture is the presenting symptom. GIST primaries as large as 30 cm have been seen.
Carney Triad The triad of gastric leiomyosarcoma, extra-adrenal paraganglioma and pulmonary chondroma was first described by Carney et al. in 1977.85 Carney subsequently reported the
UNCOMMON CANCERS OF THE STOMACH
largest series of patients, 79, none of whom were related.86 However, two unrelated patients had a sibling with paraganglioma. The majority of patients (85%) were female and all were Caucasian. The mean age at presentation is 20.2 years with 82% of cases presenting before the age of 30 and no patients over the age of 50. The antrum was the most common site of origin for GIST, involved in at least 85% of cases. The GIST tumor was the cause of presenting symptoms in 73% of patients, with similar presenting symptoms to sporadic gastric GIST. Pulmonary chondromata was the presenting diagnosis in 15%, usually as asymptomatic chest x-ray findings. Paraganglioma, manifesting as hypertension or an asymptomatic mass accounted for 10% of the presentations. Most commonly, only one tumor type is present at initial diagnosis. The median time between detection of the first and second tumor was 8.4 years. Only 22% of patients had all three tumors and only one patient had pulmonary chondroma and extra-adrenal paraganglioma without GIST. A variety of other tumor types were found in these patients, most commonly benign adrenal tumors.
Treatment The primary treatment for GIST is surgery. Wedge resection and partial gastrectomy are the preferred surgical options. After resection of the primary site, 15–45% are disease-free at 5 years.87,88 From multivariate analyses, poor prognostic factors have included tumor size, high number of mitoses, incomplete resection, and male gender. In a series of 140 gastric GIST patients who underwent resection over a three decade time period, the majority, 98, were treated with wedge resection, 1 with enucleation, 20 with partial gastrectomy and the remainder with total gastrectomy. Of the 140, all but 5 underwent resection with curative intent. Of those 135, although the paper states that only 20 patients had recurrence or metastases, an additional 23 died of GIST. On multivariate analysis, male sex, mitotic index, tumor size, and the presence of necrosis were poor prognostic factors. Resection of metastatic disease has also produced diseasefree intervals that can be prolonged.87 – 89 When patients no longer have resectable disease, chemotherapy has been ineffective. However, taking advantage of the knowledge of disease biology, agents targeted against c-KIT and PDGF have shown activity in GIST. A detailed discussion of targeted therapy for GIST may be found in Chapter 37. Imatinib mesylate targets BCR/ABL, c-KIT and platelet-derived growth factor receptor (PDGFR). In a randomized trial of two doses, 54% of patients responded and 28% had stable disease.90 Response rates varied by the type of c-KIT mutation found.91 After progression on low dose imatinib (400 mg), increasing the dose to 800 mg can produce stable disease or responses in 29% and 7%, of patients, respectively.92 More recently, in 312 patients with imatinib-refractory GIST, a second agent, SU-011248 (sunitinib) was compared to placebo in a 2 : 1, randomized, crossover design.93 Time to tumor progression significantly favored the sunitinib arm (6.3 months vs 1.5 months, respectively, p < 0.00001). Responses were seen in only 8% of patients, but stable
361
disease occurred in 58%. Other agents with similar targets are currently being studied in GIST.
Prognosis In a study from Japan of 140 patients with gastric GIST, 135 of whom had localized disease, survival at 5 and 10 years was estimated to be 86 and 81%, respectively.89 For patients with operable primary disease and no metastases at presentation, 5- and 10-year survival rates were 93% and 88%, respectively. Prognosis for metastatic disease is changing with the discovery of new targeted agents. In a randomized trial of 746 patients with metastatic GIST treated with one of two doses of imatinib mesylate, 50–53% of patients were free from progression at 2 years.92 After a median follow-up of over 2 years, median survival had not yet been reached. Survival at 2 years was estimated at 78%. Since gastric GIST comprises a significant proportion of GIST patients, it can be presumed that gastric GIST patients have similar prognosis. For patients with Carney triad, with a median follow-up of 20 years, 81% of patients were alive, 30% without tumor.86 Of the 15 patients who died, 10 died of disease and the remaining 5 died with complications of metastases or another form of cancer.
GASTRIC LYMPHOMAS Background Lymphoma of the stomach accounts for 7–10% of all lymphomas and over 30% of extranodal cases.94,95 Gastric lymphomas are the second most common malignancy of the stomach after adenocarcinoma, accounting for up to 3% of gastric tumors. Some population-based studies suggest that the incidence of gastric (and other GI) lymphomas is on the rise, while others suggest incidence is stable.95,96 Most gastric lymphomas are of the non-Hodgkin’s type, with the majority classified as B cell lymphomas.97 Low-grade gastric lymphomas are almost exclusively B cell MALT (mucosa-associated lymphoid tissue), also called marginal zone lymphomas; many of the high-grade diffuse large cell lymphomas, the most common subtype, have developed through progression of these lower grade lesions. In an evaluation of 463 patients from a British database, 43% were low-grade B cell lymphomas, 54% were high-grade B cell lymphomas and 2.5% were T cell lymphomas.96 Hodgkin’s lymphoma of the stomach has been reported but appears to be extremely rare.98 Gastric lymphoma frequently arises in the setting of gastritis. This observation led to the subsequent observation that H. pylori infection is present in up to 92% of patients with gastric MALT lymphoma.99 Other forms of lymphoma are not associated with H. pylori infection.94 The process by which H. pylori plays a role in gastric lymphoma is being elucidated. Early genetic changes in the gastric mucosa predispose the cells to be sensitive to growth stimulation by H. pylori.100 Further genetic changes may lead to a more aggressive, high-grade phenotype.
Pathology On gross examination, the gastric mucosa involved by marginal zone lymphomas shows erosions, thickened gastric
362
GASTROINTESTINAL TUMORS
Figure 3 In low-grade gastric MALT-type lymphomas, the neoplastic lymphocytes resembled marginal zone cells diffusely infiltrate the mucosa. The cells infiltrate and disrupt gastric glands, forming lymphoepithelial lesions (40x).
folds, or a diffusely infiltrative pattern, usually involving the antrum101 and is multifocal. Gastric diffuse large B cell lymphoma, in contrast, usually forms a single large exophytic mass lesion. Microscopically, low-grade MALT-type lymphomas show a diffuse, vaguely nodular infiltrate of small lymphocytes resembling marginal zone B cells, with slightly irregular nuclei and a small amount of clear cytoplasm101 (see Figure 3). Plasmacytic differentiation may be prominent. The neoplastic cells infiltrate gastric epithelium in small clusters to form so-called lymphoepithelial lesions. Infiltration of lymphoid follicles (“colonization”) is also a characteristic feature of marginal zone lymphomas. The gastric mucosa also shows evidence of chronic gastritis, and H. pylori organisms may be found. The neoplastic cells are CD20+, BCL2+, CD5−, and CD10−. Approximately 30% will demonstrate a translocation involving the API2 gene on chromosome 11 and the MLT gene on chromosome 18, which may be specific for extranodal marginal zone lymphoma.101 In diffuse large cell lymphoma, the neoplastic lymphocytes are large and atypical, with oval, vesicular nuclei, and prominent nucleoli. The gastric glands are completely destroyed by the neoplastic infiltrate, which is often transmurally invasive. Like gastric marginal zone lymphomas, diffuse large cell lymphomas are CD20+ and CD10−; they may be BCL2 negative.101
Clinical Presentation The incidence of gastric lymphoma increases with age with a peak incidence in the eighth decade of life.96 Median age at presentation is 65 years.95 Patients with lowgrade lymphoma have a slightly younger median age than those with high-grade tumors.102 Males are more commonly affected than females with a ratio up to 1.5 : 1.95,96 Symptoms at presentation include abdominal pain, early satiety, weight
loss, anorexia, nausea, and vomiting.103 Evidence of GI bleeding such as hematemesis or melena may occur in up to 30% of patients. B symptoms may also be present in ∼35% of patients.95 On physical examination, most patients have no findings, but mass, tenderness, or adenopathy may be present.103 Bone marrow involvement with lymphoma has been demonstrated in 11% of gastric lymphoma patients.95 High-grade lymphomas are more likely to present with bulky disease and more advanced stage.102 Gastric lymphomas have been observed in all gastric sites, but the body and antrum are most common. Most patients present with early stage disease. Diagnosis is generally made with endoscopy and biopsy. However, substantial biopsies need to be obtained as endoscopic biopsy evaluation of grade and type is only 73% accurate when compared to surgical specimens.104 However, it has been difficult to develop a consensus as to the amount of tissue needed.105 On CT scan, diffuse thickening of the gastric wall is the most common finding for both high and low-grade lymphomas.106 Low-grade lymphomas present without any findings on CT scan nearly half the time while high-grade lymphomas usually have findings. EUS is also commonly used for staging gastric lymphoma. Compared to surgical staging, EUS is 74 and 76% accurate, respectively, for lymph node involvement and depth of invasion.104 In a limited comparison study of 14 patients, EUS was more accurate at staging gastric MALT lymphoma.107 FDG-PET scanning is increasingly part of lymphoma staging and may be more accurate for high-grade lymphoma, but low grade, MALT lymphomas are not always FDG avid.108 Two staging systems exist for the staging of gastric and other GI lymphomas. Musshoff, modified the Ann Arbor classification to stage GI lymphomas.109 An international workshop developed a second system.110 The two staging systems use very different terminologies in defining their stages, but are very similar (Table 1). Musshoff’s criteria are designed specifically for the stomach while the international classification is written for any location in the GI tract.109,110
Treatment Surgery was historically the treatment of choice for gastric lymphoma, but the role of surgery is diminishing.108,111 – 114 Table 1 Staging systems for gastric lymphoma.109,110
Description Confined to the stomach (GI tract) Single primary or multiple noncontiguous sites Contiguous lymph node involvement Noncontiguous subdiaphragmatic lymph nodes Involvement of stomach and lymph nodes on both sides of the diaphragm Penetration of serosal layer with involvement of adjacent organs Extranodal involvement
Mosshoff109
International110
IE
I
IIE1
II1
IIE2
II2
III
IV
N/A
III
IV
IV
UNCOMMON CANCERS OF THE STOMACH
Historically, survival in surgical series has ranged from 37 to 87%.115 However, it is difficult to compare studies since staging systems used and postoperative therapies varied significantly. Low-grade disease (MALT lymphomas): Low grade, or MALT, lymphomas of the stomach appear to be dependent on H. pylori for growth stimulation.116,117 This observation led to successful attempts to treat the lymphoma by eradication of the H. pylori.118 Retrospective analyses also showed significant reduction of MALT lymphoma after eradication of H. pylori.119 In a series of 33 patients, Bayerdorffer noted complete regression in 70% of patients and partial regression in 12%.119 Of the patients who did not respond, the majority were found to have high-grade lymphoma at subsequent surgery and the remainder, a T cell lymphoma. In a report of 50 patients (these may represent an update from the Bayerdorffer article) with MALT lymphoma in which H. pylori was eradicated, 40 had complete remission and 6 did not respond at all.120 Of the nonresponders, four were found to have high-grade lymphomas at subsequent surgery. The investigators noted that 5 of the 40 patients with complete responses relapsed at median follow-up time of 24 months. Additionally, a polymerase chain–reaction assay revealed that 22 of 31 assessable patients with complete remission still had a monoclonal B cell population. In a review of multiple reports of H. pylori eradication as therapy for lowgrade lymphoma, the complete remission rate was 80%.121 Recurrences of lymphoma have been reported as having progressions to a more aggressive phenotype, even without recurrence of H. pylori.116 From the data available, lymphoma recurs at a rate of ∼5% per year.121 While eradication of H. pylori appears to be a reasonable treatment for most stage II patients, MALT lymphoma patients with t(11 : 18) appear to be more resistant to this therapy.121 High-grade disease: For early stage disease (IE or IIE disease), survival after surgery alone has been as high as 95%.122 Surgery in that series consisted of a D2 lymph node dissection. In a retrospective study of 83 patients with early stage high-grade gastric lymphomas, long-term survival patients treated with chemotherapy alone was similar to those treated with surgery and chemotherapy.111 This suggests little role for surgery in early stage high-grade disease. This is supported by another nonrandomized, but prospective trial in 185 patients with IE or IIE gastric lymphoma.123 There was no difference in survival between patients who received surgery followed by radiation and/or chemotherapy versus those who received radiation and chemotherapy without surgery.123 Recently, Aviles et al. reported the results of a randomized trial of different therapeutic strategies for stage IE and II high-grade gastric lymphomas.113 Patients were randomized to chemotherapy (CHOP: cyclophosphamide, doxorubicin, vincristine, and prednisone), surgery + CHOP, surgery + 4-Gy radiation, or surgery alone. Complete response rates were 91–93% in all arms. At 10 years, event-free survival was significantly better for the CHOP and surgery + CHOP arms (92 and 82%, respectively) compared with the surgery and surgery + radiation arms (28 and 23%, respectively,
363
p < 0.001). The numerical differences between CHOP and surgery + CHOP were not statistically different. For advanced disease (stages III or IV), chemotherapy with or without radiation has been the treatment of choice, although patients who present with obstruction or significant bleeding may require emergent surgery. In the past, concern was raised about gastric perforation in patients treated with chemotherapy alone, but in the randomized trial by Aviles et al. none of the 140 patients randomized to the CHOP arm had a perforation.113 Rituximab is a monoclonal antibody against CD20, a B cell marker. It is increasingly part of treatment for other B cell non-Hodgkin’s lymphomas. Little is known about the efficacy of rituximab in gastric lymphoma. In a retrospective study of only nine patients with MALT lymphoma with various previous therapies, five responded and four had stable disease.124 This suggests that rituximab should be evaluated further in gastric lymphoma. After therapy of localized gastric lymphoma (either highgrade or low-grade), follow-up for proof of complete remission and evidence of recurrence involves routine endoscopy with biopsy.125 EUS has been studied in comparison to endoscopy and found to be inferior for follow-up.126 Therefore, endoscopy and biopsy remain the standard.125
Prognosis Survival at 2 years was 86–89% for low-grade tumors.104 For high-grade lymphomas, survival was 83–88% for patients with negative margins or microscopic residual, but only 53% with macroscopic residual disease (p < 0.0001). Survival rates for early stage low-grade lymphomas can be over 80%.108 With chemotherapy-based therapy, 10-year actuarial survivals for high-grade lymphomas have recently been higher than 90%.113 In gastric lymphoma, advanced stage, presence of B symptoms, advanced age, and elevated LDH are predictors of poor prognosis on multivariate analysis.95
Recommendations Stages I or II low-grade MALT lymphomas without t(11 : 18) translocations should be considered for treatment with H. pylori eradication. Patients with more advanced disease or t(11 : 18) translocations should be considered for systemic chemotherapy. Patients with high-grade lymphomas should be considered for chemotherapy with or without radiation as would be standard for similar stages of other high-grade lymphomas. Surveillance endoscopy with biopsy should be considered for patients with remission of disease.
REFERENCES 1. Dickson JLB, Cunningham D. Systemic treatment of gastric cancer. Eur J Gastroenterol Hepatol 2004; 16: 255 – 63. 2. Jemal A, et al. Cancer statistics 2005. CA Cancer J Clin 2005; 55: 10 – 30. 3. Dicken DJ, et al. Gastric carcinoma: review and considerations for future directions. Ann Surg 2005; 241: 27 – 39. 4. MacDonald JS, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 2001; 345: 725 – 30.
364
GASTROINTESTINAL TUMORS
5. Cunningham D, et al. Perioperative chemotherapy in operable gastric and lower oesophageal cancer: final results of a randomised, controlled trial (the MAGIC trial, ISRCTN 93793971). Proc Am Soc Clin Oncol 2005; 24: abstract 4001. 6. Watanabe H, Enjoji M, Imal T. Gastric carcinoma with lymphoid stroma: its morphologic characteristics and prognostic correlations. Cancer 1976; 38: 232 – 43. 7. Weiss LM, Gaffey M, Shibata D. Lymphoepithelioma-like carcinoma and its relationship to Epstein-Barr virus. Am J Clin Pathol 1991; 96: 156 – 8. 8. Matsunou H, et al. Characteristics of Epstein-Barr virus-associated gastric carcinoma with lymphoid stroma in Japan. Cancer 1996; 77: 1998 – 2004. 9. Adachi Y, et al. Epstein-Barr virus-associated gastric carcinoma. J Clin Gastroenterol 1996; 23: 207 – 10. 10. Tokunaga M, et al. Association of Epstein-Barr virus in gastric carcinoma. Am J Pathol 1993; 143: 1250 – 4. 11. Cho HJ, Kim JY, Lee SS. Gastric carcinoma with lymphoid stroma: Incidence of EBV and Helicobacter pylori infection. Appl Immunohistochem Mol Morphol 2003; 11: 149 – 52. 12. Lewin KJ, Appelman HD. Tumors of the esophagus and stomach. Atlas of Tumor Pathology, Third Series, Fasc 18. Washington, District of Columbia: Armed Forces Institute of Pathology, 1997. 13. Herrera-Geopfert R, et al. Epstein-Barr virus-associated gastric carcinoma in Mexico: analysis of 135 consecutive gastrectomies in two hospitals. Mod Pathol 1999; 12: 873 – 8. 14. Wu M-S, et al. Epstein-Barr virus-associated gastric carcinomas: relation to H. pylori infection and genetic alterations. Gastroenterology 2000; 118: 1031 – 8. 15. Boureille J, et al. Existence d’alpha fetoproteine au cours d’un cancer secondaire du foie d’origine gastrique. Presse Med 1970; 78: 1277 – 8. 16. Ishikura H, et al. An AFP-producing gastric carcinomar with features of hepatic differentiation: a case report. Cancer 1985; 56: 840 – 8. 17. Akiyama S, et al. Histogenesis of hepatoid adenocarcinoma of the stomach: molecular evidence of identical origin with coexistent tubular adenocarcinoma. Int J Cancer 2003; 106: 510 – 5. 18. Nagai E, et al. Hepatoid adenocarcinoma of the stomach: a clinicopathologic and immunohistochemical analysis. Cancer 1993; 72: 1827 – 35. 19. Terraciano LM, et al. Hepatoid adenocarcinoma with liver metastases mimicking hepatocellular carcinoma: an immunohistochemical and molecular study of eight cases. Am J Surg Pathol 2003; 27: 1302 – 12. 20. Roberts CC, Colby TV, Batts KP. Carcinoma of the stomach with hepatocyte differentiation (hepatoid adenocarcinoma). Mayo Clin Proc 1997; 72: 1154 – 60. 21. Ishikura H, et al. Gastrointestinal hepatoid adenocarcinoma: venous permeation and mimicry of hepatocellular carcinoma, a report of four cases. Histopathology 1997; 31: 47 – 54. 22. Inagawa S, et al. Hepatoid adenocarcinoma of the stomach. Gastric Cancer 2001; 4: 43 – 52. 23. de Lorimier A, et al. Hepatoid adenocarcinoma of the stomach. Cancer 1993; 71: 293 – 6. 24. Chang YC, et al. Xenotransplantation of alpha-fetoprotein-producing gastric cancers into nude mice:characteristics and responses to chemotherapy. Cancer 1992; 69: 872 – 8. 25. Rolleston HD, Trevor RS. A case of columnar-celled carcinoma of the stomach showing squamous-celled metaplasia. J Pathol Bacteriol 1905; 10: 418 – 22. 26. Lee WA, et al. p53, p16, and RB expression in adenosquamous and squamous cell carcinomas of the stomach. Pathol Res Pract 1999; 195: 747 – 52. 27. Volpe CM, et al. Squamous cell carcinoma of the stomach. Am Surg 1995; 12: 1076 – 8. 28. Schwab G, et al. Primary squamous cell carcinoma of the stomach in a seventeen-year-old boy. Surg Today 1992; 22: 561 – 4. 29. Altshuler JH, Shaka JA. Squamous cell carcinoma of the stomach. Review of the literature and report of a case. Cancer 1966; 19: 831 – 8. 30. Mori M, Iwashita A, Enjoji M. Adenosquamous carcinoma of the stomach. A clinicopathologic analysis of 28 cases. Cancer 1986; 57: 333 – 9.
31. Davidsohn C. Chorion-epitheliom und magenkrebs, eine seltene verschmelzung zweier bosar-tiger geschwulste. Charite-Ann. 1905; 29: 426 – 37. 32. Krulewski T, Cohen LB. Choriocarcinoma of the stomach: pathogenesis and clinical characteristics. Am J Gastroenterol 1988; 83: 1172 – 5. 33. Liu AY, et al. Gastric choriocarcinoma shows characteristics of adenocarcinoma and gestational choriocarcinoma: a comparative genomic hybridization and fluorescence in situ hybridization study. Diagn Mol Pathol 2001; 10: 161 – 5. 34. Noguchi T, et al. A patient with primary gastric choriocarcinoma who received a correct preoperative diagnosis and achieved prolonged survival. Gastric Cancer 2002; 5: 112 – 7. 35. Jindrak K, Bochetto JF, Alpert LI. Primary gastric choriocarcinoma: case report with review of the world literature. Hum Pathol 1976; 7: 595 – 604. 36. Liu Z, Mira JL, Cruz-cadillo JC. Primary gastric choriocarcinoma∼: a case report and review of the literature. Arch Pathol Lab Med 2001; 125: 1601 – 4. 37. Corapcioglu F, et al. Immature gastric teratoma of childhood: a case report and review of the literature. J Pediatr Gastroenterol Nutr 2004; 39: 292 – 4. 38. Marina NM, et al. Complete surgical excision is effective treatment for treatment of immature teratomas with or without malignant features: a Pediatric Oncology Group/Children’s cancer Group Intergroup Study. J Clin Oncol 1999; 17: 2137 – 43. 39. Zamecnik M, Patrikova J, Comolcak P. Yolk sac carcinoma of the stomach with gastrin positivity. Hum Pathol 1993; 24: 927 – 8. 40. Suzuki T, et al. Yolk sac tumor of the stomach with adenocarcinomatous component: a case report with immunohistochemical analysis. Pathol Int 1999; 49: 557 – 62. 41. Wang L, et al. Gastric adenocarcinoma with a yolk sac component: a case report and review of the literature. J Clin Gastroenterol 2000; 31: 85 – 8. 42. Takubo K, et al. Oncocytic adenocarcinoma of the stomach: parietal cell carcinoma. Am J Surg Pathol 2002; 26: 458 – 65. 43. Caruso RA, et al. Focal parietal cell differentiation in a welldifferentiated (intestinal-type) early gastric cancer. Ultrastruct Pathol 2000; 24: 417 – 22. 44. Gaffney E. Favorable prognosis in gastric carcinoma with parietal cell differentiation. Histopathology 1987; 11: 217 – 8. 45. Godwin JD III. Carcinoid tumors: an analysis of 2837 cases. Cancer 1975; 36: 560 – 9. 46. Modlin IM, et al. A 40-year analysis of 265 gastric carcinoids. Am J Gastroenterol 1997; 92: 633 – 8. 47. Rindi G, et al. Gastric carcinoids and neuroendocrine carcinomas: pathogenesis, pathology, and behavior. World J Surg 1996; 20: 168 – 72. 48. Mulkeen A, Cha C. Gastric carcinoid. Curr Opin Oncol 2004; 17: 1 – 6. 49. Maggard MA, O’Connell JB, Ko CY. Updated population-based review of carcinoid tumors. Ann Surg 2004; 240: 117 – 22. 50. Capella C et al., World Health Organization Classification of Tumours. Endocrine tumours of the stomach. Tumours of the Digestive System: Pathology and Genetics. Lyon, France: IARC Press, 2000. 51. Delle Fave G, et al. Gastric neuroendocrine tumors. Neuroendocrinology 2004; 80(Suppl 1): 16 – 9. 52. Solcia E, et al. Hyperplastic, dysplastic and neoplastic enterochromaffin-like-cell proliferations of the gastric mucosa. Classification and pathogenesis. Am J Surg Pathol 1995; 19(Suppl 1): S1 – 7. 53. Modlin IM, Lye KD, Kidd M. Carcinoid tumors of the stomach. Surg Oncol 2003; 12: 153 – 72. 54. Granberg D, et al. Clinical symptoms, hormone profiles, treatment and prognosis in patients with gastric carcinoids. Gut 1998; 43: 223 – 8. 55. Rindi G, et al. ECL cell tumor and poorly differentiated endocrine carcinoma of the stomach: prognostic evaluation by pathological analysis. Gastroenterologia 1999; 116: 532 – 42. 56. Moertel CG. An odyssey in the land of small tumors. J Clin Oncol 1987; 10: 1503 – 22. 57. Otani Y, et al. Minimally invasive surgery for gastric carcinoid tumor. Biomed Pharmacother 2002; 56(Suppl 1): 217s – 21s.
UNCOMMON CANCERS OF THE STOMACH 58. Ichikawa J, et al. Endoscopic mucosal resection in the management of gastric carcinoid tumors. Endoscopy 2003; 35: 203 – 6. 59. Gonzalez-Ramirez A, et al. Multiple gastric carcinoid tumors∼: endoscopic management. J Clin Gastroenterol 1996; 23: 75 – 7. 60. Lechter J, Chemtob J. EUS may have limited impact on the endoscopic management of gastric carcinoids. Int J Gastrointest Cancer 2002; 31: 181 – 3. 61. Schindl M, Kasserer K, Niederle B. Treatment of gastric neuroendocrine tumors. The necessity of a type-adapted treatment. Arch Surg 2001; 136: 49 – 54. 62. Akerstrom G. Management of carcinoid tumors of the stomach, duodenum, and pancreas. World J Surg 1996; 20: 173 – 82. 63. Hirschowitz BI, et al. Rapid regression of enterochromaffinlike cell gastric carcinoids in pernicious anemia after antrectomy. Gastroenterology 1992; 102: 1409 – 18. 64. Norton JA, et al. Aggressive surgery for metastatic liver neuroendocrine tumors. Surgery 2003; 134: 1065 – 73. 65. Siperstein A, et al. Local recurrence after laparoscopic radiofrequency thermal ablation of hepatic tumors. Ann Surg Oncol 2000; 7: 106 – 13. 66. Moertel CG, et al. The management of patients with advanced carcinoid tumors and islet cell carcinomas. Ann Int Med 1994; 120: 302 – 9. 67. Diamandidou E, et al. Two-phase study of hepatic artery vascular occlusion with microencapsulated cisplatin in patients with liver metastases from neuroendocrine tumors. AJR Am J Roentgenol 1998; 170: 339 – 44. 68. Reubi JC, et al. Detection of somatostatin receptors in surgical and percutaneous needle biopsy samples of carcinoids and islet cell carcinomas. Cancer Res 1990; 50: 5969 – 77. 69. Krenning EP, et al. Somatostatin-receptor scintigraphy in gastroenteropancreatic tumors: an overview of European results. Ann N Y Acad Sci 1994; 733: 416 – 24. 70. Oberg K. Carcinoid tumors: current concepts in diagnosis and treatment. The Oncologist 1998; 3: 339 – 45. 71. Kvols LK, et al. Treatment of the malignant carcinoid syndrome. Evaluation of a long-acting somatostatin analogue. N Engl J Med 1986; 315: 663 – 6. 72. Oberg K, Eriksson B. The role of interferons in the management of cacinoid tumors. Br J Haematol 1991; 79(Suppl 1): 74 – 7. 73. Moertel CG, Rubin J, Kvols LK. Therapy of metastatic carcinoid syndrome with recombinant leucocyte A interferon. J Clin Oncol 1989; 7: 865 – 8. 74. Yao JC, et al. Improved progression free survival (PFS), and rapid, sustained decrease in tumor perfusion among patients with advanced carcinoid treated with bevacizumab. Proc Am Soc Clin Oncol 2005; abstract 4007. 75. Sun W, et al. Phase II/III study of doxorubicin with fluorouracil compared with streptozocin with fluorouracil or dacarbazine in the treatment of advanced carcinoid tumors: Eastern Cooperative Oncology Group Study E1281. J Clin Oncol 2005; 23: 4897 – 904. 76. Karr K, et al. A phase II trial of imatinib in patients with advanced carcinoid tumor. Proc Am Soc Clin Oncol 2004; abstract 4124. 77. Kulke M, et al. A phase 2 study to evaluate the efficacy and safety of SU11248 in patients (pts) with unresectable neuroendocrine tumors (NETs). Proc Am Soc Clin Oncol 2005; abstract 4008. 78. Brenner B, et al. High-grade neuroendocrine carcinomas of the gastrointestinal tract: The Memorial Sloan – Kettering Cancer Center experience of 143 cases. Proc Am Soc Clin Oncol 2004; 22: abstract 4123. 79. Miettinen M, Sobin LH, Lasota J. Gastrointestinal stromal tumors of the stomach: a clinicopathologic, immunohistochemical, and molecular genetic study of 1765 cases with long-term follow-up. Am J Surg Pathol 2005; 29: 52 – 68. 80. Miettinen M, Blay JY, Sobin LH, World Health Organization Classification of Tumours. Mesenchymal tumours of the stomach. Tumours of the Digestive System: Pathology and Genetics. Lyon, France: IARC Press, 2000. 81. Goldblum JR. Mesenchymal tumors of the GI tract. In Odze RD, Goldblum JR, Crawford JM (eds) Surgical Pathology of the GI Tract, Liver, Biliary Tract, and Pancreas. Philadelphia, Pennsylvania: Sanders, 2004.
365
82. Fletcher CDM. et al. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Hum Pathol 2002; 333: 459 – 65. 83. Mochizuki Y. et al. Treatment and risk factors for recurrence after curative resection of gastrointestinal stromal tumors of the stomach. World J Surg 2004; 28: 870 – 5. 84. Meidiros F. et al. KIT-negative gastrointestinal stromal tumors: proof of concept and therapeutic implications. Am J Surg Pathol 2004; 28: 889 – 94. 85. Carney JA, et al. The triad of gastric leiomyosarcoma, functioning extra-adrenal paraganglioma and pulmonary chondroma. N Engl J Med 1977; 296: 1517 – 8. 86. Carney JA. Gastric stromal sarcoma, pulmonary chondroma, and extra-adrenal paraganglioma (Carney Triad): natural history, adrenocortical component, and possible familial occurrence. Mayo Clin Proc 1999; 74: 543 – 52. 87. DeMatteo RP, et al. two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 2000; 231: 51 – 8. 88. Aparicio T, et al. Prognostic factors after surgery of primary resectable gastrointestinal stromal tumors. Eur J Surg Oncol 2004; 30: 1098 – 103. 89. Fujimoto Y, et al. Clinicopathologic study of primary malignant gastrointestinal stromal tumor of the stomach with special reference to prognostic factors: analysis of results in 140 surgically resected patients. Gastric Cancer 2003; 6: 39 – 48. 90. Demetri GD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med 2002; 347: 472 – 80. 91. Blackstein ME, et al. Clinical benefit of Imatinib in patients (pts) with metastatic Gastrointestinal Stromal Tumors (GIST) negative for the expression of CD117 in the S0033 trial. Proc Am Soc Clin Oncol 2005; 23: abstract 9010. 92. Rankin C, et al. Dose effect of imatinib (IM) in patients (pts) with metastatic GIST – Phase III Sarcoma Group Study S0033. Proc Am Soc Clin Oncol 2004; 22: abstract 9005. 93. Demetri GD, et al. Phase 3, multicenter, randomized, double-blind, placebo-controlled trial of SU11248 in patients (pts) following failure of imatinib for metastatic GIST. Proc Am Soc Clin Oncol 2005; 23: abstract 4000 (updated from ASCO website). 94. Parsonnet J, et al. Helicobacter pylori infection and gastric lymphoma. N Engl J Med 1994; 330: 1267 – 71. 95. D’Amore F, et al. Non-hodgkin’s lymphoma of the gastrointestinal tract: a population-based analysis of incidence, geographic distribution, clinicopathologic presentation features, and prognosis. J Clin Oncol 1994; 12: 1673 – 84. 96. Gurney KA, Cartwright RA, Gilman EA. Descriptive epidemiology of gastrointestinal non-Hodgkin’s lymphoma. in a population-based registry. Br J Cancer 1999; 79: 1929 – 34. 97. Wotherspoon A et al., World Health Organization Classification of Tumours. Lymphoma of the stomach. Tumours of the Digestive System: Pathology and Genetics. Lyon, France: IARC Press, 2000. 98. Soderstrom KO, Joensuu H. Primary Hodgkin’s disease of the stomach. Am J Clin Pathol 1988; 89: 806 – 9. 99. Wotherspoon AC, et al. Helicobacter pylori-associated gastritis and primary B-cell gastric lymphoma. Lancet 1991; 338: 1175 – 6. 100. Isaacson PG. gastric MALT lymphoma: from concept to cure. Ann Oncol 1999; 10: 637 – 45. 101. Ferry J. Lymphoid tumors of the GI tract. In Odze RD, Goldblum JR, Crawford JM (eds) Surgical Pathology of the GI Tract, Liver, Biliary Tract, and Pancreas. Philadelphia, Pennsylvania: Sanders, 2004. 102. Taal BG, et al. Primary non-Hodgkin lymphoma of the stomach: endoscopic pattern and prognosis in low versus high grade malignancy in relation to the MALT concept. Gut 1996; 39: 556 – 61. 103. Al-Akwaa AM, Siddiqui N, Al-Mofleh IA. Primary gastric lymphoma. World J Gastroenterol 2004; 10: 5 – 11. 104. Fischbach W, et al. Primary gastric B-Cell lymphoma: results of a prospective multicenter study. Gastroenterology 2000; 119: 1191 – 202. 105. De Jong D, et al. Controversies and consensus in the diagnosis, workup and treatment of gastric lymphoma: an international survey. Ann Oncol 1999; 10: 275 – 80.
366
GASTROINTESTINAL TUMORS
106. Choi D, et al. Gastric mucosa-associated lymphoid tissue lymphoma: helical CT findings and pathologic correlation. AJR Am J Roentgenol 2002; 178: 1117 – 22. 107. Vorbeck F, et al. Comparison of spiral-computed tomography with water-filling of the stomach and endosonography for gastric lymphoma of mucosa-associated lymphoid-tissue type. Digestion 2002; 65: 196 – 9. 108. Yoon SS, et al. The diminishing role of surgery in the treatment of gastric lymphoma. Ann Surg 2004; 240: 28 – 37. 109. Musshoff K. Clinical staging classification of Non-Hodgkin’s lymphomas. Strahlentherapie 1977; 153: 218 – 21. 110. Rohatiner A, et al. Report on a workshop convened to discuss the pathological and staging classifications of gastrointestinal tract lymphoma. Ann Oncol 1994; 5: 397 – 400. 111. Ferreri AJM, et al. Therapeutic management of stage I-II high-grade primary gastric lymphomas. Oncology 1999; 56: 274 – 82. 112. Aviles A, et al. Is surgery necessary in the treatment of primary gastric non-Hodgkin lymphoma? Leuk Lymphoma 1991; 5: 365 – 9. 113. Aviles A, et al. The role of surgery in primary gastric lymphoma: results of a controlled clinical trial. Ann Surg 2004; 240: 44 – 50. 114. Binn M, et al. Surgical resection plus chemotherapy versus chemotherapy alone: comparison of two strategies to treat diffuse large B-cell gastric lymphoma. Ann Oncol 2003; 14: 1751 – 7. 115. Brands F, Monig SP, Raab M. treatment and prognosis of gastric lymphoma. Eur J Surg 1997; 163: 803 – 13. 116. Isaacson PG. Gastric MALT lymphoma: from concept to cure. Ann Oncol 1999; 10: 637 – 45. 117. Hussell T, et al. The response of cells from low-grade b-cell gastric lymphomas of mucosa-associated lymphoid tissue to Helicobacter pylori . Lancet 1993; 342: 571 – 4.
118. Wotherspoon AC, et al. Regression of primary low-grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobacter pylori . Lancet 1993; 342: 575 – 7. 119. Bayerdorffer E, et al. Regression of primary gastric lymphoma of mucosa-associated lymphoid tissue type after cure of Helicobacter pylori infection. Lancet 1995; 345: 1591 – 4. 120. Neubauer A, et al. Cure of Helicobacter pylori infection and duration of remission of low-grade gastric mucosa-associated lymphoid tissue lymphoma. J Natl Cancer Inst 1997; 89: 1350 – 5. 121. Stolte M, et al. Helicobacter and gastric MALT lymphoma. Gut 2002; 50(Suppl III): III19 – 24. 122. Kodera Y, et al. The role of radical gastrectomy with systematic lymphadenectomy for the diagnosis and treatment of primary gastric lymphoma. Ann Surg 1998; 227: 45 – 50. 123. Koch P, et al. Primary gastrointestinal non-Hodgkin’s lymphoma: combined surgical and conservative or conservative management only in localized gastric lymphoma: Results of the prospective German Multicenter Study GIT NHL 01/92. J Clin Oncol 2001; 19: 3874 – 83. 124. Raderer M, et al. Rituximab for treatment of advanced extranodal marginal zone B cell lymphoma of the mucosa-associated lymphoid tissue lymphoma. Oncology 2003; 65: 306 – 10. 125. Boot H, de Jong D. Diagnosis, treatment decisions and follow up in primary gastric lymphoma. Gut 2002; 51: 621 – 2. 126. Puspock A, et al. Endoscopic ultrasound in the follow up and response assessment of patients with primary gastric lymphoma. Gut 2002; 51: 691 – 4.
Section 6 : Gastrointestinal Tumors
32
Unusual Pancreatic Tumors Ann Wexler, Roger J. Waltzman and John S. Macdonald
INTRODUCTION Cancer of the pancreas was expected to be responsible for 31 800 cancer deaths in the United States in 2005.1 The incidence is 10 per 100 000 in the general population. This disease is the fourth leading cause of cancer deaths and less than 5% of all cases have a 5-year survival.1 – 3 In both men and women, rates of pancreatic cancer increase with age in a log-linear manner. The disease is somewhat more common in males than in females. The peak incidence is in the seventh decade.4 Numerous epidemiologic studies have explored the etiologic factors in pancreatic cancer. It has been felt that there is an increased incidence in patients with diabetes3,5 – 9 and chronic pancreatitis;10 – 12 however, there are no studies that would suggest a definite causal relationship. One rare entity that does seem to carry an increased risk for pancreatic cancer is hereditary pancreatitis.13 Other factors that have been investigated include coffee,12,14 radiation exposure,15 – 19 asbestos exposure,20 – 22 diet,23 – 25 and tobacco.26 – 32 Except for tobacco, the role of these factors in the etiology of pancreatic cancer must be considered, at best, suggestive.33 Recently, there has been interest in the familial association of pancreatic cancer.34 Hereditary pancreatitis,35 Lynch syndrome II,36 von Hippel–Lindau syndrome,37 and familial melanoma syndrome,38 in which the p16 gene is mutated, have all been associated with increased risk of pancreatic adenocarcinoma. A minimum of 3% of pancreatic cancers is felt to occur on a hereditary basis.39 Recently, there has been an evolving understanding of the molecular genetic abnormalities associated with adenocarcinoma of the pancreas. Of particular interest is the fact that approximately 90% of tumors have mutations in the codon 12 of the ras oncogene.40 – 43 There have also been at least three putative tumor-suppressive gene loci defined, in which homozygous deletion commonly occurs in pancreatic cancer. These loci have been named the deleted in pancreatic cancer (DPC ) loci.44 – 47 Germline mutations in BRCA2 have also been suggested to contribute to the development of pancreatic cancer, with mutations seen in approximately 6% of cases of familial pancreatic cancer.48 Although no successful clinical strategy
in screening or treatment has yet been developed from this information, there is no doubt that the increasing knowledge of the molecular genetics of pancreatic cancer will be very important for future progress in the management of this devastating disease. The degree of lethality of adenocarcinoma of the pancreas is reflected by the fact that while it is the 11th most common cancer, it is the fourth most common cause of cancer-related deaths in the United States. The overall survival is 23% at 1 year and 4% at 5 years.49 The explanations for the grim prognosis are many, but may be distilled down to three major reasons: (i) late diagnosis due to the nonspecific, vague complaints that may not arouse suspicion of pancreatic neoplasia at a time when early diagnosis and surgical cure are possible, (ii) the fact that surgery is capable of curing only 15% of resectable patients, and (iii) ineffective therapy for advanced disease. A detailed pathological classification of pancreatic malignancies would serve no clinical purpose if all of the separate entities possessed a similar prognosis. However, there are some types of pancreatic neoplasms, which have a better prognosis than ductal adenocarcinoma. The types of pancreatic malignancies are listed in Table 1. Basically, this table divides pancreatic tumors into those of the exocrine and those of the endocrine pancreas. The tumors of the endocrine pancreas are those of islet cell origin such as insulinoma, glucagonoma etc. The exocrine pancreatic tumors are represented by tumors of presumed ductal cell origin,50,51 of which the majority are the common ductal adenocarcinoma. Also included are less-common variants such as squamous cell carcinoma, giant cell carcinoma, and cystadenocarcinoma. Other even more uncommon tumors of the exocrine pancreas include acinar cell carcinoma, lymphomas, and sarcomas.
BACKGROUND In this review, the common ductal adenocarcinoma will not be dealt with in detail except to serve as a baseline for comparison with the less-common tumors. Ductal adenocarcinoma accounts for approximately 75–90% of all pancreatic
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
368
GASTROINTESTINAL TUMORS
Table 1 Classification of pancreatic malignancies.
Tumors of presumed ductal cell origin Duct cell adenocarcinoma Giant cell carcinoma Giant cell carcinoma (epulis-osteoid type) Adenosquamous carcinoma Microadenocarcinoma Mucinous adenocarcinoma Cystadenocarcinoma Tumors of presumed acinar origin Acinar cell carcinoma Tumors of mixed acinar, ductal, and islet cell origin Tumors of uncertain origin Pancreatoblastoma Oncocytic carcinoma Papillary and cystic, solid and papillar tumor Small cell carcinoma Lymphomas Sarcomas Islet cell tumors Insulinomas Glucagonomas Somatostatinomas VIPomas Carcinoid Nonfunctioning islet cell tumor
malignancies.51 – 55 There is an overall male-to-female predominance of 1.5–2 : 1, although the ratio approaches unity with advancing age. There appears to be an increased incidence among blacks.4,52 The tumor occurs in all age-groups but is rare before the age of 25 years, most patients being in the 50–70 age-range. Approximately 70% of ductal adenocarcinomas occur in the head of the pancreas, with 20% occurring in the body, and 10% in the tail of the gland.53,54 The vast majority of patients present with at least regional metastatic disease, if not widespread disease. The most common sites of metastasis include: regional lymph nodes in 18–25%, liver in 21–80%, peritoneum in 11–23%, lungs in 7–28%, and abdomen in 25%.56 Presenting signs and symptoms are of a nonspecific nature, including weight loss and pain in over 70%; jaundice is present in 80% of patients with tumors of the head of the gland, but is uncommon in those with tumors of the body and tail. Other presenting symptoms may include anorexia, nausea, vomiting, weakness, and constipation. All of these are nonspecific, but are present in at least 25% of patients. Common presenting signs are jaundice and hepatomegaly in over 80% of patients with tumors of the head of the gland, while tumors of the body and tail more often present with hepatosplenomegaly, and abdominal mass or ascites. The current recommended diagnostic workup for a patient thought to have a possible pancreatic tumor is based on initial use of ultrasonography followed by a high-quality computed tomography (CT) scan of the upper abdomen to image the pancreas and its surrounding organs.57 If an abnormality is detected, endoscopic retrograde cholangiopancreatography (ERCP) with cytology is recommended, if available. If a lesion is identified, which appears to be localized and potentially resectable, the operating surgeon may wish to obtain a preoperative angiographic examination. If a patient has obvious metastatic disease at presentation, then diagnostic
evaluation should be limited to establishing a tissue diagnosis from the most available site. This may include fine needle biopsy of the pancreas.58 Morphologically, pancreatic adenocarcinomas present as ill defined, indurated masses gray-white or gray in color on cut section. Microscopically, these neoplasms are composed of different-sized ductlike structures, with an abundant stromal reaction. They vary from well-to-very poorly differentiated neoplasms. Mucin stains are usually positive.59,60 Immunohistochemistry (IHC) and genetic analysis have been undertaken and information concerning almost each pancreatic tumor histology is now available. Overexpression, mutations, and loss of heterozygosity (LOH) are among the specific abnormalities seen.
UNCOMMON EXOCRINE TUMORS OF THE PANCREAS Cystadenocarcinoma Cystadenocarcinomas represent 1–1.5% of pancreatic cancers.51,52 There are no known epidemiologic or etiologic factors that would separate this tumor from the ductal adenocarcinoma. Pathologically, these tumors are generally large, ranging in size from 2 to 30 cm in diameter, with a mean diameter of approximately 10 cm.61,62 Grossly, these tumors tend to be round; they may be lobulated or smooth, and usually are well encapsulated, although occasionally, the tumor is found to break the surface. One-half to two-thirds of these cysts are multilocular. The cyst walls range from 1 to 20 mm in thickness, and may be fibrous. Commonly, the cyst wall contains prominent vessels, and calcifications are not unusual. The cysts are filled with fluid, which may be mucoid or gelatinous. Microscopically, the cysts are lined by columnar cells, forming papillary structures. It appears that the majority of cystadenocarcinomas occur with coexisting areas of cytologically benign-appearing epithelium. This raises the question of whether a benign cystadenoma can give rise to its malignant counterpart.63,64 Most authors agree that the mucinous cystadenomas and cystadenocarcinomas are of ductal cell origin,62,65 while the serous cystadenomas are felt to be of acinar origin with little malignant potential. Confirmation of the malignant nature of a cystadenomatous lesion is based on (i) local invasion, (ii) involvement of regional nodes, (iii) cyst wall invasion, and (iv) histologic dedifferentiation. Most of these tumors tend to be of low biologic grade as evidenced by 11 of 21 tumors being graded as Broders’ grade 1 and 9 of 21 as grade 2 lesions in one series.64 Clinically, these tumors occur in patients aged between 20 and 80 years. The mean age is between 50 and 60 years and the median is 48 years.61 There is a slight female predominance, although not to the degree reported in the benign cystadenoma, where the female-to-male sex ratio varies between 4.6 : 1 and 9 : 1.61,63 Cystadenocarcinoma is remarkable for its long duration of symptoms prior to diagnosis, averaging 22 months in one report.66 The common presenting symptom is abdominal pain, usually in the left upper quadrant. Jaundice is remarkable for its absence,63 although jaundice has been reported to occur from mucin
UNUSUAL PANCREATIC TUMORS
obstructing the biliary tree without direct compression of the bile ducts by tumor.67 The clinical history is also remarkable for the absence, as a rule, of alcohol abuse, trauma, cholelithiasis, or prior pancreatitis.62,64,66 To make the diagnosis of this neoplasm, clinicians must have a high index of suspicion for a low-grade pancreatic tumor in patients with long duration of symptoms, vagueness of abdominal complaints, no jaundice, and no history of cholelithiasis, trauma, or alcohol or drug abuse. Diagnostic studies may be helpful in diagnosing cystadenocarcinoma. The plain abdominal films are reported to show a mass effect in 50–60% of cases, with calcifications being noted in 10%.63 Routine upper gastrointestinal series using barium are positive in 50–95% of cases, the most common abnormalities being either indentation of the lesser curvature of the stomach or C-loop deformities of the duodenum.61 As most of the literature dealing with these tumors antedates widespread use of ultrasonography or CT scanning, the exact value of these procedures cannot be stated on the basis of the available literature. However, on the basis of the utility of CT scanning in the diagnostic evaluation of the common ductal adenocarcinoma, one would presume that it would be the diagnostic procedure of choice, currently. It has been reported that the sunburst calcifications typical of these lesions are best appreciated with CT scanning.68 Endoscopic ultrasound (EUS) has also been shown to correctly predict malignant cysts in 89% of cases, and is, thus, a useful diagnostic tool.69,70 Once a cystic lesion of the pancreas has been demonstrated, the differential diagnosis must include benign cystadenoma, cystadenocarcinoma, and the more common pseudocyst. Because of the frequent coexistence of benign areas in the malignant cystadenocarcinoma, needle biopsies are felt to be inadequate and may not rule out the possibility of malignancy. The risk of seeding the biopsy tract has also been raised as an objection to this procedure, although the exact incidence of this complication is not known.61 If the diagnosis is in doubt, surgical exploration should be undertaken. At the time of surgery, approximately 50% of patients will be found to have localized disease without evidence of metastatic disease. Around 20–25% of patients will be found to have liver, nodal, or peritoneal spread at the time of exploration. In contrast with the more common ductal adenocarcinoma, curative resection may be attempted even in the presence of local nodal involvement or direct local extension to adjacent organs.61 The surgical procedure of choice is partly dictated by the location of the lesion, which is slightly more common in the body and tail than in the head of the gland. Therefore, distal pancreatectomy, and “cystectomy” with a rim of normal tissue have been the most common surgeries performed. Whipple procedures and total pancreatectomies have also been performed. Because of the small number of reported cases, the preferred surgery cannot be clearly determined. Survival in patients who are able to undergo a total excision is markedly better than those in the case of ductal adenocarcinoma.52 Reported 5-year survival rates range from 38 to 68% with 23% 10-year survivors in one series.61,66 In addition, patients who have a subtotal resection still have a
369
mean survival of 30 months from diagnosis to death, reinforcing the apparently more “benign” course of this tumor type.61 Reports of the use of chemotherapy and radiotherapy are too few to draw meaningful conclusions.
Adenosquamous Carcinoma Adenosquamous carcinoma, also known as adenocanthoma, has been reported to represent from 3.8%51 to as much as 11%71 of all pancreatic tumors. Most series report 3–4%. There appears to be a male-to-female ratio of approximately 3 : 1.53 These tumors tend to occur most frequently in the head of the pancreas, but can occur in more than one anatomic region.72 Because of the extremely small number of cases, it is difficult to define any specific epidemiologic or etiologic association. It has been suggested that there may be an association with chronic pancreatitis;73 this needs further substantiation. It has also been suggested that there may be an association with prior radiation exposure, with a relatively long latency prior to the appearance of the tumor, similar to radiation-induced thyroid disease.54 The validity of this theory awaits further reports. Pathologically, these tumors exist as a mix of squamous and adenocarcinomatous elements.74 Pure squamous cell carcinomas are very rare. Virtually all tumors show areas of transition from adenocarcinoma to squamous metaplasia to frank squamous cell carcinoma. As these tumors are felt to be of ductal origin, this implies that the adenocarcinoma cells are able to undergo metaplastic changes into squamous forms.71,74,75 It should be noted that, some authors74 have found small areas of squamous carcinoma in a high proportion of carefully sectioned tumors originally felt to be pure ductal adenocarcinoma. The gross appearance and size, 4–10 cm in diameter,71,74,76 of these tumors do not aid in differentiating them from the more common ductal adenocarcinoma. Immunoreactivity with keratin (AE1:AE3 and CK1) and expressions of CA 19-9 (84%) and carcinoembryonic antigen (CEA, 74%) are seen in the majority of cases of pancreatic adenosquamous carcinoma. Approximately 50% of these tumors contain K-ras oncogene mutations.72,77 There are no unique features in the clinical presentation of these tumors. Their diagnostic evaluation and management are little different from the approaches used for ductal adenocarcinoma. One feature of note is the unique angiographic appearance of these tumors. They are reported to have marked hypervascularity with an obvious tumor blush. This is said to distinguish the adenosquamous carcinoma and angiosarcoma from the common ductal carcinoma, which usually is hypovascular.78 Although the vascularity of adenosquamous tumors may be helpful in diagnosis, one should be aware that neuroendocrine tumors (carcinoid and islet cell neoplasms) involving the pancreas are also hypervascular and will blush upon angiography. Unfortunately, the prognosis of adenosquamous carcinoma is also similar, if not worse, to that of common ductal cancer. The mean survival in one series was reported to be 9 months from the onset of symptoms. In another series, none of 18 patients survived for 1 year.50 Whether chemotherapy regimens that are relatively potent in other gastrointestinal squamous cell cancers, such as those of the anus79,80 and esophagus,81,82
370
GASTROINTESTINAL TUMORS
will have benefit in adenosquamous cancers of the pancreas is unknown. Because of the relatively uncommon occurrence of this tumor, it will be very difficult to accumulate a series to test this hypothesis.
Acinar Cell Carcinoma Acinar cell carcinoma has been reported to represent 1–13% of all pancreatic tumors,51,83 with most series reporting 1–2%.51,53 There is a male predominance for this tumor. There are again no proven unique epidemiologic or etiologic features of this tumor when compared with ductal adenocarcinoma. It is of interest that acinar cell neoplasms can be induced in rats by using the carcinogenic agent, azaserine.84 To date, no definite environmental carcinogens have been linked to human acinar cell tumors. Grossly, these neoplasms are usually large, ranging from 2 to 15 cm in diameter. They are lobulated, soft, and fleshy in consistency, usually yellowish in color with obvious zones of necrosis and hemorrhage on gross sectioning. Microscopically, these tumors appear in an acinar arrangement of regular round or columnar cells with basally situated nuclei, a deeply situated eosinophilic cytoplasm, and occasional zymogen granules. Mitotic figures are usually sparse.85 These tumors may bear a slight resemblance to an islet cell tumor or small cell carcinoma. On electron microscopy, the cytoplasm is shown to contain abundant rough endoplasmic reticulum and numerous mitochondria. Electron-dense zymogen granules are also usually apparent.85 Between 30 and 40% of tumors have a mixed ductal and acinar composition.86 Clinically, acinar cell tumors occur in all age-groups, and are thought to occur at a slightly younger mean age than the common ductal adenocarcinomas.53,85 They have rarely been reported in children. Acinar cell tumors are located more commonly in the tail of the gland than in the head. Common clinical presentations include pain, weight loss, and/or an abdominal mass. Jaundice is uncommon, reflecting this tumor’s usual location away from the pancreatic head. A unique syndrome occurring in elderly patients, predominantly males, has been described in association with acinar cell carcinoma.86 – 88 Clinical features include malaise, eosinophilia, nodular skin lesions, and polyarthritis. Microscopically, the skin nodules reveal fat necrosis.85 The skin lesions first appear on the lower extremities, and later on the upper extremities, the chest, abdomen, and scalp.87 This wide distribution helps differentiate these lesions from erythema nodosum, with which they may be confused clinically.87 The arthritis involves both small and large joints. There is pain with motion, but little swelling, erythema, or warmth of the affected joints. Fat necrosis has been seen in or about the articular tissue.87 An eosinophilia ranging from 15 to 21% of total white blood cell count was reported in 70% of patients in one series.87 It is postulated that high circulating levels of pancreatic lipase are responsible for the fat necrosis.87 It is not unusual for patients with acinar cell tumors to present with obvious metastatic disease and an occult primary tumor. The paraneoplastic syndrome of thrombotic nonbacterial endocarditis occurred in 3 of 11 patients in one series, suggesting a high incidence of this complication in acinar cell
carcinoma.85 Diagnostically and therapeutically, these tumors do not differ from ductal adenocarcinoma. There are reports of markedly elevated CEA levels being found in patients with acinar cell carcinoma,89 but it is unclear whether this CEA elevation is a unique feature of this particular histology of pancreatic malignancy.88 Markedly elevated serum lipase levels and various chromosomal abnormalities (i.e. LOH at the pancreatic acinar cell carcinoma locus (APC) can help differentiate acinar cell carcinoma from other subtypes of pancreatic cancer.90 – 92 Mean survival from diagnosis is reported to range from 4.3 to 6.5 months, which is not significantly different from that of common ductal tumor.53,85 The largest retrospective review on pancreatic acinar cell carcinoma reports significantly improved overall survival in a series of 39 patients.93 Median survival was reported at 19 months and was prolonged in those with localized versus metastatic disease at diagnosis (38 vs 14 months, respectively). As expected, patients survived longer if they were amenable to surgical resection at diagnosis (36 vs 14 months). However, poor responses to chemotherapy were seen, suggesting the need for novel therapies.
Pleomorphic Adenocarcinoma Pleomorphic adenocarcinoma, also known as giant cell carcinoma or sarcomatoid carcinoma, represents 2.1–12.8% of pancreatic malignancies.56 This tumor has no unique epidemiologic features. The exact cell of origin is debated, with some authors favoring a true mesenchymal origin based on the tumor’s sarcomatous appearance. The acinar cell origin has been proposed on the basis of electron microscopy suggestive of this diagnosis.94 Still, the most widely accepted hypothesis is that this neoplasm originates from the ductal cells.95 – 97 Clinically, there is a slight male over female predominance.98 This tumor most commonly occurs in the seventh decade of life with the mean age being 6799 and the median age 65.98 There is a pronounced tendency for these tumors to occur in the pancreatic body and tail,53,97,98,100 although in one series there was an equal distribution among head, body, and tail. The clinical presentation of pleomorphic carcinoma resembles that of other tumors occurring in similar locations in the pancreas. Abdominal pain, weight loss, anorexia, and left upper quadrant mass are all common presenting complaints. Ascites may occur.98 Jaundice is uncommon, as one would expect. Most patients are found to have widely metastatic disease at the time of diagnosis. Locoregional disease and distant hematogenous spread are both common. Common sites of metastases include the liver, lungs, and regional nodes; of interest in one series was a significant incidence of spread to mediastinal and hilar nodes, which is unusual for common ductal adenocarcinoma.98 Pancreatic carcinomas also tend to be locally invasive.97 Pathologically, these tumors are usually large, with a mean size of 8.9 cm in diameter, with a range of 4.5–18 cm.98 Areas of gross hemorrhage and necrosis are common. Microscopically, they grow in a loose sarcomatoid pattern with mono- and multinucleated giant cells. The giant cells themselves are large pleomorphic cells, having vacuolated
UNUSUAL PANCREATIC TUMORS
nuclei with prominent nucleoli and coarse chromatin and eosinophilic cytoplasm. The giant cells are interspersed on a background of malignant tumor cells, frequently in a spindle cell configuration with fibrous septa of reticulin fibers around the tumor cells.98 Cytophagocytosis of red blood cells (RBCs) by the tumor cells is common, as is cannibalism of tumor cells by the malignant giant cells. Care must be taken to differentiate this tumor from melanoma, choriocarcinoma, sarcoma, hepatocellular carcinoma, angiosarcoma, and liposarcoma.98,99,101 On angiography, these tumors appear to be hypervascular, helping to differentiate them from ductal adenocarcinoma.102 There is nothing atypical about the diagnostic or therapeutic approach to this tumor type. However, IHC and molecular analysis have shown giant cell carcinoma to express osteopontin and (vascular endothelial growth factor) VEGF-C. Mutations of K-ras were also seen. Her-2/neu was not found to be overexpressed and neither microsatellite instability (MSI), p53 or p16 mutations, nor LOH were seen in these tumors.103,104 APC and DPC4 gene loci were not found. Identification of IHC markers and molecular analysis may aid in the diagnosis and treatment of these tumors in the future. Mean survival from diagnosis is usually 3–4 months,98,100 although occasional long-term survivors have been reported. There is a question whether these long-term survivors represent a subgroup of the giant cell carcinoma patients, in that the giant cells are of a morphologically different variety than in the more common form. The giant cells in this subgroup resemble osteoclasts and therefore this subgroup has been referred to as the epulis-osteoid type of giant cell carcinoma.53,105 Whether this represents a unique subtype of neoplasm with a better prognosis has not been resolved.59
Microadenocarcinoma Microadenocarcinoma represents 1–3% of pancreatic tumors.53 The mean age of patients in one small series was 39.7 years.59 These tumors tend to be large, with a median diameter of 14 cm. Both the head and the tail of the pancreas have been reported as common sites of occurrence for this neoplasm.53,59 Microscopically, small glands of more uniform and smaller size than those in the more common ductal tumor are present. These glandular structures are centered in sheets of tumor cells. The nuclei of the cells are uniform and of intermediate size. The cytoplasm is scant and pale-staining. Mucin stains are positive. There is more necrosis and less fibrosis than is seen with ductal adenocarcinoma.53 Care must be taken to distinguish the microadenocarcinoma from carcinoid tumor. Clinically, microadenocarcinoma patients typically present with disseminated disease and their prognosis is grim, with a mean survival of 2–11 months.53,59
Cystic Neoplasms Cystic neoplasms of the pancreas include mucinous and colloid adenocarcinomas, papillary and cystic carcinomas, and intraductal papillary-mucinous tumors (IPMT). Mucinous and colloid adenocarcinoma represents approximately
371
2% of pancreatic malignancies.53 There is a marked male predominance. The median age is similar to that of the more common ductal tumor. A history of chronic pancreatitis and ethanol abuse is common in these patients, although there is no proof of a causal relationship. The second cystic neoplasm variant, papillary and cystic carcinoma, also known as solid and papillary carcinoma, is rare; the exact incidence is unknown, but is only at the most 1% of pancreatic tumors. It is most common in young females, with a mean age of 24 in one series.106 This tumor may be more common in AfricanAmericans. Patients most commonly present with complaints of an abdominal mass with or without pain, most often in the left upper quadrant. The mass may have been present for a prolonged period, in one case greater than 2 years. The most common location is in the head of the gland. Thirdly, IPMTs of the pancreas are characterized by dilated pancreatic ducts and ductules lined with tall columnar mucin-producing neoplastic epithelial cells. It is suggested that these tumors are less aggressive and have a better prognosis than pancreatic ductal carcinoma.107 This difference in aggressiveness has been postulated to be related to the fact that the DPC protein is strongly and universally expressed in IPMTs, whereas this protein is not expressed in the majority of pancreatic ductal adenocarcinomas.108 Mucinous and colloid adenocarcinomas tend to present with a large, soft mass in the head of the pancreas. Grossly, they have a mucoid or gelatinous appearance. Microscopically, there are large cystic spaces filled with mucin, and lined by tall columnar, nonpapillary glandular epithelium. Nests of tumor cells are found floating in lakes of mucin.53 Grossly, papillary and cystic carcinoma are large, with a mean diameter of 10 cm.106 They are encapsulated by fibrous tissue109 and are obviously hemorrhagic on cut section. Microscopically, solid and cystic areas with hemorrhage are noted. Sheets of cells surround thin-walled vessels containing intact RBCs. Other areas show a distinct papillary configuration. Nuclei are round with inconspicuous nucleoli, and the chromatin is finely dispersed.53 Cholesterol granulomas are common. Tumors have been reported to stain positive for α 1 antitrypsin.109,110 periodic-acid Schiff (PAS)-positive material has also been noted in some cystic spaces.109 Electron microscopy reveals abundant mitochondria. Some reports demonstrate prezymogen granules,111 although others report an absence of zymogen granules. The exact cell of origin is debated; some feel this is an acinar cell-derived tumor,109,111 while others believe that it arises from the cells of the smallest pancreatic ducts. Intraductal papillary-mucinous histology can be associated with invasive carcinoma112,113 . Noninvasive IPMTs are classified as adenoma, borderline, or carcinoma in situ. Invasive IPMTs are of the tubular, colloid, mixed, or anaplastic types. Mucinous adenocarcinoma, as well as ductal adenocarcinoma have been shown to overexpress KOC, a K homology domain protein. Recently, a monoclonal antibody, specific for KOC, has been shown to be very sensitive and specific for pancreatic dysplastic lesions.114 KOC staining has been seen in 97% of carcinomas, 94% having moderate-to-strong staining. Thus, immunostains directed against KOC may be of diagnostic utility in mucinous, as well as other types of,
372
GASTROINTESTINAL TUMORS
pancreatic cancer. Additionally, MSI (mutations in mismatch repair genes), typically seen with mucinous carcinomas in other solid organs, are not normally exhibited in pancreatic mucinous carcinomas.115 Different types of IPMTs exhibit different levels of mucin gene expression. On the basis of the patterns of mucin expression, invasive IPMTs can be further classified into two types: a colloid form and an ordinary form.116 The ordinary form shows a pattern of mucin gene expression similar to that of ductal adenocarcinoma and has a greater tendency to metastasize. When examining other areas of gene expression profiling, besides mucin staining, significantly higher levels of expression were seen in IPMTs than in normal pancreatic tissue, and even higher levels of transcript expression in invasive versus noninvasive IPMTs. Other studies support these differences in gene expression between invasive and noninvasive IPMTs.107,117 Preoperative assessment of gene expression profiles may be a useful tool, helping in the differentiation of patients with invasive components to their disease. The median survival in patients with mucinous adenocarcinoma has been reported to be 9–11 months;53 in one small series, three of five patients were alive at 1 year but none at 5 years. It has been suggested that this tumor may have a slightly better prognosis than the common ductal adenocarcinoma, but this would not appear to be significantly better. Papillary and cystic carcinoma has a better prognosis than the common ductal adenocarcinoma, even when metastatic disease is found. Exact survival figures are not available but many series report most, if not all, patients alive and free of disease after simple resection.106,118 Radical pancreatectomy is not felt to be necessary by most authors. Deaths due to metastatic disease have been reported, however. There is one report of a patient with locally advanced unresectable disease treated with 4000 cGy external beam irradiation, who experienced significant tumor regression.119 The exact role that radiation may play in this tumor is unclear, but it could be considered in cases where palliation is appropriate. The overall 5-year survival for patients having IPMTs without invasive cancer was 77%, compared to 43% in those patients with invasion.113 For patients with invasive IPMTs, 2-year survival was 40% when margins were positive and 60% when margins were free of tumor. IPMTs of the colloid type were associated with improved survival, compared to those of the tubular type (5-year survival rates of 83 and 24%, respectively). Similar survival estimates have been reported by multiple investigators.120 Predictors of poor outcome were presentation with elevated bilirubin levels, invasive carcinoma, increasing size and percentage of invasive cancer, histologic type of invasive carcinoma, positive lymph nodes, and vascular invasion.
Extrapulmonary Small Cell Carcinoma Extrapulmonary small cell carcinoma accounts for 1% of pancreatic cancers. There were 5 out of 485 (1.0%) and 7 out of 508 (1.4%) small cell cancers reported in two large series of pancreatic malignancy.53,121 There is a male predominance. The age-range in one series was 42–73 years with a mean age of 60.121 Most patients present with jaundice and weight loss. Eighty percent of tumors are located in
the head of the pancreas, with a mean diameter of 4.2 cm. Grossly, the cancer is firm, gray-white, and with areas of necrosis and hemorrhage. Local and regional dissemination of disease is extensive.121 Metastatic disease is noted in regional and distant nodes, liver, and lungs. Microscopically, the tumor consists of sheets and nests of small, round cells with markedly hyperchromatic nuclei and poorly defined cytoplasm. A scant fibrous stroma is also noted. Mitotic activity is prominent. Serum neuron-specific enolase (NSE) levels may be elevated in certain cases and may be a useful diagnostic tool that aids in evaluating response to treatment.122 Clinically, patients with small cell cancer have done very poorly, with all patients reported dead within 2 months in one series.121 Use of aggressive chemotherapy has been reported in two patients and a pathologic complete remission by laparotomy was documented in one case.123 In view of the advances in the therapy for pulmonary small cell carcinoma, it is reasonable to approach pancreatic small cell cancer in a similar manner. The use of combination chemotherapy should be the initial approach, supplemented with radiation or surgery should the clinical situation so indicate.124 There are at least two reported cases of ectopic hormone production by a pancreatic small cell tumor. One was shown to produce adrenocorticotropic hormone (ACTH), while the other produced an unidentified substance, causing hypercalcemia.125,126 Several authors warn that the presence of a lung primary must always be ruled out, since the vast majority of patients with a pancreatic small cell cancer actually have pancreatic metastases from a lung primary.124,127 Small cell carcinoma of the gastrointestinal tract is also addressed in Chapter 38, Small Cell Carcinomas of the Gastrointestinal Tract.
Pancreatoblastoma Pancreatoblastoma is a very uncommon pancreatic malignancy, representing less than 1% of pancreatic cancers.53 It presents in a very young population, with an age-range of 15 months to 13 years in one series.128 Patients present with an abdominal mass and discomfort. Grossly, these tumors present as encapsulated masses, most commonly in the head of the pancreas.129 They may be large in size, some reported to be greater than 10 cm in diameter.73,128 Microscopically, these neoplasms have small cells with scant cytoplasm; these cells occur in sheets, but with focal glandular arrangements. Mucin stains are positive in the glandular lumina. Mixed with the small cells are larger, deeply eosinophilic cells with zymogen granules present in the cytoplasm. Irregular nests of well-differentiated epidermoid cells without keratinization are also present. The exact cellular origin of these tumors is unclear.73,128 – 132 Gene analysis has revealed expression of certain proliferation markers (i.e. topoisomerase II α) and aberrant expression of tumor suppressor genes.133 Clinically, patients are reported to do well after surgical removal although the follow-up in most cases has been short.
Oncocytic Carcinoma Oncocytic carcinoma is very rare, representing much less than 1% of pancreatic malignancies. Cases have been too
UNUSUAL PANCREATIC TUMORS
few to state age or sex trends with certainty. Presenting signs and symptoms are usually related to an enlarging mass, with or without abdominal pain. Grossly, the tumors are usually large, with diameters of 7–12 cm being reported.59,130 These neoplasms may be more common in the tail of the gland. The cut surface is yellow-white with focal hemorrhage.134 The tumor is composed of solid sheets of cells with abundant granular, eosinophilic cytoplasm. The nuclei are large and irregular in size and shape. The chromatin pattern is coarse and nuclei are prominent. Fine bands of fibrous tissue are present. Rare glandular structures may be seen.134 Remarkable features on electron microscopy include abundant mitochondria and a scarcity of other cytoplasmic organelles. Also notable are the lack of zymogen and neurosecretory granules.134 Molecular analysis shows no mutations at tumor suppressor loci.135 The clinical course of this tumor is marked by malignant behavior as evidenced by local splenic invasion, perineural invasion, and peripancreatic nodal involvement. The prognosis cannot be compared to that of the more common ductal cancer because of the rarity of oncocytic carcinoma.
Lymphoma Primary lymphomas of the pancreas are extremely uncommon. However, pancreatic involvement of intra-abdominal lymphoma is not very rare.131,132,134,136,137 One series of primary pancreatic lymphomas found a male predominance, with a mean age of 65 at diagnosis. This disease occurs typically in the head of the pancreas and presents with abdominal pain and weight loss.138,139 Most lymphomas involving the pancreas are large cell tumors or small-cleaved cell tumors. When dealing with a possible pancreatic lymphoma, the problem of differentiation from an anaplastic carcinoma must be kept in mind, and adequate tissue for appropriate immunohistochemical and/or molecular analysis must be obtained. The use of needle biopsies or fine needle aspirations (FNAs) in these instances may therefore be unrewarding.140 Therapy should be based on stage and histology, as with other lymphomas.
Sarcoma Sarcomas clearly originating in and localized to the pancreas are rare. Pancreatic metastases from retroperitoneal sarcomas occur much more commonly than primary sarcoma of the pancreas. In reviewing case reports of sarcoma involving the pancreas, it is frequently difficult to be sure of the exact histologic diagnosis and whether the tumor is indeed primary to the pancreas.141,142 It is important to realize that diagnostic distinction among sarcoma, lymphoma, and anaplastic carcinoma is difficult and requires obtaining adequate tissue (perhaps by open biopsy) for IHC and electron microscopy. Needle biopsy and FNA cytology technologies are frequently inadequate for diagnosis.
Mixed Cell Tumor Mixed cell tumors of the pancreas have been reported.143,144 Combinations of acinar, ductal, and islet cell differentiations, including at least one case of trilineage differentiation,
373
occur. This is not totally unexpected, as experimental embryologic studies have provided evidence for differentiation of endocrine cells within the endodermal epithelium.145 – 147 Cases of these neoplasms have been far too few to make accurate statements about clinical associations or behavior, although reported cases have shown malignant behavior with infiltrative growth, perineural invasion, and nodal and distant metastases.144,145
ENDOCRINE TUMORS OF THE PANCREAS The pancreas has two functions, an exocrine function and an endocrine function. The previously discussed neoplasms of the pancreas have all arisen from one or another of the cellular components of the exocrine pancreas. Tumors of the endocrine pancreas are described in this section. The tumors of the endocrine pancreas fall into two groups: (i) islet cell tumors, and (ii) carcinoid tumors. As a general entity, endocrine neoplasms of the pancreas have been of great interest to pathologists, oncologists, and endocrinologists because of their unique clinical and pathologic features and the protean clinical manifestations that can be caused by the biologically active materials secreted by these neoplasms. Telomerase, a ribonuclear protein inducing cell immortality, is present in the vast majority of pancreatic adenocarcinomas and in other exocrine tumors (i.e. acinar cell carcinomas) and is associated with aggressive tumor behavior. However, telomerase activity is present in virtually no endocrine pancreatic tumor.148 This finding may be responsible for the less aggressive and more indolent course of these endocrine tumors and may be useful in distinguishing these tumors from other histologic subtypes. Additionally, the DPC genes affected in virtually all pancreatic ductal adenocarcinomas are not mutated or deleted in the majority of islet cell tumors of the endocrine pancreas, possibly accounting for differences in aggressive behavior between these tumors.149
Islet Cell Tumors Islet cell tumors represent the most common endocrine neoplasms of the pancreas. The vast majority of these tumors remain asymptomatic, since there is a marked difference in prevalence of islet cell tumors in autopsy series versus clinical series. Incidence rates as high as 1500 per 100 000 have been reported in autopsy series, whereas the clinical incidence of islet cell neoplasia is approximately 1 per 100 000 population.150,151 The pathology of islet cell tumors depends not only on the light microscopic identification of neoplasia of islet cells but also on sophisticated staining and/or other techniques that define what particular hormone or hormones the neoplasms produce. The classic Grimelius silver stain152 was one of the early approaches to defining a tumor as a pancreatic endocrine tumor. Now, however, electron microscopy, which is capable of defining the morphology of granules within the islet cells, and IHC, which may stain for particular polypeptides produced by endocrine tumors, are very helpful techniques. In poorly differentiated tumors, histochemical staining for NSE153 and chromogranins,154 and specific techniques for detecting neuroendocrine tissue are
374
GASTROINTESTINAL TUMORS
Table 2 Pancreatic islet cell tumors.
Cell type (immunohistochemical staining)
Product
Syndrome
A B D
Glucagon Insulin Gastrin, somatostatin
D1 PP
VIP Pancreatic polypeptide Chromogranin A and B
Glucagonoma Insulinoma Zollinger-Ellison syndrome, somatostatinoma Pancreatic cholera No clearly defined syndrome No syndrome
Nonfunctioning
relatively simple. Since islet cells are of neuroendocrine origin, this approach can be useful in defining these neoplasms. Table 2 defines the islet cell neoplasms on the basis of the polypeptides produced by these tumors and the clinical syndromes that may be associated with them. Occasionally, more than one product may be secreted by islet cell tumors. Islet cell tumors generally occur sporadically. The etiology of these malignancies is unknown. However, there are familial syndromes in which neuroendocrine tumors occur. Wermer155 and Sipple156 describe two inherited syndromes of multiple neuroendocrine neoplasia, which clinicians must be aware of. In Wermer’s syndrome or multiple endocrine neoplasia (MEN type 1), patients may have multiple, small tumors157 of the pituitary gland, thyroid gland, and pancreas, and most commonly have islet cell tumors. In Sipple’s syndrome (MEN type 2A), patients present with parathyroid hyperplasia, pheochromocytomas, and medullary carcinoma of the thyroid, but uncommonly have pancreatic tumors. In MEN type 2B, parathyroid hyperplasia does not occur but bony abnormalities, neuromas, and marfanoid features may occur. The importance of these inherited neuroendocrine tumor syndromes concerns screening for the detection of families at risk for neuroendocrine neoplasms. Patients with islet cell, thyroid, parathyroid, adrenal, or pituitary neoplasms should be questioned in regard to any familial history and, if present, family members should be evaluated for the possible presence of a MEN syndrome. There is clear evidence that molecular genetic abnormalities may be useful in screening for MEN 1 and MEN 2. In MEN 1 a genetic locus on the long arm of chromosome 11 (11q12-13) has been recognized as being associated with the syndrome.158 Mutations involving the Ret oncogene in the pericentromeric region of chromosome 10 have been demonstrated in patients with MEN 2A. It is possible to screen asymptomatic family members for these abnormalities.159 Individuals with the Ret mutation should be considered for early thyroidectomy to prevent death from medullary carcinoma of the thyroid.
Islet Cell Tumor Syndromes Zollinger-Ellison Syndrome
Zollinger-Ellison syndrome (ZES) results from the secretion of gastrin by an islet cell tumor containing δ granules. The
typical patient presents with symptoms marked by recurrent gastric ulceration due to hypersecretion of gastrin and resultant gastric acid production,160 gastrointestinal bleeding, and pain.161 Occasionally, patients will also have a syndrome of diarrhea, which is secondary to rapid intestinal transit time.162 ZES may also be a manifestation of the MEN 1 syndrome. Diagnosis of ZES depends on documenting elevations in gastric acid output and fasting serum gastrin levels. A fasting serum gastrin level of greater than 1000 pg mL−1 is considered diagnostic of ZES.163 In patients suspected of this syndrome with less elevated serum gastrin levels, a secretin stimulation test may be performed (intra-arterial injection of secretin, combined with venous sampling).164 If the serum gastrin increases to greater than 200 pg mL−1 over baseline after administration of secretin, this is strong evidence for the existence of ZES. Although the vast majority of patients with ZES will have islet cell tumors of the pancreas, occasionally the syndrome is caused by extrapancreatic tumors, typically in the duodenal wall.162 CT scanning and ultrasonography will occasionally demonstrate a mass in the pancreas and also may demonstrate liver metastases. Prominent gastric body folds on endoscopy have been noted in 94% of ZES patients.161 Techniques including portal and other selective venous sampling for gastrin, EUS, and intraoperative ultrasonography (IOUS) of the pancreas and duodenum to detect small tumors are frequently useful.163 EUS aids in the evaluation of the number, size, and anatomic characteristics of gastrinomas. Octreoscan, a diagnostic scan utilizing a radiolabeled somatostatin analog, is best for evaluation of metastases to the liver, mediastinum, and bones.165 Gastrinomas are rich in somatostatin receptors and screening with radiolabeled octreotide preoperatively may define the location of the tumor. A genetic abnormality (LOH) associated with chromosome 1 has been documented in a subset of patients with gastrinoma.166 This genetic abnormality is associated with aggressive growth, presence of liver metastases, and postoperative development of liver metastases. Often, the assessment of whether a gastrinoma is benign or malignant can only be made at the time of surgery. Approximately 75% of patients with ZES have malignant tumors. The malignancy or benignity of an islet cell tumor is not defined by histopathology but rather by the size, the presence or absence of local invasion, and/or metastases. Tumors greater than 3 cm in size are often associated with liver metastases. The treatment of ZES entails antitumor therapy and therapy aimed at the inhibition of the physiologic effect of gastrin. In the past before the advent of potent H2 histamine receptor blockers and Na+ /K+ membrane pump inhibitors, total gastrectomy was a frequent therapeutic approach to patients with ZES.167 Obviously, with the removal of the end organ affected by hypersecretion of gastric acid, symptoms could be relieved. Now with the use of pharmacologic methods including H2 blockers and the Na+ /K+ membrane pump inhibitor omeprazole, inhibition of gastric acid secretion secondary to hypergastrinemia can be achieved, pharmacologically.163 Frequently, the doses required to effectively suppress gastric acid secretion in ZES
UNUSUAL PANCREATIC TUMORS
are considerably higher than the standard dosages of H2 blockers. Omeprazole is highly effective and is considered the drug of choice. Effective antitumor therapy168 can be carried out by resection of the tumor mass that secretes gastrin and by the use of partial cytoreduction with surgery and cytotoxic chemotherapy. Since the chemotherapy for all islet cell tumors is similar, this will be discussed at the end of this section. Insulinoma
These are the most common islet cell tumors, with an incidence of one in 250 000, and are neoplasms of β cells. They occur more frequently in women than men, with a mean age at occurrence of 45 years. Over 90% of insulinomas are single, with even distribution between the head, body, and tail of the pancreas.169 – 171 Immunohistochemically, these tumors are strongly immunoreactive to insulin, have atypical membranous electron-dense granules, and have a high Ki67 labeling index.172 Whipple’s triad (symptoms of hypoglycemia, glucose level below 50 mg dL−1 , and relief of symptoms by administration of glucose) and an index of insulin release >0.3 are necessary for the diagnosis of insulinoma.169,171,173,174 The diagnosis of insulinoma depends on the demonstration of hypoglycemia with inappropriately elevated plasma levels of immunoreactive insulin.168 Fasting hypoglycemia with elevation of serum plasma insulin levels, particularly if “big” insulin (proinsulin) is elevated, makes diagnosis of insulinoma highly likely. Occasionally, patients will have hypoglycemia and elevated insulin levels secondary to the surreptitious self-administration of insulin. In the past, if this was suspected, the plasma would be assayed for anti-insulin antibodies, which develop when porcine or beef insulin is injected.168 However, with the widespread use of recombinant human insulin, the diagnosis of factitious hypoglycemia is more problematic. Preoperative diagnosis of small insulinomas is aided by a selective arterial calcium stimulation test and hepatic venous sampling, using an intra-arterial calcium injection.175 Despite these tests, intraoperative localization using IOUS is superior to preoperative localization.170 Insulinomas are usually benign, with only 10% being malignant. Resection of the primary tumor will effectively manage this syndrome. Glucagonoma
Islet cell tumors containing α granules may produce glucagon.168 The physiologic effect of glucagon is to increase serum glucose levels by stimulating gluconeogenesis and glycogenolysis in the liver. In patients with glucagonoma, the hypersecretion of glucagon results in carbohydrate intolerance and hyperglycemia. The hyperglycemia is usually mild. In addition to metabolic abnormalities, these patients may also present with a characteristic dermatitis termed necrolytic migratory erythema.176,177 Other features including weight loss, anemia, diarrhea, and hypoaminoacidemia have also been noted.178,179 It should be emphasized that glucagonomas are very rare tumors. On presentation, this tumor is commonly malignant with most patients having metastatic disease to the liver.168,180 The diagnosis of glucagonoma depends on demonstrating that a
375
patient with hyperglycemia and other clinical manifestations of this syndrome has grossly elevated plasma glucagon levels. In many instances, glucagon levels are in excess of 2000 pg mL−1 .168,181 Somatostatinoma Another extremely rare islet cell tumor, somatostatinoma was originally described in 1977.181 Fewer than 30 patients with somatostatinoma have been described in the world literature. Somatostatin-producing tumors contain δ granules and are associated with mild diabetes, steatorrhea, and cholelithiasis.182 Elevated somatostatin levels may be measured by serum radioimmunoassay in these patients. The overall natural history of this disease is unknown, although most patients have malignant disease with metastases to the liver.183 Since somatostatin produces very mild hormonally mediated symptoms, patients may present only after symptoms caused by bulk metastatic disease have occurred. It is thus likely that patients will have a natural history similar to that seen in ZES, where most patients have hepatic metastases at the time of diagnosis.184 There is little information available for the treatment of this syndrome. Certainly, surgical resection of the tumor should be attempted if it is technically feasible, and nonsurgical approaches such as chemotherapy may have some beneficial effect. VIPoma This tumor is responsible for the watery diarrhea, hypokalemia, and achlorhydria (WDHA) syndrome.168 Another name for this condition is pancreatic cholera syndrome. Although profuse diarrhea may be seen with ZES, all those patients have increased gastric acid production and thus do not exhibit achlorhydria. In the past, there has been a debate in the literature with regard to the particular hormone responsible for the WDHA syndrome. Now, there is general agreement that vasoactive intestinal peptide (VIP) is the major polypeptide producing this syndrome.183 Strong presumptive evidence for causation by VIP has been recorded from studies in which normal volunteers have been given VIP infusions and the syndrome has been reproduced.183 The clinical manifestations of this disease are characterized by profuse watery diarrhea and hypokalemia. A characteristic of the WDHA syndrome is that diarrhea persists even when all oral intake has been stopped, thus defining the diarrhea as a secretory diarrhea. The hypokalemia is presumably based on the potassium loss in diarrheal stools. In one series, periodic backache, skin rash, and polyps of the colon were reported with VIPoma. Backache and skin rash were noted to disappear following surgery, as did the elevation in VIP level.185 The great majority of patients with VIPoma present with metastatic disease to the liver. Antitumor therapy (surgery and chemotherapy) is similar to that for other islet cell tumors. Because of the secretory nature of the diarrhea, patients with WDHA may have excellent symptomatic response to the somatostatin analog octreotide (described below). Nonfunctioning Islet Cell Tumors Nonfunctioning islet cell tumors (NIT) are rare pancreatic neoplasms, characterized by nonspecific clinical symptoms.
376
GASTROINTESTINAL TUMORS
Tumors are typically located in the pancreatic head, but may be seen in the body or tail. In one series, mean tumor diameter was 10.7 cm. Similar to pancreatic adenocarcinoma, approximately one-half of NITs reveals point mutations or deletions in the DPC genes.149,186 Curative resections, with either enucleation or pancreaticoduodenectomy, were seen in 88% of patients, with low rates of local recurrence or metastatic disease. Pancreatic polypeptide (PP) is secreted in patients with islet cell tumors without syndromes of hormone excess.187 In such patients, metastatic malignant islet cell tumor may present with hepatomegaly and concomitant elevation of serum levels of pancreatic peptide,168 which produces no clinical hormonal abnormality. Other hormones measured, including insulin, glucagon, somatostatin, VIP, and gastrin, are low, as is urinary excretion of 5-hydroxyindoleacetic acid (5-HIAA). This islet cell tumor may have a significant relationship with the MEN 1 syndrome. While screening family members in the MEN 1 families, the presence of elevated serum levels of PP may be an indication of clinically occult pancreatic tumors. Removal of these tumors resulted in normalization of the PP levels.188 In other patients with islet cell tumors, no hormone syndrome exists and PP is not secreted. These tumors probably produce other nonfunctional polypeptides or fragments of polypeptides. It has also been demonstrated that almost all NITs produce elevated plasma levels of chromogranin A and B.154
Treatment of Islet Cell Tumors In a patient with a hormonally active islet cell tumor, surgical intervention is the primary treatment with the goal being the removal of as much tumor mass as possible.168 Vigorous attempts should be made to achieve preoperative localization of the tumor. Such attempts would include localization of the tumor within the pancreas by use of CT scan, ultrasonography, arteriography, and selective venous sampling for the localization of hormone-producing tissue. IOUS has also been shown to aid in tumor localization.157 In patients without metastatic disease, once the tumor is localized in the body and tail of the pancreas and if it is greater than 1 cm in size, a distal pancreatectomy may be performed. If the tumor is less than 1 cm in diameter or if it is located in the head of the pancreas, enucleation of the tumor mass may be performed. In patients with metastatic disease to the liver, the most likely site of metastasis,189 who have significant symptoms from polypeptide production one may still consider surgical resection of functioning metastases. Other cytoreductive approaches that may be useful are embolization of metastases via the hepatic artery, ligation of hepatic artery, radiofrequency ablation (RFA) and, rarely, hepatic radiation therapy.190 – 192 Factors related to achieving a sustained response following treatment are surgical resection of the primary tumor, four or more embolization procedures, and liver metastases of 5 cm or less.193 In one series, median survival following cytoreductive treatment was 32 months.193 Another recent 5-year series involving 20 patients followed up for 19 months, showed 90% of patients alive and 60%
disease-free, following various surgical procedures.194 Overall survival at 5 years was 80% and morbidity and mortality associated with surgical intervention was low. Despite acceptable toxicity and excellent survival rates, most patients will develop a tumor recurrence. The reason why partial resection and incomplete cytoreductive therapy are viable alternatives in the treatment of islet cell tumors is the natural history of these tumors. Growth of islet cell neoplasms may be very slow and the patient’s major morbidity may be related to the hormonal products being produced by the tumor. Thus, partial resection may give patients long periods of significant remission from the morbidity of their disease.168 Medical Management of Islet Cell Tumors
When surgical resection of an islet cell tumor fails or is contraindicated, medical management is considered. There is clear evidence that the somatostatin analog octreotide is useful in the treatment of islet cell tumors.195 Octreotide is a synthetic octapeptide that has all the physiologic characteristics of somatostatin. Somatostatin is able to decrease peptide release from islet cell tumors. However, the normal plasma half-life of somatostatin is very short so that, apart from approaches using continuous infusion, therapy for islet cell tumors would be impractical. Octreotide LAR, a longacting somatostatin analog, has been evaluated in patients with islet cell malignancies, ill with symptoms from tumorproduced polypeptides.196 Beneficial effects, as defined by decrease in the secretion of tumor-produced polypeptides and symptoms, have been seen in patients with VIPomas, ZES, carcinoid tumors, and glucagonoma196,197 following the use of somatostatin analog. There is evidence that insulin levels are reduced in 60–70% of patients with insulinoma treated with octreotide.198 A small percentage of patients treated with the drug experience objective tumor regression.199 Separate from octreotide, patients with ZES can experience relief from all symptoms of gastrinoma with high doses of H2 blockers or preferably the Na+ /K+ pump inhibitors.200 Therapy with diazoxide201,202 may effectively prevent hypoglycemic crises in insulinomas. Biochemical partial responses or stabilization in all endocrine pancreatic tumor subtypes have been reported with single-agent α-interferon and, in patients progressing on treatment with a somatostatin analog, the addition of α-interferon may be of added value.203 This therapy does, however, have significant toxicity that has limited its widespread use in clinical practice. The Chemotherapy of Islet Cell Tumors
A characteristic of all islet cell tumors is the relative indolence of their progression. When compared with other malignant tumors of the gastrointestinal tract, such as the common adenocarcinoma of stomach, pancreas, and colon, metastatic islet cell carcinoma is a very slowly progressive disease. For this reason, patients are considered candidates for cytotoxic chemotherapy only when other procedures have failed. Such palliative approaches include hepatic-directed therapies, for example, RFA, chemoembolization, surgical
UNUSUAL PANCREATIC TUMORS
debulking, as well as antihormonal therapy.150 Because these patients may then have a poor performance status and be quite ill, they may be relatively poor candidates in whom response to chemotherapy cannot be expected. The major drug that has been shown to be useful in islet cell tumors is the naturally occurring methyl nitrosourea streptozotocin. This drug is known to have a toxic effect on the pancreatic β cells in animals and was therefore thought to be beneficial for humans with insulinoma.204 Clinical trials with streptozotocin used alone showed an approximate 10–40% response rate in metastatic islet cell carcinomas.205 Combinations of streptozotocin and 5-fluorouracil (5-FU) have also been shown to be beneficial, producing response in 50–60% of patients.206 Dacarbazine and cyclophosphamide also appear to have some activity in pancreatic endocrine tumors.207 In 1992, the Eastern Cooperative Oncology Group published the results of a study comparing 5-FU plus streptozotocin with doxorubicin plus streptozotocin and a third arm of chlorozotocin alone in105 cases of islet cell carcinoma. This study208 demonstrated that doxorubicin plus streptozotocin was significantly superior to 5-FU plus streptozotocin in response rate (69 vs 45%, p = 0.05) and in the interval to tumor progression (median 20 vs 6.9 months, p = 0.004). The chlorozotocin arm produced a response rate of 30%. These data suggest that doxorubicin plus streptozotocin is the treatment of choice for patients with islet cell tumors who reach a stage in this disease in which cytotoxic chemotherapy is required. In addition, preliminary data from a phase II study of sunitinib (SU011248) suggest activity for this multitargeted tyrosine kinase inhibitor in patients with metastatic islet cell tumors.209
Carcinoid Tumors Carcinoid tumors, like islet cell tumors, develop from neuroectodermal tissue and may also be associated with other endocrine tumors in the MEN 1 and MEN 2 syndromes.210 Fully 85% of these tumors occur in three organ sites: the rectum and sigmoid, appendix, and small bowel. Pancreatic carcinoids are rare and represent less than 5% of the total carcinoid tumors reported. Carcinoid tumors are of interest because they may be associated with a syndrome characterized by diarrhea and flushing and thought to be due to production of a wide variety of biologically active compounds by the tumors. A variety of clinically significant vasoactive substances and hormonally active molecules have been shown to be released by carcinoids. These include histamine, bradykinin, serotonin, and prostaglandins.211 Elevations in the biochemical markers chromogranin A and neurokinin A are useful prognostic indicators for midgut carcinoid tumors, with higher levels associated with increased tumor burden or activity.212 Carcinoids from the gastrointestinal tract that metastasize, almost always metastasize to the liver. Pancreatic carcinoids, as with other gastrointestinal carcinoids metastasizing to the liver, will almost always be associated with an elevation of urinary 5-HIAA. Symptoms due to the carcinoid syndrome, such as diarrhea, flushing, and retroperitoneal or endomyocardial fibrosis, are relatively
377
uncommon. In one series of210 patients with documented carcinoid tumors, only 14 (6.7%) had evidence of carcinoid syndrome.213 Even in patients with liver metastases, fewer than 50% will have evidence of carcinoid syndrome. With the rarity of primary pancreatic carcinoid tumors and the infrequent occurrence of carcinoid syndrome, it can be seen that a pancreatic tumor producing the carcinoid syndrome would be a distinctly rare event. The treatment of carcinoid tumors is similar to the therapy of islet cell neoplasms since patients may need pharmacologic therapy to antagonize the physiologic effects of biologically active substances produced by these tumors, along with cytoreductive therapy. Cytoreductive therapies include surgery, chemoembolization, and to some extent chemotherapy; although, well-differentiated carcinoid tumors are only modestly sensitive to chemotherapy. In the past, a variety of approaches including adrenergic blocking agents, kinin antagonists, serotonin synthesis inhibitors, histamine receptor antagonists, and peripheral serotonin antagonists have been used to treat symptoms of carcinoid syndrome.214 Frequently, mild diarrhea and/or flushing can be treated with simple symptomatic therapy. In patients with florid carcinoid syndrome or in those who develop a marked exacerbation of their carcinoid syndrome, octreotide is widely used.198,215 Investigators at the Mayo Clinic215 have reported complete amelioration of the carcinoid syndrome in 22 of 25 patients refractory to other therapeutic measures. In 18 of 25 cases, there was a greater than 50% reduction in urinary 5-HIAA. There was also some suggestion that somatostatin analog may produce objective tumor regression in some patients with metastatic carcinoid tumors. Octreotide may also delay the time to progression of carcinoid tumors and the use of octreotide LAR may prolong survival. Chemotherapy of carcinoid tumors is similar to that of islet cell tumors. Single agents including doxorubicin, 5-FU, and dacarbazine have activity in 15–25% of patients.213 5-FU plus streptozotocin has been compared to cyclophosphamide and streptozotocin in this disease by the Eastern Cooperative Oncology Group study.215 One-third of the patients receiving 5-FU plus streptozotocin and 27% of those receiving cyclophosphamide combined with streptozotocin responded. These results indicate that the chemotherapy of carcinoid syndrome is not as effective as the chemotherapy of islet cell tumors. Although α-interferon has been used to treat metastatic carcinoid,216 it is unclear how effective a strategy this is. Mayo Clinic investigators reported only a 20% tumor regression rate in a phase II study of α-interferon in carcinoid cases.217 Future clinical trials should explore combination therapy approaches of somatostatin analog, perhaps combined with chemotherapy and, potentially, interferon. It should also be emphasized that for both, islet cell tumors and carcinoid tumors, other cytotoxic agents, including taxanes, topoisomerase-1 inhibitors, and newer nucleoside analogs (e.g. gemcitabine) have not been evaluated. The receptor tyrosine kinase inhibitor SU011248 and the VEGFbinding monoclonal antibody bevacizumab were both shown in recent abstract presentations (ASCO 2005)209,218 to delay progression of carcinoid and islet cell tumors. These results are early and will require confirmation.
378
GASTROINTESTINAL TUMORS
The surgical management of patients with carcinoid tumor is of great importance since this approach may be curative. It is unlikely that a patient with a primary carcinoid of the pancreas will present to a physician at a time when the tumor is still localized. However, if such is the case, appropriate resection should be curative. In patients with metastatic carcinoid, the basic principles for surgical management are similar to those discussed with islet cell tumors. Carcinoid tumors are like islet cell neoplasms, with very indolent malignant processes. Therefore, partial cytoreduction by surgery, such as resection of hepatic tumors or the removal of intra-abdominal metastases, may give the patient excellent, relatively long-term palliation. One should consider the use of noncurative but cytoreductive surgical procedures as palliative approaches for patients with carcinoid tumors who are significantly symptomatic with either bulk disease or carcinoid syndrome not controllable by medical means. There is good evidence that carcinoid tumor patients who have resection of bulk disease may attain palliation from symptoms and be given months or, in many cases, years of remission.213 Other approaches such as hepatic artery interruption for metastatic liver disease should also be considered as palliative. In the Mayo Clinic experience, 18 of 25 patients developed significant remission of symptoms from this procedure.213 Recent experience suggests that selective hepatic arterial tumor embolization with the additive of cytotoxic chemotherapy by the embolizing material (chemoembolization) is superior to embolization alone in patients with either islet cell or carcinoid tumors metastatic to the liver.219 RFA has also been shown to result in significant tumor debulking, with improvement in patient symptoms. Long-term follow-up on this issue is awaited.192
REFERENCES 1. Jemal A, Cancer statistics, 2005. CA Cancer J Clin 2005; 55(1): 10. 2. Williamson RCN. Pancreatic cancer: the greatest oncological challenge. Br Med J 1991; 296: 445. 3. Gordis L, Gold EB. Epidemiology of pancreatic cancer. World J Surg 1984; 8: 808. 4. Buncher CR. Epidemiology of pancreatic cancer. In Moossa AR (ed) Tumors of the Pancreas. Baltimore, Maryland: Williams & Wilkins, 1980: 415 – 417. 5. Ellinger F, Landsman H. Frequency and course of cancer in diabetics. N Y State J Med 1944; 44: 259 – 65. 6. Bell ET. Carcinoma of the pancreas. Am J Pathol 1957; 33: 499 – 523. 7. Cohen GF. Early diagnosis of pancreatic neoplasms in diabetics. Lancet 1965; ii: 1267 – 74. 8. Bauer FW, Robbins SL. An autopsy study of cancer patients. JAMA 1972; 221: 1471 – 4. 9. Kessler II. Cancer mortality among diabetics. J Natl Cancer Inst 1970; 44: 673 – 86. 10. Burch GE, Ansari A. Chronic alcoholism and carcinoma of the pancreas. Arch Intern Med 1968; 122: 273 – 5. 11. Wynder EL. A case control study of cancer of the pancreas. Cancer 1973; 31: 641 – 8. 12. Warshaw AL, Fernandez-Del Castillo C. Pancreatic carcinoma. N Engl J Med 1992; 326: 455 – 65. 13. Kattwinkel J, Loepy A, DiSant’Agnese PA. Hereditary pancreatitis: three new kindreds and a critical review of the literature. Pediatrics 1973; 51: 55 – 69. 14. MacMahon B, Yen S, Trichopoulos D. Coffee and cancer of the pancreas. N Engl J Med 1981; 304: 630 – 3.
15. Tavassoli FA, Lynch RG. Occult adenocarcinoma of the pancreas in a 17-year old patient with immunosuppressed leukemia. Gastroenterology 1974; 66: 1054 – 7. 16. Jablon S, Kato H. Studies of the mortality of the A-bomb survivors, 5. Radiation dose and mortality, 1950 – 1970. Radiat Res 1972; 50: 649 – 98. 17. Matanoski GM, Seltser R, Sartwell PE. The current mortality rates of radiologists and other physicians specialists: specific causes of death. Am J Epidemiol 1975; 101: 199 – 210. 18. Hutchinson GB. Late neoplastic changes following medical irradiation. Cancer 1976; 37: 1102 – 10. 19. Court-Brown WM, Doll R. Mortality from cancer and other causes after radiotherapy for ankylosing spondylitis. Br Med J 1965; ii: 1327 – 32. 20. Levy BS, Sigurdson E, Mandel J. Investigating possible effects of asbestos in city water: surveillance of gastrointestinal cancer incidence in Duluth, Minnesota. Am J Epidemiol 1976; 103: 362 – 8. 21. Kanarek MS. Asbestos in drinking water and cancer incidence in the San Francisco Bay area. Am J Epidemiol 1980; 112: 54 – 72. 22. Selikoff IJ, Seidman H. Cancer of the pancreas among asbestos insulation workers. Cancer 1981; 47: 1469 – 73. 23. Bjelke E. Epidemiologic studies of cancer of the stomach, colon, and rectum, with special emphasis on the role of diet. Scand J Gastroenterol 1974; 9: 1 – 53. 24. Kark JD. Serum vitamin A (retinal) and cancer incidence in Evans County, Georgia. J Natl Cancer Inst 1981; 66: 7 – 16. 25. Wald N. Low serum vitamin A and subsequent risk of cancer. Lancet 1980; ii: 813 – 5. 26. Kahn HA. The Dorn study of smoking and mortality among US veterans: report on eight and one-half years of observation. Natl Cancer Inst Monogr 1966; 19: 1 – 125. 27. Hammond EC. Smoking in relation to the death rates of one million men and women. Natl Cancer Inst Monogr 1966; 19: 127 – 204. 28. Weir JM, Dunn JE. Smoking and mortality: a prospective study. Cancer 1970; 25: 105 – 12. 29. Best EWR. A Canadian Study of Smoking and Health. Ottawa, Canada: Department of National Health and Welfare, 1966. 30. Cederlof R. The Relationship of Smoking and Some Social Covariables to Mortality and Cancer Morbidity. Stockholm, Sweden: The Karolinska Institute, 1975. 31. Doll R, Peto R. Mortality in relation to smoking: 20 years observations on male British doctors. Br Med J 1976; 2: 1525 – 36. 32. Hirayama T. Prospective studies on cancer epidemiology based on census population in Japan. In Nieburgs HE (ed) Prevention and Detection of Cancer. New York: Marcel Dekker, 1978: 1139 – 1148. 33. Lyon JL, et al. Dietary intake as a risk factor for cancer of the exocrine pancreas. Cancer Epidemiol Biomarkers Prev 1993; 2: 513. 34. Lynch HT, et al. Familial pancreatic cancer: a review. Semin Oncol 1996; 23: 251 – 75. 35. Davidson P, et al. Hereditary pancreatitis: a kindred without gross aminoaciduria. Ann Intern Med 1968; 68: 88. 36. Lynch HT, et al. Pancreatic carcinoma and hereditary nonpolyposis colorectal cancer: a family study. Br J Cancer 1985; 52: 271. 37. Neumann HP, et al. Pancreatic lesions in the von Hippel – Lindau syndrome. Gastroenterology 1991; 101: 465. 38. Lynch HT, Fusaro RM. Pancreatic cancer and the familial atypical multiple mole (FAMMM) syndrome. Pancreas 1991; 6: 127. 39. Fernandez E, et al. Family history and the risk of liver, gallbladder and pancreatic cancer. Cancer Epidemiol Biomarkers Prev 1994; 3: 209. 40. Pellegata NS, et al. Detection of K-ras mutations by denaturing gradient gel electrophoresis (DGGE): a study on pancreatic cancer. Anticancer Res 1992; 12: 1731. 41. Almoguera C, et al. Most human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell 1988; 53: 549. 42. Smit VT, et al. KRAS codon 12 mutations occur very frequently in pancreatic adenocarcinomas. Nucleic Acids Res 1988; 16: 7773. 43. Masson P, Andren-Sandberg A. Crude isolation of DNA from unselected human pancreatic tissue and amplification by the polymerase chain reaction of Ki-ras oncogene to detect point mutations in pancreatic cancer. Acta Oncol 1992; 31: 421.
UNUSUAL PANCREATIC TUMORS 44. Hahn SA, et al. DPC4, a candidate tumor suppressor gene at human chromosome 18q21.1. Science 1996; 271: 350. 45. Schutte M, et al. Identification by representational difference analysis of a homozygous deletion in pancreatic carcinoma that lies within the BRCA2 region. Proc Natl Acad Sci USA 1995; 92: 5950. 46. Caldas C, et al. Frequent somatic mutations and homozygous deletions of the p16 (MTS1) gene in pancreatic adenocarcinoma. Nat Genet 1994; 8: 27. 47. Hahn SA, et al. Homozygous deletion map at 18q21.1 in pancreatic cancer. Cancer Res 1996; 56: 490. 48. Martin ST, et al. Increased prevalence of the BRCA2 polymorphic stop codon K3326X among individuals with familial pancreatic cancer. Oncogene 2005; 24(2): 3652 – 6. 49. Detailed Guide: Pancreatic Cancer 2005. American Cancer Society. 50. Cubilla AL, Fitzgerald PJ. Morphological patterns of primary nonendocrine human pancreas carcinoma. Cancer Res 1975; 35: 2234 – 48. 51. Morohoski KT, Held G, Kloppel G. Exocrine pancreatic tumors and their histological classification. A study based on 167 autopsy and 97 surgical cases. Histopathology 1983; 6: 645 – 61. 52. Ahlgren JD, Hill MC, Roberts IM. Pancreatic cancer: patterns, diagnosis, and approaches to treatment. In Ahlgren JD, Macdonald JS (eds) Gastrointestinal Oncology. Philadelphia, Pennsylvania: JB Lippincott, 1992: 197 – 207. 53. Cubilla AL, Fitzgerald PJ. Cancer of the pancreas (nonendocrine): a suggested morphologic classification. Semin Oncol 1979; 6: 285 – 97. 54. Cubilla AL, Fitzgerald PJ. Classification of pancreatic cancer (nonendocrine). Mayo Clin Proc 1979; 54: 449 – 58. 55. Cubilla AL, Fitzgerald PJ. Surgical pathology of tumors of the exocrine pancreas. In Moossa AR (ed) Tumors of the Pancreas. Baltimore, Maryland: Williams & Wilkins, 1980: 159 – 193. 56. Kloppel G. Pancreatic: non-endocrine tumors. In Kloppel G, Heitz PU (eds) Pancreatic Pathology. New York: Churchill Livingstone, 1984: 79 – 113. 57. Bastidas JA, Niederhuber JE. Pancreas. In Abeloff MD, et al. (eds) Clinical Oncology. New York: Churchill Livingstone, 1995: 1373 – 1403. 58. Moossa AR. Pancreatic cancer: approach to diagnosis, selection for surgery and choice of operation. Cancer 1982; 50: 2689 – 98. 59. Chen J, Baithun SI. Morphological study of 391 cases of exocrine pancreatic tumors with special reference to the classification of exocrine pancreatic carcinoma. J Pathol 1985; 146: 65 – 76. 60. Tracey KJ et al. Signet ring carcinoma of the pancreas, a rare variant with very high CEA values. Dig Dis Sci 1984; 29: 573 – 6. 61. Hodinson DJ, ReMine WH, Weiland LH. A clinicopathologic study of 21 cases of pancreatic cystadenocarcinoma. Ann Surg 1978; 188: 679 – 84. 62. Compagno J, Oertel JE. Mucinous cystic neoplasms of the pancreas with overt and latent malignancy (cystadenocarcinoma and cystadenoma). Am J Clin Pathol 1978; 69: 573 – 80. 63. Becker WF, Welsh RA, Pratt HS. Cystadenoma and cystadenocarcinoma of the pancreas. Ann Surg 1965; 161: 845 – 63. 64. Cullen PK, ReMine WH, Dahlin DC. A clinicopathological study of cystadenocarcinoma of the pancreas. Surg Gynecol Obstet 1963; 117: 189 – 95. 65. Setia U, Bhatia G. Pancreatic cystadenocarcinoma associated with strongyloides. Am J Med 1984; 77: 173 – 5. 66. Warren KW, Hardy KJ. Cystadenocarcinoma of the pancreas. Surg Gynecol Obstet 1968; 127: 734 – 6. 67. Ito Y, et al. Mucinous biliary obstruction associated with a cystic adenocarcinoma of the pancreas. Gastroenterology 1977; 73: 1410 – 2. 68. Mullens JE, Barr JR, Barron PT. Cystadenoma and cystadenocarcinoma of the pancreas. Can J Surg 1983; 26: 529 – 31. 69. Madura JA, et al. Mucin secreting cystic lesions of the pancreas: treatment by enucleation. Am Surg 2004; 70(2): 106. 70. Hernandez LV, et al. Role of endoscopic ultrasound (EUS) and EUSguided fine needle aspiration in the diagnosis and treatment of cystic lesions of the pancreas. Pancreas 2002; 25(3): 222. 71. Ishikawa O, et al. Adenosquamous carcinoma of the pancreas. Cancer 1980; 46: 1192 – 6. 72. Kardon DE, et al. Adenosquamous carcinoma of the pancreas: a clinicopathologic series of 25 cases. Mod Pathol 2001; 14(5): 443.
379
73. Horic A, et al. Morphogenesis of pancreatoblastoma, infantile carcinoma of the pancreas. Cancer 1977; 39: 247 – 54. 74. Cihak RW, Kawashima T, Steer A. Adenocanthoma (adenosquamous carcinoma) of the pancreas. Cancer 1971; 29: 1133 – 40. 75. Cubilla AL, Fitzgerald PJ. Morphological lesions associated with human primary invasive nonendocrine pancreas. Cancer Res 1976; 36: 2690 – 8. 76. Willozynski SP, Valente PT, Atkinson BF. Cytodiagnosis of adenosquamous carcinoma of the pancreas. Acta Cytol 1984; 28: 733 – 6. 77. Murakami Y, et al. Adenosquamous carcinoma of the pancreas: preoperative diagnosis and molecular alterations. J Gastroenterol 2003; 38(12): 1171. 78. Sprayregen S, Schoenbaum SW, Messinger NH. Angiographic features of squamous cell carcinoma of the pancreas. J Can Assoc Radiol 1975; 26: 122 – 4. 79. Leichman L, et al. Cancer of the anal canal. Am J Med 1985; 78: 211 – 5. 80. Nigyro ND, et al. Combined pre-operative radiation and chemotherapy for squamous cell carcinoma of the anal canal. Cancer 1983; 51: 1826 – 9. 81. Franklin R, et al. Combined modality therapy for esophageal squamous cell carcinoma. Cancer 1983; 51: 1062 – 71. 82. Leichman L, et al. Pre-operative chemotherapy and radiation therapy for patients with cancer of the esophagus: a potentially curative approach. J Clin Oncol 1984; 2: 75 – 9. 83. Miller JR, Baggenstoss AH, Comfort MW. Carcinoma of the pancreas. Cancer 1951; 4: 233 – 41. 84. Longecker DS, Curphey TJ. Adenocarcinoma of the pancreas in azaserine treated rats. Cancer Res 1975; 35: 2249 – 58. 85. Webb JN. Acinar cell neoplasms of the exocrine pancreas. J Clin Pathol 1977; 30: 103 – 12. 86. MacMahon HE, Brown PA, Shen EM. Acinar cell carcinoma of the pancreas with subcutaneous fat necrosis. Gastroenterology 1965; 49: 555 – 9. 87. Mullin GT, et al. Arthritis and skin lesion resembling erythema nodosum in pancreatic disease. Ann Intern Med 1968; 68: 75 – 87. 88. Burns WA, et al. Lipase-secreting acinar cell carcinoma of the pancreas with polyarthopathy. Cancer 1974; 33: 1002 – 9. 89. Horie Y, et al. Plasma carcinoembryonic antigen and acinar cell carcinoma of the pancreas. Cancer 1984; 53: 1137 – 42. 90. Ohori NP, et al. Multiple loss of heterozygosity without K-ras mutation identified by molecular analysis on fine-needle aspiration cytology specimens of acinar cell carcinoma of pancreas. Diagn Cytopathol 2002; 27(1): 42. 91. Rigaud G, et al. Allelotype of pancreatic acinar cell carcinoma. Int J Cancer 2000; 88(5): 772. 92. Abraham SC, et al. Genetic and immunohistochemical analysis of pancreatic acinar cell carcinoma: frequent allelic loss on chromosome 11p and alterations in the APC/beta-catenin pathway. Am J Pathol 2002; 160(3): 953. 93. Holen KD, et al. Clinical characteristics and outcomes from an institutional series of acinar cell carcinoma of the pancreas and related tumors. J Clin Oncol 2002; 20(24): 4673. 94. Rosai J. Carcinoma of the pancreas simulating giant cell tumor of bone. Cancer 1968; 22: 333 – 44. 95. Pour RM. Induction of unusual pancreatic neoplasms with morphologic similarity to human tumors, and evidence of their ductal/ductular origin. Cancer 1985; 55: 2411 – 6. 96. Robinson L, Damjenow I, Brezina P. Multinucleated giant cell neoplasm of pancreas. Arch Pathol Lab Med 1977; 101: 590 – 3. 97. Alguacil-Garcia A, Weiland LH. The histologic spectrum, prognosis, and histogenesis of the sarcomatoid carcinoma of the pancreas. Cancer 1977; 39: 1181 – 9. 98. Tschang TP, Garza-Garza R, Kissane JM. Pleomorphic carcinoma of the pancreas. Cancer 1977; 39: 2114 – 26. 99. Guillan RA, McMahon J. Pleomorphic adenocarcinoma of the pancreas. Am J Gastroenterol 1973; 60: 379 – 86. 100. Guillan RA. Pleomorphic adenocarcinoma of the pancreas. Cancer 1968; 21: 1072 – 9. 101. Freund U. Pleomorphic giant cell tumor of the pancreas. Isr J Med Sci 1973; 9: 84 – 8.
380
GASTROINTESTINAL TUMORS
102. Yamaguchi K, et al. Pleomorphic carcinoma of the pancreas: reappraisal of surgical resection. Am J Gastroenterol 1998; 93(7): 1151. 103. Sedivy R, et al. Osteoclast-like giant cell tumor in mucinous cystadenocarcinoma of the pancreas: an immunohistochemical and molecular analysis. Cancer Detect Prev 2005; 29(1): 8. 104. Imai Y, et al. Immunohistochemical and molecular analysis of giant cell carcinoma of the pancreas: a report of three cases. Pancreas 1999; 18(3): 308. 105. Kay S, Harrison JM. Unusual pleomorphic carcinoma of the pancreas featuring production of osteoid. Cancer 1969; 23: 1158 – 62. 106. Compagno J, Oertel JE, Kremzar M. Solid and papillar epithelial neoplasms of the pancreas, probably of small duct origin: a clinicopathologic study of 52 cases. Lab Invest 1979; 40: 248 – 9. 107. Sasaki S, et al. Differential roles of alterations of p53, p16, and SMAD4 expression in the progression of intraductal papillarymucinous tumors of the pancreas. Oncol Rep 2003; 10(1): 21. 108. Iacobuzio-Donahue C, et al. Dpc-4 protein is expressed in virtually all human intraductal papillary mucinous neoplasms of the pancreas: comparison with conventional ductal adenocarcinomas. Am J Pathol 2000; 157(3): 755. 109. Kloppel G, et al. Solid and cystic acinar cell tumor of the pancreas. Virchows Arch Pathol Anat 1981; 392: 171 – 83. 110. Colombo P, Arizzi C, Roncalli M. Acinar cell cystadenocarcinoma of the pancreas: report of rare case and review of the literature. Hum Pathol 2004; 35(12): 1568. 111. Bombi JA, et al. Papillary-cystic neoplasm of the pancreas. Cancer 1984; 54: 780 – 4. 112. Sato N, et al. Gene expression profiling identifies genes associated with invasive intraductal papillary mucinous neoplasms of the pancreas. Am J Pathol 2004; 164(3): 903. 113. Sohn TA, et al. Intraductal papillary mucinous neoplasms of the pancreas: an updated experience. Ann Surg 2004; 239(6): 788. 114. Yantiss RK, et al. KOC (K homology domain containing protein overexpression in cancer): a novel molecular marker that distinguishes between benign and malignant lesions of the pancreas. Am J Surg Pathol 2005; 29(2): 188. 115. Luttges J, et al. Pancreatic mucinous noncystic (colloid) carcinomas and intraductal papillary mucinous carcinomas are usually microsatellite stable. Mod Pathol 2003; 16(6): 537. 116. Terris B, et al. Mucin gene expression in intraductal papillarymucinous pancreatic tumours and related lesions. J Pathol 2002; 197(5): 632. 117. Terris B, et al. Characterization of gene expression profiles in intraductal papillary-mucinous tumors of the pancreas. Am J Pathol 2002; 160(5): 1745. 118. Kuo TT, Su IJ, Chien CH. Solid and papillary neoplasm of the pancreas. Cancer 1984; 54: 1469 – 74. 119. Fried P, et al. A role for radiotherapy in the treatment of solid and papillary neoplasms of the pancreas. Cancer 1985; 56: 2783 – 5. 120. D’Angelica M, et al. Intraductal papillary mucinous neoplasms of the pancreas: an analysis of clinicopathologic features and outcome. Ann Surg 2004; 239(3): 400. 121. Reyes CV, Wang T. Undifferentiated small cell carcinomas of the pancreas. Cancer 1981; 47: 2500 – 2. 122. Nakamura Y, et al. Changes to levels of serum neuron-specific enolase in a patient with small cell carcinoma of the pancreas. J Hepatobiliary Pancreat Surg 2005; 12(1): 93. 123. Fer MF, Levenson RM, Cohen MH. Extrapulmonary small cell carcinoma. In Greco FA, Oldham RK, Bunn PA (eds) Small Cell Lung Cancer. London, England: Grune & Stratton, 1981: 301 – 325. 124. Richardson RL, Weiland LH. Undifferentiated small cell carcinomas in extrapulmonary sites. Semin Oncol 1982; 9: 484 – 96. 125. Hobbs RD, et al. Hypercalcemia in small cell carcinoma of the pancreas. Cancer 1984; 53: 1552 – 4. 126. Corrin B, et al. Oat cell carcinoma of the pancreas with ectopic ACTH secretion. Cancer 1973; 31: 1523 – 7. 127. Ibrahim NBN, Briggs JC, Corbindhley CM. Extrapulmonary oat cell carcinoma. Cancer 1984; 54: 1645 – 61. 128. Buchino JJ, Castello FM, Nagaraj HS. Pancreatoblastoma. Cancer 1984; 53: 963 – 9.
129. Taxy JB. Adenocarcinoma of the pancreas in childhood. Cancer 1976; 37: 1508 – 18. 130. Hamoudi AB, et al. Papillary epithelial neoplasm of pancreas in a child. Cancer 1970; 26: 1126 – 34. 131. Frable WJ, Still WJS, Kay S. Carcinoma of the pancreas. Infantile type. Cancer 1971; 27: 667 – 73. 132. Grosfeld JL, Clatworthy HW, Hamoudi AB. Pancreatic malignancy in children. Arch Surg 1970; 101: 370 – 5. 133. Cao D, et al. Expression of novel markers of pancreatic ductal adenocarcinoma in pancreatic nonductal neoplasms: additional evidence of different genetic pathways. Mod Pathol 18(6): 2005; 752 – 61. 134. Huntrakoon M. Oncocytic carcinoma of the pancreas. Cancer 1983; 51: 332 – 6. 135. Patel SA, et al. Genetic analysis of invasive carcinoma arising in intraductal oncocytic papillary neoplasm of the pancreas. Am J Surg Pathol 2002; 26(8): 1071. 136. Gray GM, et al. Lymphomas involving the gastrointestinal tract. Gastroenterology 1982; 82: 152 – 3. 137. Rosenfelt F, Rosenberg SA. Diffuse histiocytic lymphoma presenting with gastrointestinal tract lesions. Cancer 1980; 45: 2188 – 93. 138. Arcari A, et al. Primary pancreatic lymphoma. Report of five cases. Haematologica 2005; 90(2): ECR09. 139. Boni L, et al. Primary pancreatic lymphoma. Surg Endosc 2002; 16(7): 1107. 140. Ackerman NB, et al. Problems in differentiating between pancreatic lymphoma and anaplastic carcinoma and their management. Ann Surg 1976; 184: 705 – 8. 141. Neibling HA. Primary sarcoma of the pancreas. Am Surg 1968; 34: 690 – 3. 142. Brooke WS, Maxwell JG. Primary sarcoma of the pancreas. Am J Surg 1966; 112: 657 – 61. 143. Schron DS, Mendelsohn G. Pancreatic carcinoma with duct, endocrine, an acinar differentiation. Cancer 1984; 54: 1766 – 70. 144. Reid JD, et al. Ductuloinsular tumors of the pancreas. Cancer 1982; 49: 908 – 15. 145. Andrew A. An experimental investigation into the possible neural crest origin of pancreatic APUD (islet cells). J Embryol Exp Morphol 1976; 35: 577 – 93. 146. Fontaine J, DeDovarin NM. Analysis of endoderm formation in avian blastoderm by the use of quail chick chimeras: problem of the neuroecto-origin of the cells of APUD series. J Embryol Exp Morphol 1977; 41: 209 – 22. 147. Nicolesco S, et al. Morphological significance of association of epithelial cords and duct-like structures in certain tumors of the pancreatic islets. Endocrinology 1978; 16: 227 – 30. 148. Tang SJ, et al. Telomerase activity in pancreatic endocrine tumors. Am J Gastroenterol 2002; 97(4): 1022. 149. Bartsch D, et al. Mutations of the DPC4/Smad4 gene in neuroendocrine pancreatic tumors. Oncogene 1999; 18(14): 2367. 150. Haller DG. Carcinoid and islet cell tumors of the gastrointestinal tract. In Ahlgren JD, Macdonald JS (eds) Gastrointestinal Oncology. Philadelphia, Pennsylvania: JB Lippincott, 1992: 449 – 460. 151. Moldow RE, Connelly RR. Epidemiology of pancreatic cancer in Connecticut. Gastroenterology 1978; 55: 667 – 86. 152. Grimelius L. A silver nitrate stain for alpha 2 cells in human pancreatic islets. Acta Soc Med Ups 1968; 73: 243. 153. D’Alessandro M, Mariani P, Lomanto D. Serum neuron-specific enolase in diagnosis and follow-up of gastrointestinal neuroendocrine tumors. Tumour Biol 1992; 13: 352. 154. Eriksson B, Oberg K. PPomas and nonfunctioning endocrine pancreatic tumors: clinical presentation, diagnosis, and advances in management. In Mignon M, Jensen RT (eds) Endocrine Tumors of the Pancreas: Recent Advances in Research and Management. Basel, Switzerland: Karger, 1995: 208. 155. Wermer P. Genetic aspects of adenomatosis of endocrine glands. Am J Med 1954; 16: 363 – 71. 156. Sipple JH. The association of pheochromocytoma with carcinoma of the thyroid gland. Am J Med 1961; 31: 163 – 8. 157. Caronna R, et al. Surgical management of pancreatic endocrine tumors in patients with MEN 1 syndrome. Considerations on one case observed. Ann Ital Chir 2004; 75(3): 369.
UNUSUAL PANCREATIC TUMORS 158. Radford DM, et al. Loss of heterozygosity of markers on chromosome 11 in tumors with patients with multiple endocrine neoplasia syndrome. J Cancer Res 1991; 50: 1154. 159. Wu J, et al. The genetic defect in multiple endocrine neoplasia type 2A maps next to the centromere of chromosome 10. Am J Hum Genet 1990; 46: 624 – 30. 160. Jensen RT. Zollinger – Ellison syndrome: current concepts and management. Ann Intern Med 1983; 98: 5 – 78. 161. Roy PK, et al. Zollinger-Ellison syndrome. Clinical presentation in 261 patients. Medicine 2000; 79(6): 379. 162. Regan PT, Malagelada JR. A reappraisal of clinical, roentgenographic and endoscopic features of the Zollinger – Ellison syndrome. Mayo Clin Proc 1978; 53: 19 – 23. 163. McCarthy DM, Jensen RT. Zollinger – Ellison syndrome – current issues. In Cohen S, Soloway RD (eds) Hormone-Producing Tumors of the Gastrointestinal Tract. New York: Churchill Livingstone, 1985: 25 – 55. 164. Honda M, Ishibashi M. The diagnosis and treatment of insulinoma and gastrinoma. Gan To Kagaku Ryoho 2004; 31(3): 337. 165. Mignon M. Diagnostic and therapeutic strategies in Zollinger-Ellison syndrome associated with multiple endocrine neoplasia type I (MEN1): experience of the Zollinger-Ellison Syndrome Research Group: Bichat 1958 – 1999. Bull Acad Natl Med 2003; 187(7): 1249. 166. Chen YJ, et al. Loss of heterozygosity of chromosome 1q in gastrinomas: occurrence and prognostic significance. Cancer Res 2003; 63(4): 817. 167. Brennan MF, Macdonald JS. Cancer of the endocrine system. In DeVita VT, Hellman S, Rosenberg SH (eds) Cancer: Principles and Practice of Oncology. Philadelphia, Pennsylvania: JB Lippincott, 1985: 1179 – 1241. 168. Stefanini P, Carboni M, Petrassi N. Beta islet cell tumors of the pancreas. Results of a study on 1067 cases. Surgery 1974; 75: 597 – 609. 169. Vasquez QE. The surgical management of insulinoma. Bol Asoc Med P R 2004; 96(1): 33. 170. Dyaczynski M, et al. Insulinoma exclusively localized during intraoperative ultrasonography. Wiad Lek 2004; 57(7 – 8): 385. 171. Zhao Y, et al. Pancreatic insulinomas: experience in 220 patients. Zhonghua Wai Ke Za Zhi 2000; 38(1): 10. 172. Lee CH, et al. A large malignant insulinoma: case report with endosonographic, immunohistochemical and ultrastructural features. Korean J Intern Med 2003; 18(1): 45. 173. Feng LS, et al. Diagnosis and treatment of insulinoma: report of 105 cases. Hepatobiliary Pancreat Dis Int 2002; 1(1): 137. 174. de Herder WW. Insulinoma. Neuroendocrinology 2004; 80(1): 20. 175. Reynolds LR, et al. Combined use of calcium infusion localization and a minimally invasive surgical procedure in the management of insulinoma. Endocr Pract 2002; 8(5): 329. 176. Becker WS, Kahn D, Rothman S. Cutaneous manifestations of internal malignant tumors. Arch Derm Syphilol 1972; 45: 1069. 177. Du Jardin P, Cools P, Van der Stighelen Y. Necrolytic migratory erythema: first symptom of a glucagonoma. A case report. Acta Chir Belg 2004; 104(4): 468. 178. Malinson CN, Bloom SR, Warin AP. A glucagonoma syndrome. Lancet 1974; ii: 1 – 5. 179. Holst JJ. Glucagon-producing tumors. In Cohen S, Soloway RD (eds) Hormone-Producing Tumors of the Gastrointestinal Tract. New York: Churchill Livingstone, 1985: 57 – 84. 180. Leichter SB. Clinical and metabolic aspects of glucagonoma. Medicine 1980; 59: 100. 181. Ganda OP, Weir GC, Soeldner JS. Somatostatinoma: a somatostatincontaining tumor of the endocrine pancreas. N Engl J Med 1977; 296: 963 – 98. 182. Boden G, Shimoyama R. Hormone-producing tumors of the gastrointestinal tract. In Cohen S, Soloway RD (eds) HormoneProducing Tumors of the Gastrointestinal Tract. New York: Churchill Livingstone, 1985: 85 – 99. 183. Kane MG, O’Dorisio TM, Krejs GL. Intravenous VIP infusion causes secretory diarrhea in man. N Engl J Med 1983; 309: 1501 – 6. 184. House MG, Yeo CJ, Schulick RD. Periampullary pancreatic somatostatinoma. Ann Surg Oncol 2002; 9(9): 869.
381
185. Peng SY, et al. Diagnosis and treatment of VIPoma in China: (case report and 31 cases review) diagnosis and treatment of VIPoma. Pancreas 2004; 28(1): 93. 186. Guo KJ, et al. Surgical treatment of nonfunctioning islet cell tumor: report of 41 cases. Hepatobiliary Pancreat Dis Int 2004; 3(3): 469. 187. Glasser B, Vinik AL. Clinical findings in patients with malignant tumors secreting pancreatic polypeptide (PP). Clin Res 1979; 27: 627a. 188. Friesen SR, Kimmel JR, Tomita T. Pancreatic polypeptide as a screening marker for pancreatic polypeptide apudomas in multiple endocrinopathies. Am J Surg 1980; 139: 61 – 72. 189. Wang L, et al. Diagnosis and treatment of malignant pancreatic endocrine tumour. Chin Med Sci J 2004; 19(2): 130. 190. Clouse ME, Lee RGL, Duszlak EJ. Hepatic artery embolization for metastatic endocrine-secreting tumors of the pancreas. Gastroenterology 1983; 85: 1183 – 6. 191. Tochner ZA, Kinsella TJ, Glatstein E. Hepatic irradiation in the management of metastatic hormone-secreting tumors. Cancer 1985; 56: 20 – 4. 192. Atwell TD, et al. Treatment of neuroendocrine cancer metastatic to the liver: the role of ablative techniques. Cardiovasc Intervent Radiol 2005; 28(4): 409 – 21. 193. Yao KA, et al. Indications and results of liver resection and hepatic chemoembolization for metastatic gastrointestinal neuroendocrine tumors. Surgery 2001; 130(4): 677. 194. Norton JA, et al. Morbidity and mortality of aggressive resection in patients with advanced neuroendocrine tumors. Arch Surg 2003; 138(8): 859. 195. Kvols L, Schutt A, Buck M. Treatment of metastatic islet cell carcinomas with a long acting somatostatin analogue. Proc Am Soc Clin Oncol 1986; 5: 85. 196. Evers BM, et al. Somatostatin and analogues in the treatment of cancer. Ann Surg 1991; 213: 190 – 8. 197. Tomassetti P, et al. Treatment of gastroenteropancreatic neuroendocrine tumours with octreotide LAR. Aliment Pharmacol Ther 2000; 14(5): 557. 198. Maton PN. The use of the long-acting somatostatin analogue, octreotide in patients with islet cell tumors. Gastroenterol Clin North Am 1989; 18: 897. 199. Anthony LB, Kang T, Shyr Y. Malignant carcinoid syndrome: survival I the octreotide LAR era. Am Soc Clin Oncol 2005; 23(16S): 328S (Abstract #4084). 200. McGuigan JE, Wolfe MM. Secretin injection test in the diagnosis of gastrinoma. Gastroenterology 1980; 789: 1324. 201. Bleecker JJ, Chowdhury F, Goldner MG. Thiazide therapy in hypoglycemia of metastatic insulinoma. Clin Res 1964; 12: 456. 202. Mateu MV, et al. Treatment of insulinoma with diazoxide. Medicina 2003; 63(1): 51. 203. Fjallskog ML, et al. Treatment of malignant endocrine pancreatic tumors with a combination of alpha-interferon and somatostatin analogs. Med Oncol 2002; 19(1): 35. 204. Murray-Lyon IM, Eddleston ALWF, Williams R. Treatment of multiple hormone producing malignant islet cell tumors with streptozotocin. Lancet 1968; ii: 895. 205. Broder LE, Carter SK. Results of therapy with streptozotocin in 52 patients. Ann Intern Med 1973; 79: 108 – 18. 206. Moertel CG, Hanley JA, Johnson LA. Streptozotocin alone compared with streptozotocin plus fluorouracil in the treatment of advanced islet cell carcinoma. N Engl J Med 1980; 303: 1189. 207. Schott M, Scherbaum WA, Feldkamp J. Drug therapy of endocrine neoplasms. Part II: malignant gastrinomas, insulinomas, glucagonomas, carcinoids and other tumors. Med Klin 2000; 95(2): 81. 208. Moertel CG, et al. Streptozotochin-doxorubicin, streptozotocinfluorouracil or chlorozotocin in the treatment of advanced islet cell carcinoma. N Engl J Med 1992; 326: 519 – 23. 209. Kulke M, et al. A phase II study to evaluate the efficacy and safety of SU11248 in patients with unresectable neuroendocrine tumors. Am Soc Clin Oncol 2005; 23(16S): 310S (Abstract #4008). 210. O’Dorisio TM, Vinik AL. Pancreatic polypeptide and mixed polypeptide-producing tumors of the gastrointestinal tract. In Cohen S, Soloway RD (eds) Hormone-Producing Tumors of the Gastrointestinal Tract. New York: Churchill Livingstone, 1985: 117 – 128.
382
GASTROINTESTINAL TUMORS
211. Moertel CG. Treatment of the carcinoid and the malignant carcinoid syndrome. J Clin Oncol 1983; 1: 727 – 40. 212. Ardill JE, Erikkson B. The importance of the measurement of circulating markers in patients with neuroendocrine tumours of the pancreas and gut. Endocr Relat Cancer 2003; 10(4): 459. 213. Moertel CG. Small intestine. In Holland JF, Free E (eds) Cancer Medicine. Philadelphia, Pennsylvania: Lea & Febiger, 1973: 1574 – 1584. 214. Kvols LK, et al. Treatment of the malignant carcinoid syndrome: evaluations of a long-acting somato-statin analogue. N Engl J Med 1986; 315: 663 – 704. 215. Moertel CG, Hanley JA. Combination chemotherapy trials in metastatic carcinoid tumor and the malignant carcinoid syndrome. Clin Cancer Trials 1979; 2: 327 – 34.
216. Oberg K, Norheim I, Alm G. Treatment of malignant carcinoid tumors: a randomized controlled study of streptozotocin plus 5-FU and human leukocyte interferon. Eur J Cancer Clin Oncol 1989; 25: 1475 – 9. 217. Moertel CG, Rubin J, Kvols LK. Therapy of metastatic carcinoid tumor and the malignant carcinoid syndrome with recombinant leukocyte A interferon. J Clin Oncol 1989; 7: 865. 218. Yao JCm, et al. Improved progression-free survival and rapid, sustained decrease in tumor perfusion among patients with advanced carcinoid treated with bevacizumab. Am Soc Clin Oncol 2005; 23(16S): 309S (Abstract #4007). 219. Moertel CG, Johnson CM, McKusick MA. The management of patients with advanced carcinoid tumors and islet cell carcinomas. Ann Intern Med 1994; 120: 302.
Section 6 : Gastrointestinal Tumors
33
Uncommon Hepatobiliary Tumors Steven J. Cohen and Natalie E. Joseph
INTRODUCTION Many tumors of the hepatobiliary system are uncommon. Cure is possible for select patients, and evaluation in a multidisciplinary setting is essential. As systemic therapy has a minimal impact on outcome, enrollment in clinical trials is preferred. This chapter summarizes management of tumors of the gallbladder, bile ducts, and less common primary hepatic lesions.
GALLBLADDER TUMORS Epidemiology/Risk Factors Approximately 7500 cases of gallbladder and extrahepatic biliary cancer are expected in the United States in 2005.1 Gallbladder cancer is more prevalent in Native Americans and South America.2 Incidence increases with age, with a female : male ratio of 3 : 1. The most closely linked risk factor is gallstones. A relationship between gallstone size and risk has been reported in some3 but not all studies.4 However, only a small percentage of patients with gallstones develop cancer.5 Other risk factors include Salmonella infection, choledochal cysts, calcified gallbladder, and anomalous pancreatic duct architecture.
Pathology Nearly all gallbladder cancers are carcinomas, with 90% adenocarcinomas.6 A sequence of histopathologic changes from chronic inflammation to invasive carcinoma with a metaplasia –dysplasia –carcinoma sequence has been described.7 Common mutations in gallbladder cancer include p53, K-ras, and regulators of the cell cycle. Less common subtypes include papillary, tubular, mucinous, and signet ring. Other carcinomas, including squamous cell, adenosquamous, anaplastic (undifferentiated), clear cell, and small cell make up only a small percentage. Noncarcinoma histologies, such as sarcoma (and carcinosarcomas), carcinoid, lymphoma, and melanoma are extremely rare.
Clinical Presentation/Workup The most common symptoms of gallbladder cancer are right upper quadrant abdominal pain (75%), weight loss, nausea/vomiting (50%), and anorexia.8 An incidental finding at surgery for benign disease is rare. Transabdominal ultrasonography is a good initial study, although computed tomography (CT) scan is superior to detect local liver extension or metastases.9 For patients with distant metastases, biopsy confirms the diagnosis. For resectable patients, definitive diagnosis is obtained surgically. CA 19-9 lacks specificity for diagnosis. CT scan of the abdomen should be obtained, as involvement of liver, regional lymph nodes, and abdominal cavity is common.10
Treatment Considerations – Resectable Disease Surgical resection is the only treatment modality associated with long-term survival, and prognosis is closely linked to stage (Table 1). For early stage disease, surgical resection can be curative. However, most patients present with locally advanced or distant disease. The updated staging system simplifies T3 (potentially resectable) and T4 (unresectable) disease. Approximately 1–2% of cholecystectomies for benign disease will harbor a malignancy.5 For T1a lesions, no additional surgery is recommended, as lymph node metastases are rare. For T1b lesions radical cholecystectomy is controversial, with 15% harboring lymph node metastases.11 For patients with T2 tumors, more extensive resection is recommended, as lymph node metastases are noted in 62% of patients and 5-year survival drops considerably to 30–40%.12,13 This involves an extended cholecystectomy which includes resection of the gallbladder bed with a margin of normal liver and regional lymphadenectomy.14 For T3/T4 tumors or extensive lymph node metastases, 5-year survivors are rare. Radical surgical resection has resulted in occasional cures, and extent of surgery must be evaluated with each individual patient. For laparoscopic cholecystectomies, subsequent resection of the port site to prevent tumor seeding is important.
Adjuvant Therapy No large randomized studies of adjuvant therapy for resected gallbladder cancer have been reported. Historically, local
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
384
GASTROINTESTINAL TUMORS
Table 1 Staging of gallbladder cancer.15
T-stage T0 Tis T1a T1b T2 T3 T4 N-stage N0 N1 M-stage M0 M1 Stage grouping Stage 0 Stage IA Stage IB Stage IIA Stage IIB Stage III Stage IV
No evidence of primary Carcinoma in situ Tumor invades lamina propria Tumor invades muscle layer Tumor invades perimuscular connective tissue Tumor perforates the serosa and/or directly invades liver and/or other adjacent organ Tumor invades main portal vein or hepatic artery or multiple extrahepatic organs No regional lymph node metastases Regional lymph node metastases No distant metastases Distant metastases Tis T1 T2 T3 T1/2/3 T4 Any T
N0 N0 N0 N0 N1 Any N Any N
M0 M0 M0 M0 M0 M0 M1
Reproduced from reference 15. 2002 Springer-Verlag.
recurrences were thought to affect the majority of patients.16 Single center series of adjuvant external beam radiation therapy with 5-fluorouracil (5-FU) suggested improved local control and survival when compared to historical controls.17,18 However, a recent series demonstrates that initial distant recurrence occurs in 85% of patients.19 Thus, systemic failure is an important concern. We offer external beam radiation therapy with continuous infusion 5-FU to patients with >T1 tumors as per National Comprehensive Cancer Network (NCCN) guidelines.20 Systemic chemotherapy after chemoradiation could be considered. A phase III study of patients with a variety of pancreaticobiliary cancers randomized patients to receive mitomycin-C and 5-FU or no therapy after resection.21 For 112 patients with gallbladder cancer, 5-year survival was significantly better for the chemotherapy arm (26.0% vs 14.4%, p = 0.037). However, this difference was predominantly in patients undergoing palliative operations. Thus, consideration can be given to a 5-FU or gemcitabine-based regimen after chemoradiation to complete a 6-month course of therapy.
Advanced Disease For patients with locally advanced gallbladder cancer, external beam radiotherapy with continuous infusion 5-FU is appropriate. There is limited data to support “downstaging” of tumors, although we reconsider resection after chemoradiation. Patients with noncontiguous liver metastases or distant disease are not resection candidates and palliative chemotherapy is offered. Fluoropyrimidine-based regimens have been commonly utilized. However, impact on survival has not been shown, and combination regimens are not clearly better than monotherapy.22 Other chemotherapy agents with activity include platinum agents,23,24 gemcitabine,25,26 irinotecan,27 taxanes,28 and the topoisomerase-I inhibitor DX-8951,29 but impact on survival is questionable (Table 2).
Table 2 Selected reports of chemotherapy for advanced gallbladder/bile duct cancer.
Author
N (GB/BD) Chemotherapy RR
Falkson22
53 (GB) 34 (BD)
Doval23
30 (GB)
Glover24
26 (total)
Gallardo25 Knox26
26 22 23 23 13 26 17 18 21
Alberts27 Kuhn28 Abou-Alfa29
(GB) (GB) (BD) (GB) (BD) (GB) (BD) (GB) (BD)
Oral 5-FU ± streptozotocin ± methyl CCNU Cisplatin, gemcitabine Capecitabine, oxaliplatin Gemcitabine Gemcitabine, capecitabine Irinotecan Gemcitabine, docetaxel DX-8951
Median survival
11% (GB) 8% (BD)
(5-FU alone) 21 weeks (GB) 26 weeks (BD)
37%
20 weeks
27%
Not reported
36% 31%
30 weeks 14 months
8%
6 months
9%
11 months
5%
7.8 months
GB, gallbladder; BD, bile duct; RR, response rate; 5-FU, 5-fluorouracil.
One must be cautious in interpreting these data regarding the activity of systemic agents, given the small size of these studies. All patients should be offered a clinical trial if available. Otherwise, fluoropyrimidines or gemcitabine are most commonly used.
Uncommon Histologies Squamous cell or adenosquamous cell gallbladder carcinomas account for 5–10% of gallbladder malignancies.30 Squamous cells may develop from metaplastic squamous epithelium or squamous differentiation of adenocarcinoma. They are more common in women and present with bulky local disease and direct liver invasion.31 Thus, resectability is lower and survival worse than for adenocarcinomas. For resectable tumors with negative margins, long-term survival has been demonstrated.32 Radiation therapy with 5-FU should be considered for locally unresectable disease or adjuvant therapy. If a clinical trial is unavailable, a fluoropyrimidine and platinum combination is reasonable. Papillary carcinomas, a subtype of adenocarcinomas, comprise 5% of gallbladder cancer and have an improved prognosis, attributed to exophytic growth, delayed invasion, and earlier obstructive symptoms.33 Noninvasive papillary carcinomas are cured with resection, but invasive tumors have as poor a prognosis as adenocarcinomas (52% 5-year survival if organ confined, <10% 5-year survival with lymph node metastases). Less common carcinomas include mucinous and the signet ring subtype,34 and clear cell carcinomas which must be distinguished from metastatic renal cell carcinoma by immunohistochemistry.35 Surgical resection should be pursued if appropriate. For advanced disease, a trial of chemotherapy utilized for adenocarcinoma is reasonable. Small cell carcinomas of the gallbladder are extremely rare.36 Most present with unresectable, metastatic disease, and chemotherapy similar to that used for pulmonary small cell tumors is warranted. For localized disease, platinum and etoposide with radiation therapy is reasonable.
UNCOMMON HEPATOBILIARY TUMORS
Nonepithelial tumors of the gallbladder are even less common. Less than 50 carcinoid tumors of the gallbladder have been reported.37 Most are localized,38 due to early symptoms such as jaundice and pain, and 5-year survival is greater than adenocarcinomas (61%). Adjuvant therapy is of unproven value and treatment of distant disease follows that for any primary site. (Systemic therapy of carcinoid tumors is considered in detail in Chapter 34; Chapter 35 and Chapter 36.) Less than 30 carcinosarcomas (mixed carcinomas and sarcomas) of the gallbladder have been reported.39 Most present with right upper quadrant pain and jaundice and locally advanced or metastatic disease. They are clinically aggressive and almost all patients die within a year. Effectiveness of chemotherapy or radiotherapy is unknown. Approximately 100 primary sarcomas of the gallbladder have been reported, representing a range of histologies.40 Most have extensive local invasion and/or metastases. Surgical resection, if possible, is appropriate for both sarcomas and carcinosarcomas. For unresectable or metastatic disease, chemotherapy regimens commonly utilized for soft tissue sarcomas would be reasonable, although data are lacking for this site. Finally, less than 20 cases of primary gallbladder melanoma have been reported. Most patients are in their 40s or 50s. Surgical excision is appropriate in the absence of metastases. Adjuvant immunotherapy might be considered. Metastatic melanoma to gallbladder is more common with most discovered at autopsy.41
BILE DUCT TUMORS Epidemiology/Risk Factors Cholangiocarcinoma is the most common bile duct tumor, with approximately 5000 new cases yearly in the United States.42 Two-thirds are extrahepatic (involving the right and left hepatic ducts, common hepatic, and common bile ducts), and one-third intrahepatic.42 Most present in the sixth and seventh decades of life, with an increased incidence in Asia. Primary sclerosing cholangitis is the most closely linked risk factor, with a 6–30% lifetime risk.43 Pathogenesis relates to chronic inflammation. Colonization with the liver flukes Opisthorchis viverrini and Clonorchis sinensis (from undercooked freshwater fish) is strongly associated. Possible mechanisms of carcinogenesis include chronic irritation, nitric oxide formation, and activation of drug-metabolizing enzymes.44 Other risk factors include choledochal cysts (10–30% risk),45 hepatolithiasis (particularly for intrahepatic cholangiocarcinomas), and exposure to Thorotrast.42 More recently, hepatitis C has been linked.46
385
DPC4, p16, and p53 are often inactivated.47 Among oncogenes, K-ras is the most commonly mutated,48 particularly in hilar cholangiocarcinomas (75%). Perturbations in signal transduction and adhesion pathways, including Her-2/neu, c-Met, COX-2, and MUC1, have been described.49
Clinical Presentation/Workup Extrahepatic bile duct tumors consist of hilar (Klatzkin) tumors (60%), middle, and distal tumors. Patients with extrahepatic cholangiocarcinoma often present with obstructive jaundice while intrahepatic lesions frequently present with constitutional symptoms, loss of appetite, and malaise, or a mass found on routine imaging.42 Workup includes an ultrasound, CT scan, and/or MRCP (magnetic resonance cholangiopancreatography). CA 19-9 is elevated in most patients with cholangiocarcinoma.50 Imaging studies reveal a discrete mass for intrahepatic cholangiocarcinomas, but often only ductal dilatation for extrahepatic lesions. Thus, painless jaundice may be confused with pancreatic cancer. Pancreatic lesions cause extrahepatic ductal dilatation while extrahepatic cholangiocarcinomas cause only intrahepatic ductal dilatation. Histological confirmation of malignancy is important prior to surgical intervention. Cytology yield with endoscopic retrograde cholangiopancreatography (ERCP) is less than 50%.51 Endoscopic ultrasound may have increased yield in patients with negative brush cytology from ERCP.52 An example of a diagnostic ERCP is shown in Figure 1. For intrahepatic tumors, the differential diagnosis includes metastatic tumor. For a solitary intrahepatic adenocarcinoma, a physical exam, CT scan of chest, abdomen, and pelvis, mammography (women), and upper and lower endoscopy are reasonable. In the absence of distant disease, a solitary liver lesion with pathology consistent with a cholangiocarcinoma should be treated with curative surgery.
Pathology Nearly all bile duct tumors are carcinomas, with 92% adenocarcinomas.6 The adenoma –carcinoma sequence is rarely seen. Similar to gallbladder adenocarcinomas, subtypes include papillary, mucinous, clear cell, and signet ring carcinomas. Surrounding desmoplasia is common and confounds cytologic diagnosis. Small cell tumors are exceedingly rare. Noncarcinomas include lymphomas, sarcomas, and carcinoid tumors. Tumor suppressor genes such as
Figure 1 ERCP image of a malignant extrahepatic biliary stricture. Brushings for cytology and subsequent surgical pathology specimen confirmed adenocarcinoma. (Image courtesy of Jeffrey Tokar, M.D., Department of Medical Oncology, Fox Chase Cancer Center.)
386
GASTROINTESTINAL TUMORS
Treatment Considerations – Resectable Disease Intrahepatic cholangiocarcinomas typically present as large masses. Contraindications to surgical resection include distant disease, bilobar liver involvement, or hepatic artery or portal vein invasion. For resected patients (approximately one-third), 5-year survival is 25–35%53 Adverse features include larger tumor size, vascular invasion, nodal involvement, multiple tumors, and positive margin.54 No long-term survivors in one series had positive lymph nodes.55 Liver transplantation is not routinely performed. Surgery for extrahepatic bile duct tumors at the biliary hilum involves partial hepatectomy. Determination of resectability includes extent of tumor (hepatic duct involvement), vascular invasion (encasement of portal vein proximal to bifurcation), hepatic lobe atrophy, and presence of metastatic disease. Patients with positive margins do particularly poorly.56 Resection of distal bile duct tumors (between the duodenum and ampulla of Vater, 20–30%) involves pancreaticoduodenectomy. These tumors are more often resectable than pancreatic cancer, with less frequent positive margins and nodal involvement, and 5-year survival of up to 40%. The most critical determinant of outcome is lymph node status.57
Adjuvant/Neoadjuvant Therapy Rationale exists to consider adjuvant therapy of biliary tumors. Despite surgical resection, the majority of patients have recurrence, indicating the presence of microscopic metastatic disease. Adjuvant therapy is of benefit for more common regional tumors such as pancreas and gastric cancer.58 – 60 Finally, many recurrences are loco-regional,19 suggesting a role for radiation therapy. A 500 patient study of hepatobiliary tumors randomized 139 patients, with cholangiocarcinoma after palliative or curative resection, to observation or mitomycin-C and 5-FU.21 Chemotherapy did not improve overall survival. Other data consist of case series utilizing 5-FU-based regimens.61 Retrospective reviews suggest improved local control and survival with postoperative radiotherapy alone or with 5-FU.62 – 64 NCCN guidelines offer the option of adjuvant radiation therapy with 5-FU chemotherapy for resected extrahepatic (but not intrahepatic) cholangiocarcinomas.20 The use of subsequent systemic chemotherapy is of unknown benefit. Extrapolating from adjuvant gastric and pancreatic cancer, 4 months of 5-FU or gemcitabine-based chemotherapy could be considered.
Locally Advanced/Metastatic Disease Most clinical trials of chemotherapy for advanced biliary cancers have included gallbladder cancer (Table 2). Singleagent response rates are 5–35%, without clear survival impact. Large, randomized studies have not been reported. If a clinical trial is unavailable, we typically utilize a fluoropyrimidine or gemcitabine-based regimen. For localized, unresectable extrahepatic cholangiocarcinoma, we use concurrent 5-FU with external beam radiation therapy. Data is primarily retrospective. In one series, despite
therapy with radiation alone or with 5-FU, the first site of disease progression was local in nearly three-fourth of patients and a median survival of only 10 months.65 A second series treated with concurrent chemoradiation noted a similar survival of 10.7 months for unresectable patients.66 Of nine patients undergoing posttreatment surgery, all had negative margins and three, a complete pathologic response. In our practice, patients with resectable localized cholangiocarcinomas proceed to surgical resection. If unresectable, concurrent 5-FU with radiation therapy is reasonable with reevaluation for posttherapy resection. Extrapolation is made from pancreas cancer, in which many67 but not all68 studies support a benefit for combination therapy compared to either modality alone.
Supportive Care The leading causes of death from cholangiocarcinoma are liver failure and infection. For patients with fever and jaundice, immediate antibiotics and prompt relief of obstruction is critical. ERCP with mechanical stenting is cost effective, easier, and comparable in efficacy to surgical bypass.69 Photodynamic therapy (PDT) may improve outcome for patients refractory to traditional stenting.70 Most patients have successful relief of biliary obstruction and infection endoscopically.
Uncommon Histologies Epithelial variants of adenocarcinoma were described for gallbladder cancers and are approached similarly. Papillary carcinomas comprise up to 4% of bile duct epithelial tumors. A common presentation is obstructive jaundice, abdominal pain, and weight loss. They have a more indolent course than infiltrating adenocarcinomas with an intraductal growth pattern71 and excellent prognosis after resection. Chemotherapy after surgical resection is unlikely to be of benefit. Papillary carcinomas usually lack the multifocal nature and extracellular mucin production of pancreatic intraductal papillary neoplasms.72 Squamous cell carcinomas of the biliary tract typically involve the intrahepatic biliary tree.73 Many are associated with liver fluke infestation, intrahepatic lithiasis, and primary sclerosing cholangitis. Surgical resection can be curative, although most are locally advanced. Mixed adenosquamous carcinomas comprise approximately 5% of extrahepatic bile duct tumors. After surgical resection 5-year survival is only 16%, with pancreatic invasion and extensive lymph node metastases being the negative prognostic features.74 The role of adjuvant chemotherapy is unknown. For metastatic disease, therapy with fluoropyrimidine and platinum is reasonable. Only a handful of small cell cancers of the biliary tract have been reported. A combination of platinum-based chemotherapy and local radiation therapy for extrahepatic tumors is appropriate. For metastatic disease, chemotherapy alone is recommended. Finally, there are few reports of undifferentiated tumors, comprised of spindle-shaped and pleomorphic tumor cells resembling sarcoma.75 Approximately 30 bile duct carcinoid tumors have been reported, presenting largely with jaundice in younger women
UNCOMMON HEPATOBILIARY TUMORS
(mean age 47 years) in the common bile duct (58%).76 Prognosis is excellent for patients undergoing complete resection. Adjuvant therapy is not recommended. Treatment for metastatic disease follows general guidelines for carcinoids at other sites. (Systemic therapy of carcinoid tumors is considered in detail in Chapter 34, Chapter 35, and Chapter 36.) Less than 20 cases of primary non-Hodgkin’s lymphoma (NHL) of the bile duct have been reported. More common is systemic dissemination to bile ducts or liver. NHL, mimicking cholangiocarcinoma, presents with weight loss and fever, usually without a prominent mass lesion.77 Most are of B cell origin. Therapy involves anthracycline-based chemotherapy. Finally, sarcomas and mixed carcinosarcomas of the biliary tract have been described. Carcinosarcomas are typically polypoid lesions which may cause early symptoms with potential for curative resection.78
HEPATIC TUMORS Hepatocellular carcinoma (HCC) is one of the most common cancers worldwide, with up to one million cases annually.79 Uncommon primary hepatic tumors include fibrolamellar carcinoma (FC), mesenchymal tumors, and hepatoblastoma, and will be considered here.
Fibrolamellar Carcinoma Epidemiology/Pathology
Debate exists whether FC should be classified as an HCC variant or a separate entity.80 FC accounts for less than 5% of HCC and arises in younger Caucasian patients in their second to third decade.79 FC arises in an otherwise normal liver, in contrast to most patients with HCC. A characteristic lamellar connective tissue is noted with macrohepatocytes.81 Clinical Presentation/Workup
Most patients present with local symptoms such as right upper quadrant pain. Imaging reveals a large, lobulated, heterogeneous mass with a central scar.82 α-Fetoprotein (AFP) levels are normal. Although more indolent than HCC, FC can metastasize, and one-fourth have metastases at diagnosis. Common sites include lung and abdominal cavity83 with bone reported. Treatment
FC has a better prognosis than HCC. FC is associated with a 46% reduction in risk of 5-year mortality compared to HCC.80 However, a pediatric analysis showed no difference in event-free survival between FC and HCC.84 In general, FC has greater resectability and improved survival compared to HCC.85 When patients with FC have unresectable or distant disease, survival may not be significantly greater. One-third of pediatric patients have long-term disease-free survival.84 No prospective data compare resection versus transplant for FC. In one small series, patients undergoing resection had a 3-year survival of 100% (n = 11), compared to 76% with liver transplantation (n = 9). Overall survival was 50% at 5 years.86 Tumor-free survival is 33% at 5 years and 29% at 10 years for patients treated with partial hepatectomy
387
or liver transplantation.87 Prognostic features include stage, positive lymph nodes, and vascular invasion. Thus, hepatic transplant is an option for FC not amenable to surgical resection. Patients should be evaluated in centers with surgical experience with this uncommon cancer. Despite resection or transplantation, many patients will have recurrence. Thus, neoadjuvant or adjuvant chemotherapy can be considered. For HCC, chemotherapy is without clear benefit. In FC, cisplatin-based chemotherapy may have minimal activity.84 5-FU with INF-α has shown responses and warrants additional study.88 For patients with metastatic disease, not eligible for a clinical study, a 5-FU –based regimen is reasonable. For locally advanced, unresectable FC, chemoembolization can be considered.89
Mixed Hepatocellular and Cholangiocarcinoma These tumors are divided into those coincidentally containing HCC and cholangiocarcinoma in the same liver (type I), “transitional tumors” from HCC to cholangiocarcinoma (type II), and “fibrolamellar tumors” which resemble FC but contain mucin-producing pseudoglands (type III).90 However, classification has been inconsistent. A recent series demonstrated that the majority of these tumors arise in noncirrhotic livers, with equal male/female distribution.91 AFP levels are lower than with HCC. However, other series demonstrate more similar features to HCC.92 Thus, this diagnosis should be considered in patients with adenocarcinoma in a liver biopsy with an elevated AFP. Most series suggest the prognosis is similar or worse than either HCC or cholangiocarcinoma for resected patients. One review demonstrated a median survival of 1.8 years, with 5and 10-year survival of 23 and 11% respectively.92 Another demonstrated a 24% 5-year survival for resected patients, but all unresectable patients died within 18 months.91 The most common recurrence site was the liver. For locally unresectable disease, chemoembolization for a solitary or dominant disease focus is appropriate, given its efficacy in HCC.93 For multifocal or metastatic disease, consideration of 5-FU or gemcitabine-based chemotherapy is reasonable outside of a clinical trial.
Mesenchymal Tumors of the Liver Primary liver angiosarcomas account for 2% of primary liver tumors.94 Approximately one-fourth are associated with chemical exposure, including vinyl chloride, Thorotrast (latency period 20–42 years), and arsenic.94 Patients present with abdominal pain, weakness, and fatigue. Imaging reveals a hepatic mass which may have intense enhancement from Thorotrast. Prognosis is poor, due to limited resectability. Patients often die of liver failure or tumor rupture with hemorrhage. Chemotherapy efficacy is unclear. Responses occur with ifosfamide and doxorubicin-containing regimens95 and paclitaxel (scalp and facial lesions).96 In the absence of an available study, use of these agents is reasonable. Hepatic epithelioid hemangioendotheliomas are vascular tumors of endothelial origin.97 More than 200 cases have been reported, with a link to oral contraceptives.98 There is a female predominance with nonspecific symptoms (right upper quadrant pain, weight loss), although
388
GASTROINTESTINAL TUMORS
patients may present with liver failure, Budd-Chiari syndrome, or portal hypertension.99 In over 40% the tumor is found incidentally.99 Liver imaging demonstrates multifocal, bilobar involvement.100 Histologically, tumors contain dendritic and epithelioid cells with immunohistochemistry positive for at least one endothelial cell marker (FVIII-RAg, CD34, and/or CD31).99 Clinical course is variable, with tumors acting in a range from indolent to aggressive. Resection or liver transplantation can be curative, with half the patients alive at 5 years. In a large series, 27% of patients developed metastases99 and nearly half of all deaths occurred within 16 months of diagnosis. Given its rarity and spectrum of clinical behavior, the efficacy of systemic chemotherapy or liver-directed therapies is unknown. Expectant observation for indolent tumors can be considered. Approximately 50 primary liver leiomyosarcomas are documented, with a mean age of 53 and equal male/female distribution.101 No causative factor is known.94 Presenting symptoms relate to a hepatic mass (usually right lobe) and ascites. CT scan demonstrates heterogeneous enhancement, in contrast to metastatic liver leiomyosarcomas.102 Metastatic disease is present in 40% at diagnosis. Supportive care yields median survival of less than one year. Patients undergoing resection have survival of more than 3 years. For patients with metastatic disease, therapy with an adriamycincontaining regimen is reasonable. While common in soft tissue, malignant fibrous histiocytomas of the liver are extremely rare, with 29 cases reported in the international literature (mean age 51, 16 men, 13 women).103 Most tumors are large (mean diameter 12 cm), of pleomorphic storiform histology, and over one-third invade adjacent organs. Surgical resection is recommended, although only one-fourth of patients are disease-free in the long-term.103 The impact of chemotherapy is unknown. Rhabdomyosarcomas of liver typically appear in children and present with early jaundice due to biliary obstruction. Cure is possible with resection. The role of chemotherapy is unclear.94 Less than 10 case reports of primary hepatic osteosarcoma104 and primary hepatic schwannoma105 have been reported. The latter is typically associated with von Recklinghausen’s disease.
patients in whom complete resection is not possible or with intrahepatic recurrence, liver transplantation can result in long-term survival. In one large analysis, 6-year survival was 82% in 106 patients undergoing initial liver transplantation and 30% for 41 patients undergoing salvage transplant.108 For patients with metastatic disease, chemotherapy with activity includes cisplatin, doxorubicin, cyclophosphamide, vincristine, and 5-FU.106
Hepatoblastoma
1. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55(1): 10 – 30. 2. Lazcano-Ponce EC, et al. Epidemiology and molecular pathology of gallbladder cancer. CA Cancer J Clin 2001; 51(6): 349 – 64. 3. Diehl AK. Gallstone size and the risk of gallbladder cancer. JAMA 1983; 250(17): 2323 – 6. 4. Moerman CJ, et al. Gallstone size and the risk of gallbladder cancer. Scand J Gastroenterol 1993; 28(6): 482 – 6. 5. Misra S, et al. Carcinoma of the gallbladder. Lancet Oncol 2003; 4(3): 167 – 76. 6. Carriaga MT, Henson DE. Liver, gallbladder, extrahepatic bile ducts, and pancreas. Cancer 1995; 75(1 Suppl): 171 – 90. 7. Yamagiwa H, Tomiyama H. Intestinal metaplasia-dysplasia-carcinoma sequence of the gallbladder. Acta Pathol Jpn 1986; 36(7): 989 – 97. 8. Su WC, et al. A clinical study of 130 patients with biliary tract cancers and periampullary tumors. Oncology 1996; 53(6): 488 – 93. 9. Gore RM, et al. Imaging benign and malignant disease of the gallbladder. Radiol Clin North Am 2002; 40(6): 1307 – 23, vi. 10. Kokudo N, et al. Strategies for surgical treatment of gallbladder carcinoma based on information available before resection. Arch Surg 2003; 138(7): 741 – 50; discussion 750.
Hepatoblastomas are almost exclusively found in children between 6 months and 3 years, with a male predominance.106 They are derived from undifferentiated embryonal tissue and thought to develop from pluripotent hepatic stem cells.106 They are found more commonly in families with familial adenomatous polyposis.107 Patients typically present with an asymptomatic abdominal mass, with weight loss, anorexia, and abdominal pain indicative of advanced disease. Distant metastases are found in 20% of patients, most commonly in the lungs. Lesions are often a solitary mass in the right lobe of the liver. Curative therapy involves surgery, but given the exquisite chemosensitivity of this tumor, preoperative chemotherapy has become a standard. In one study, preoperative cisplatin and doxorubicin resulted in less extensive liver resection in approximately one-fourth of patients.106 Five-year survival for all patients is approximately 75%. For
Other Hepatic Tumors Less than 100 primary hepatic carcinoid tumors have been reported. These tumors may arise from a pluripotent stem cell.109 Radiolabeled-octreotide imaging is essential to exclude an alternative primary, as the liver is the most common metastatic site for GI carcinoids. Survival at 5 (78%) and 10 (59%) years is high.110 For resected patients, 10-year survival is 68%. Adjuvant therapy is not of proven benefit. Primary hepatic NHL constitutes 0.016% of all NHL.111 It is associated with hepatitis C infection. Typical symptoms are abdominal pain and B symptoms in middle age patients, with a male:female ratio of 2.3 : 1. Imaging commonly reveals a solitary liver mass, although a third have multiple lesions. The most common histology is diffuse large B cell. Typical chemotherapy for NHL, including CHOP plus rituximab, is commonly employed, although median survival is 15 months.111 Nearly 200 cystadenomas and 100 cystadenocarcinomas have been reported. Cystadenomas are usually found in middle-aged women.112 The cause is unknown. The typical appearance is of a lobulated, multiloculated mass. Histologically, most contain an “ovarian-like” stroma.113 CA 19-9 may be elevated. Treatment requires complete excision, as malignant transformation has been reported.112 Approximately 100 cystadenocarcinomas of the liver have been reported. It similarly has a female predominance with nonspecific symptoms. Formal hepatic resection is preferred to partial excision, as two-thirds of patients undergoing local excision have recurrences with a 5-year survival of only 36%, in contrast with 65–100% 5-year survival for hepatic lobectomy or extensive hepatic resection.114
REFERENCES
UNCOMMON HEPATOBILIARY TUMORS 11. Ogura Y, et al. Radical operations for carcinoma of the gallbladder: present status in Japan. World J Surg 1991; 15(3): 337 – 43. 12. Shimada H, et al. The role of lymph node dissection in the treatment of gallbladder carcinoma. Cancer 1997; 79(5): 892 – 9. 13. Fong Y, Jarnagin W, Blumgart LH. Gallbladder cancer: comparison of patients presenting initially for definitive operation with those presenting after prior noncurative intervention. Ann Surg 2000; 232(4): 557 – 69. 14. Shoup M, Fong Y. Surgical indications and extent of resection in gallbladder cancer. Surg Oncol Clin N Am 2002; 11(4): 985 – 94. 15. Greene FLPD, et al. AJCC Cancer Staging Handbook, 6th ed. Philadelphia, Pennsylvania: Lippincott Raven Publisher, 2002. 16. Kopelson G, et al. Patterns of failure after curative surgery for extrahepatic biliary tract carcinoma: implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 1981; 7(3): 413 – 7. 17. Kresl JJ, et al. Adjuvant external beam radiation therapy with concurrent chemotherapy in the management of gallbladder carcinoma. Int J Radiat Oncol Biol Phys 2002; 52(1): 167 – 75. 18. Czito BG, et al. Adjuvant external-beam radiotherapy with concurrent chemotherapy after resection of primary gallbladder carcinoma: A 23year experience. Int J Radiat Oncol Biol Phys 2005; 62(4): 1030 – 4. 19. Jarnagin WR, et al. Patterns of initial disease recurrence after resection of gallbladder carcinoma and hilar cholangiocarcinoma: implications for adjuvant therapeutic strategies. Cancer 2003; 98(8): 1689 – 700. 20. http://www.nccn.org/professionals/physician gls/PDF/ hepatobiliary.pdf, 2005. 21. Takada T, et al. Is postoperative adjuvant chemotherapy useful for gallbladder carcinoma? A phase III multicenter prospective randomized controlled trial in patients with resected pancreaticobiliary carcinoma. Cancer 2002; 95(8): 1685 – 95. 22. Falkson G, MacIntyre JM, Moertel CG. Eastern Cooperative Oncology Group experience with chemotherapy for inoperable gallbladder and bile duct cancer. Cancer 1984; 54(6): 965 – 9. 23. Doval DC, et al. A phase II study of gemcitabine and cisplatin in chemotherapy-naive, unresectable gall bladder cancer. Br J Cancer 2004; 90(8): 1516 – 20. 24. Glover KY, et al. A Phase II Study of Oxaliplatin and Capecitabine (XELOX) in Patients with Unresectable Cholangiocarcinoma, including Carcinoma of the Gallbladder and Biliary Tract. Proc Am Soc Clin Oncol, 2005; 23(16S): 338s, abst 4123. 25. Gallardo JO, et al. A phase II study of gemcitabine in gallbladder carcinoma. Ann Oncol 2001; 12(10): 1403 – 6. 26. Knox JJ, et al. Combining gemcitabine and capecitabine in patients with advanced biliary cancer: a phase II trial. J Clin Oncol 2005; 23(10): 2332 – 8. 27. Alberts SR, et al. CPT-11 for bile-duct and gallbladder carcinoma: a phase II North Central Cancer Treatment Group (NCCTG) study. Int J Gastrointest Cancer 2002; 32(2 – 3): 107 – 14. 28. Kuhn R, et al. Outpatient therapy with gemcitabine and docetaxel for gallbladder, biliary, and cholangio-carcinomas. Invest New Drugs 2002; 20(3): 351 – 6. 29. Abou-Alfa ORE, et al. A phase II study of intravenous DX-8951f administered daily for five days, every three weeks to patients with biliary tree cancer (cholangiocarcinoma and gallbladder cancer). Proceedings of Gastrointestinal Cancers Symposium, San Francisco, CA, 2004, abst 94. 30. Henson DE, Albores-Saavedra J, Corle D. Carcinoma of the gallbladder. Histologic types, stage of disease, grade, and survival rates. Cancer 1992; 70(6): 1493 – 7. 31. Waisberg J, et al. Squamous cell carcinoma of the gallbladder. Sao Paulo Med J 2001; 119(1): 43. 32. Oohashi Y, et al. Adenosquamous carcinoma of the gallbladder warrants resection only if curative resection is feasible. Cancer 2002; 94(11): 3000 – 5. 33. Albores-Saavedra J, et al. Papillary carcinomas of the gallbladder: analysis of noninvasive and invasive types. Arch Pathol Lab Med 2005; 129(7): 905 – 9. 34. Albores-Saavedra J, Molberg K, Henson DE. Unusual malignant epithelial tumors of the gallbladder. Semin Diagn Pathol 1996; 13(4): 326 – 38.
389
35. Bittinger A, Altekruger I, Barth P. Clear cell carcinoma of the gallbladder. A histological and immunohistochemical study. Pathol Res Pract 1995; 191(12): 1259 – 65; discussion 1266. 36. Fujii H, et al. Small cell carcinoma of the gallbladder: a case report and review of 53 cases in the literature. Hepatogastroenterology 2001; 48(42): 1588 – 93. 37. Modlin IM, Shapiro MD, Kidd M. An analysis of rare carcinoid tumors: clarifying these clinical conundrums. World J Surg 2005; 29(1): 92 – 101. 38. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carcinoid tumors. Cancer 2003; 97(4): 934 – 59. 39. Huguet KL, Hughes CB, Hewitt WR. Gallbladder carcinosarcoma: a case report and literature review. J Gastrointest Surg 2005; 9(6): 818 – 21. 40. Zeig DA, et al. Leiomyosarcoma of the gallbladder – a case report and review of the literature. Acta Oncol 1998; 37(2): 212 – 4. 41. Dong XD, et al. Melanoma of the gallbladder: a review of cases seen at Duke University Medical Center. Cancer 1999; 85(1): 32 – 9. 42. Lazaridis KN, Gores GJ. Cholangiocarcinoma. Gastroenterology 2005; 128(6): 1655 – 67. 43. Narayanan Menon KV, Wiesner RH. Etiology and natural history of primary sclerosing cholangitis. J Hepatobiliary Pancreat Surg 1999; 6(4): 343 – 51. 44. Watanapa P, Watanapa WB. Liver fluke-associated cholangiocarcinoma. Br J Surg 2002; 89(8): 962 – 70. 45. Metcalfe MS, Wemyss-Holden SA, Maddern GJ. Management dilemmas with choledochal cysts. Arch Surg 2003; 138(3): 333 – 9. 46. Yamamoto S, et al. Hepatitis C virus infection as a likely etiology of intrahepatic cholangiocarcinoma. Cancer Sci 2004; 95(7): 592 – 5. 47. Kang YK, Kim WH, Jang JJ. Expression of G1-S modulators (p53, p16, p27, cyclin D1, Rb) and Smad4/Dpc4 in intrahepatic cholangiocarcinoma. Hum Pathol 2002; 33(9): 877 – 83. 48. Tada M, Omata M, Ohto M. High incidence of ras gene mutation in intrahepatic cholangiocarcinoma. Cancer 1992; 69(5): 1115 – 8. 49. Sirica AE. Cholangiocarcinoma: molecular targeting strategies for chemoprevention and therapy. Hepatology 2005; 41(1): 5 – 15. 50. Paganuzzi M, et al. CA 19-9 and CA 50 in benign and malignant pancreatic and biliary diseases. Cancer 1988; 61(10): 2100 – 8. 51. de Bellis M, et al. Influence of stricture dilation and repeat brushing on the cancer detection rate of brush cytology in the evaluation of malignant biliary obstruction. Gastrointest Endosc 2003; 58(2): 176 – 82. 52. Byrne MF, et al. Yield of endoscopic ultrasound-guided fine-needle aspiration of bile duct lesions. Endoscopy 2004; 36(8): 715 – 9. 53. Jarnagin WR, Shoup M. Surgical management of cholangiocarcinoma. Semin Liver Dis 2004; 24(2): 189 – 99. 54. Weber SM, et al. Intrahepatic cholangiocarcinoma: resectability, recurrence pattern, and outcomes. J Am Coll Surg 2001; 193(4): 384 – 91. 55. Isa T, et al. Predictive factors for long-term survival in patients with intrahepatic cholangiocarcinoma. Am J Surg 2001; 181(6): 507 – 11. 56. Jarnagin WR. Cholangiocarcinoma of the extrahepatic bile ducts. Semin Surg Oncol 2000; 19(2): 156 – 76. 57. Fong Y, et al. Outcome of treatment for distal bile duct cancer. Br J Surg 1996; 83(12): 1712 – 5. 58. Macdonald JS, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 2001; 345(10): 725 – 30. 59. Neoptolemos JP, et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med 2004; 350(12): 1200 – 10. 60. Neuhaus P, et al. A randomised, prospective, multicenter, phase III trial of adjuvant chemotherapy with gemcitabine vs. observation in patients with resected pancreatic cancer. Proc Am Soc Clin Oncol 2005; 23(16S): 311s, abst 4013. 61. Todoroki T. Chemotherapy for bile duct carcinoma in the light of adjuvant chemotherapy to surgery. Hepatogastroenterology 2000; 47(33): 644 – 9. 62. Todoroki T, et al. Benefits of adjuvant radiotherapy after radical resection of locally advanced main hepatic duct carcinoma. Int J Radiat Oncol Biol Phys 2000; 46(3): 581 – 7.
390
GASTROINTESTINAL TUMORS
63. Mahe M, et al. Radiation therapy in extrahepatic bile duct carcinoma. Radiother Oncol 1991; 21(2): 121 – 7. 64. Kim S, et al. Role of postoperative radiotherapy in the management of extrahepatic bile duct cancer. Int J Radiat Oncol Biol Phys 2002; 54(2): 414 – 9. 65. Crane CH, et al. Limitations of conventional doses of chemoradiation for unresectable biliary cancer. Int J Radiat Oncol Biol Phys 2002; 53(4): 969 – 74. 66. McMasters KM, et al. Neoadjuvant chemoradiation for extrahepatic cholangiocarcinoma. Am J Surg 1997; 174(6): 605 – 8; discussion 608 – 9. 67. Moertel CG, et al. 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 1981; 48(8): 1705 – 10. 68. Cohen SJ, et al. A randomized phase III study of radiotherapy alone or with 5-fluorouracil and mitomycin-C in patients with locally advanced adenocarcinoma of the pancreas: Eastern Cooperative Oncology Group study E8282. Int J Radiat Oncol Biol Phys 2005; 62(5): 1345 – 50. 69. Martin RC II, et al. Cost comparison of endoscopic stenting vs surgical treatment for unresectable cholangiocarcinoma. Surg Endosc 2002; 16(4): 667 – 70. 70. Ortner ME, et al. Successful photodynamic therapy for nonresectable cholangiocarcinoma: a randomized prospective study. Gastroenterology 2003; 125(5): 1355 – 63. 71. Albores-Saavedra J, et al. Noninvasive and minimally invasive papillary carcinomas of the extrahepatic bile ducts. Cancer 2000; 89(3): 508 – 15. 72. Abraham SC, et al. Molecular and immunohistochemical analysis of intraductal papillary neoplasms of the biliary tract. Hum Pathol 2003; 34(9): 902 – 10. 73. Sewkani A, et al. Squamous cell carcinoma of the distal common bile duct. Jop 2005; 6(2): 162 – 5. 74. Okabayashi T, et al. Adenosquamous carcinoma of the extrahepatic biliary tract: clinicopathological analysis of Japanese cases of this uncommon disease. J Gastroenterol 2005; 40(2): 192 – 9. 75. Nagai E, et al. Undifferentiated carcinoma of the common bile duct: case report and review of the literature. J Hepatobiliary Pancreat Surg 2002; 9(5): 627 – 31. 76. Chamberlain RS, Blumgart LH. Carcinoid tumors of the extrahepatic bile duct. A rare cause of malignant biliary obstruction. Cancer 1999; 86(10): 1959 – 65. 77. Das K, et al. Primary non-Hodgkin’s lymphoma of the bile ducts mimicking cholangiocarcinoma. Surgery 2003; 134(3): 496 – 500. 78. Sodergren MH, et al. Carcinosarcoma of the biliary tract: two case reports and a review of the literature. Eur J Gastroenterol Hepatol 2005; 17(6): 683 – 5. 79. Pawlik TM, et al. Advances in the surgical management of liver malignancies. Cancer J 2004; 10(2): 74 – 87. 80. El-Serag HB, Davila JA. Is fibrolamellar carcinoma different from Hepatocellular carcinoma? A US population-based study. Hepatology 2004; 39(3): 798 – 803. 81. Balis EJLG. Pathology and natural history of Hepatocellular carcinoma. In Abbruzzese JL, et al. (eds) Gastrointestinal Oncology: Oxford University Press, 2004: 507 – 514. 82. McLarney JK, et al. Fibrolamellar carcinoma of the liver: radiologicpathologic correlation. Radiographics 1999; 19(2): 453 – 71. 83. Epstein BE, et al. Metastatic nonresectable fibrolamellar hepatoma: prognostic features and natural history. Am J Clin Oncol 1999; 22(1): 22 – 8. 84. Katzenstein HM, et al. Fibrolamellar hepatocellular carcinoma in children and adolescents. Cancer 2003; 97(8): 2006 – 12. 85. Okuda K. Natural history of hepatocellular carcinoma including fibrolamellar and hepato-cholangiocarcinoma variants. J Gastroenterol Hepatol 2002; 17(4): 401 – 5. 86. El-Gazzaz G, et al. Outcome of liver resection and transplantation for fibrolamellar hepatocellular carcinoma. Transpl Int 2000; 13(Suppl 1): S406 – 9. 87. Pinna AD, et al. Treatment of fibrolamellar hepatoma with subtotal hepatectomy or transplantation. Hepatology 1997; 26(4): 877 – 83.
88. Patt YZ, et al. Phase II trial of systemic continuous fluorouracil and subcutaneous recombinant interferon Alfa-2b for treatment of hepatocellular carcinoma. J Clin Oncol 2003; 21(3): 421 – 7. 89. Czauderna P, et al. Preliminary experience with arterial chemoembolization for hepatoblastoma and hepatocellular carcinoma in children. Pediatr Blood Cancer 2005. 90. Goodman ZD, et al. Combined hepatocellular-cholangiocarcinoma. A histologic and immunohistochemical study. Cancer 1985; 55(1): 124 – 35. 91. Jarnagin WR, et al. Combined hepatocellular and cholangiocarcinoma: demographic, clinical, and prognostic factors. Cancer 2002; 94(7): 2040 – 6. 92. Yano Y, et al. Combined hepatocellular and cholangiocarcinoma: a clinicopathologic study of 26 resected cases. Jpn J Clin Oncol 2003; 33(6): 283 – 7. 93. Llovet JM, et al. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial. Lancet 2002; 359(9319): 1734 – 9. 94. Mani H, Van Thiel DH. Mesenchymal tumors of the liver. Clin Liver Dis 2001; 5(1): 219 – 57, viii. 95. Budd GT. Management of angiosarcoma. Curr Oncol Rep 2002; 4(6): 515 – 9. 96. Fata F, et al. Paclitaxel in the treatment of patients with angiosarcoma of the scalp or face. Cancer 1999; 86(10): 2034 – 7. 97. Weiss SW, Enzinger FM. Epithelioid hemangioendothelioma: a vascular tumor often mistaken for a carcinoma. Cancer 1982; 50(5): 970 – 81. 98. Dean PJ, Haggitt RC, O’Hara CJ. Malignant epithelioid hemangioendothelioma of the liver in young women. Relationship to oral contraceptive use. Am J Surg Pathol 1985; 9(10): 695 – 704. 99. Makhlouf HR, Ishak KG, Goodman ZD. Epithelioid hemangioendothelioma of the liver: a clinicopathologic study of 137 cases. Cancer 1999; 85(3): 562 – 82. 100. Lyburn ID, et al. Hepatic epithelioid hemangioendothelioma: sonographic, CT, and MR imaging appearances. AJR Am J Roentgenol 2003; 180(5): 1359 – 64. 101. Gates LK Jr, et al. Primary leiomyosarcoma of the liver mimicking liver abscess. Am J Gastroenterol 1995; 90(4): 649 – 52. 102. Soyer P, et al. Hepatic leiomyosarcomas: CT features with pathologic correlation. Eur J Radiol 1995; 19(3): 177 – 82. 103. Anagnostopoulos G, et al. Malignant fibrous histiocytoma of the liver: a case report and review of the literature. Mt Sinai J Med 2005; 72(1): 50 – 2. 104. Govender D, Rughubar KN. Primary hepatic osteosarcoma: case report and literature review. Pathology 1998; 30(3): 323 – 5. 105. Lederman SM, et al. Hepatic neurofibromatosis, malignant schwannoma, and angiosarcoma in von Recklinghausen’s disease. Gastroenterology 1987; 92(1): 234 – 9. 106. Schnater JM, et al. Where do we stand with hepatoblastoma? A review. Cancer 2003; 98(4): 668 – 78. 107. Hirschman BA, Pollock BH, Tomlinson GE. The Spectrum of APC mutations in children with Hepatoblastoma from Familial Adenomatous Polyposis Kindreds. J Pediatr 2005; 147(2): 263 – 6. 108. Otte JB, et al. Liver transplantation for hepatoblastoma: results from the International Society of Pediatric Oncology (SIOP) study SIOPEL1 and review of the world experience. Pediatr Blood Cancer 2004; 42(1): 74 – 83. 109. Andreola S, et al. A clinicopathologic study of primary hepatic carcinoid tumors. Cancer 1990; 65(5): 1211 – 8. 110. Knox CD, et al. Long-term survival after resection for primary hepatic carcinoid tumor. Ann Surg Oncol 2003; 10(10): 1171 – 5. 111. Noronha V, et al. Primary non-Hodgkin’s lymphoma of the liver. Crit Rev Oncol Hematol 2005; 53(3): 199 – 207. 112. Devaney K, Goodman ZD, Ishak KG. Hepatobiliary cystadenoma and cystadenocarcinoma. A light microscopic and immunohistochemical study of 70 patients. Am J Surg Pathol 1994; 18(11): 1078 – 91. 113. Vogt DP, Henderson JM, Chmielewski E. Cystadenoma and cystadenocarcinoma of the liver: a single center experience. J Am Coll Surg 2005; 200(5): 727 – 33. 114. Lauffer JM, et al. Biliary cystadenocarcinoma of the liver: the need for complete resection. Eur J Cancer 1998; 34(12): 1845 – 51.
Section 6 : Gastrointestinal Tumors
34
Cancer of the Small Bowel
Robert R. McWilliams, Thomas C. Smyrk and Axel Grothey
INTRODUCTION The small bowel is arbitrarily divided into the duodenum, jejunum, and ileum, with a gradual transition of histologic anatomy and biologic functions. All regions of the small intestine have in common a villous mucosa, submucosa, and muscularis propria. The epithelium that lines the villi and crypts of Lieberk¨uhn is a single cell layer. The epithelium absorbs nutrients, maintains ionic balance, and provides antigen exclusion. The proliferative stem cell compartment of the epithelium is located in the base of the crypt. The stem cells proliferate, giving rise to daughter cells, which undergo replication. With luminal axial migration, the cells differentiate and, upon reaching the epithelial surface, ultimately undergo apoptotic cell death with subsequent phagocytosis by the underlying lamina propria macrophages. Admixed with the enterocytes at the base of the crypt are the neuroendocrine cells and Paneth cells. The duodenum is distinguished by the submucosally located Brunner’s glands, while the terminal ileum is rich in mucosa-associated lymphoid tissue (MALT) arranged in dome-shaped follicles and broad Peyer’s patches. While malignant tumors of the small bowel may arise from any of these cell types, the probability of neoplastic transformation differs widely among them. Malignant tumors of the small bowel are uncommon.1 – 3 Approximately 5400 new cases of small bowel cancer will be diagnosed in 2005 in the United States, resulting in slightly more than 1000 deaths. The incidence of small bowel cancer increases with age, is greater in males, and is increased by conditions such as celiac disease, human immunodeficiency virus (HIV), Crohn’s disease, and familial colon cancer syndromes.4 – 7 The most common primary small bowel malignancies are adenocarcinomas, carcinoid tumors, and lymphomas.5 Primary small bowel sarcomas or stromal tumors are less common. More than half of adenocarcinomas are present in the duodenum, most sarcomas in the jejunum, and most carcinoids in the ileum.8 Metastatic tumors to the small bowel are more common than primary small bowel malignancies in certain subgroups of patients (e.g. patients with known or suspected metastases).3
CLINICAL PRESENTATION OF SMALL BOWEL CANCERS Patients with cancer of the small bowel frequently present with symptoms prior to diagnosis. Unfortunately, the symptoms usually develop when the tumor is at an advanced stage. There may be a significant delay averaging more than 8 months from the onset of symptoms until the diagnosis is secured.1 Abdominal pain, nausea, vomiting, weight loss, and bleeding occur frequently; diarrhea in the absence of bleeding suggests carcinoid.8 Not uncommonly, the tumor is discovered during emergency or exploratory surgery for unspecific abdominal symptoms. Obstruction and perforation occur in the minority of patients. A palpable mass on physical examination is also a less common finding. Patients may occasionally present with systemic symptoms, such as fevers and night sweats from lymphoma or the flushing syndrome associated with carcinoid tumors.
DIAGNOSIS OF CANCERS OF THE SMALL BOWEL The preoperative diagnosis of small bowel cancers may be made on radiographic studies. A plain abdominal radiograph may demonstrate small bowel obstruction, but is not specific for small bowel cancer because of a myriad of potential causes. A small bowel follow-through may demonstrate an intraluminal neoplasm.2 During this study, the patient ingests 700–1000 mL of barium, and the radiologist performs interval fluoroscopy to examine the intestine. Forty to sixty percent of small bowel tumors may be accurately characterized with this technique.9,10 However, enteroclysis is a superior examination for the detection of primary small bowel cancers. During this study, a catheter is manipulated past the ligament of Treitz for instillation of contrast. A single-contrast enteroclysis employs only low-density, lowviscosity barium. A double-contrast enteroclysis uses air or methlycellulose as a second contrast agent to achieve luminal distension. In contrast to a small bowel followthrough examination, enteroclysis results in superior luminal distension and superior evaluation of fold thickness and mucosal detail. Small bowel tumors may be detected during computed tomography (CT) performed for abdominal pain or
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
392
GASTROINTESTINAL TUMORS
distension. CT scans can also demonstrate mesenteric, nodal, or hepatic metastases.3 Endoscopic evaluation is frequently used in the diagnosis of small bowel cancers. Esophagogastric-duodenoscopy (EGD) with push enteroscopy may localize and diagnose proximal small bowel tumors, such as periampullary carcinomas and proximal jejunal tumors. Colonoscopy with retroileoscopy may detect lymphoma as well as other malignancies of the ileum.4 However, these endoscopic imaging techniques still leave a significant length of small bowel unevaluable. Capsule endoscopy has gained increasing utilization, and appears to have a higher sensitivity and specificity than push enteroscopy and barium studies for causes of small bowel blood loss.11,12 However, biopsy is not possible with capsule endoscopy, obstructive symptoms remain a contraindication, and the technology is not yet widely available. Laboratory testing is frequently nonspecific. Gastrointestinal tract bleeding may lead to iron deficiency anemia. Abnormalities of liver function tests may occur with hepatic metastases. Carcinoid syndrome may be diagnosed by an elevated urinary 5-hydroxyindoleacetic acid (5-HIAA) level, but this is usually elevated only in the presence of liver metastases.13
SPECIFIC SMALL BOWEL NEOPLASMS Adenocarcinoma Biology and Pathology
Adenocarcinoma of the small bowel is the most common primary small bowel malignancy, accounting for approximately 45–60% of cases.3,8 Adenocarcinomas are derived from the epithelium through a series of genetic events leading first to dysplasia and then to carcinoma with the potential for invasion.4,14 – 17 Dysplasia is defined on the basis of nuclear histologic features that are indicative of the first series of genetic events. Dysplastic epithelium shows elongated, hyperchromatic nuclei with a pseudostratified arrangement. This aggregate of dysplastic epithelium is often, but not always, identifiable as a polypoid lesion termed an adenoma. With the acquisition of invasive capability, the term carcinoma is applied. Carcinomas may display varying degrees of differentiation. Tumors that maintain a high degree of glandular architecture are well differentiated, whereas those that are formed of discohesive cells with little cell-to-cell adherence are poorly differentiated. Most adenomas and carcinomas occur in the proximal small intestine, extending from the papilla of Vater to the second loop of the jejunum.8,18,19 The reasons for this geographic predilection are unclear, though the proliferative and possible carcinogenic effects of bile have been suggested as possibilities.20 Exceptions include the carcinomas that arise in association with surgical anastomotic sites and underlying Crohn’s disease.21 – 24 The carcinomas arising at anastomotic sites may occur up to 25 years after surgery. The carcinomas occurring in Crohn’s disease arise in disease-involved segments of the gut.25 – 27 Small bowel carcinomas associated with Crohn’s disease also follow the dysplasia –carcinoma
sequence. Morphologically, these adenocarcinomas display the same range of differentiation as other small bowel carcinomas. The genetics of these tumors seem to involve, in part, mutations in TGF (transforming growth factor)-β receptors. Additionally, DNA mismatch repair defects may also be involved in Crohn’s disease tumor pathogenesis at a phenotypic level as microsatellite instability, but no mutations in hMLH1 or hMSH2 have been found in the nondysplastic epithelium. There is also an increased incidence of adenocarcinoma arising in long-standing gluten-sensitive enteropathy (celiac disease).28 – 32 In gluten-sensitive enteropathy, there is increased antigenic stimulation with increased intestinal permeability and chronic inflammation. The mechanisms for the development of carcinoma arising in celiac disease are not known, but a recent report found that 8 of 11 adenocarcinomas arising in the setting of celiac disease were microsatellite instability-high.33 It may be that chronic inflammation or chronic exposure to something in the gut lumen increases the likelihood for hypermethylation of the hMLH1 promotor site in celiac disease. Other possible contributors to carcinogenesis include increased crypt mitosis, penetration of the epithelium by carcinogens or viruses, or the underlying immune deficiency inherent in celiac disease. Adenocarcinomas may also arise in the setting of various polyposis syndromes. Familial adenomatous polyposis (FAP) coli patients are born with a germ line mutation in one of their APC genes.28 While multiple colon adenomas and carcinomas commonly arise, the small bowel can also develop adenomas and carcinomas, albeit less commonly. As with sporadic small bowel adenomas and carcinomas, the periampullary region is the site most commonly involved in patients with FAP. Patients with a germ line mutation in the DNA mismatch repair genes (hMLH1 and hMSH2 ), termed the “Lynch syndrome” or “hereditary nonpolyposis colorectal cancer” (HNPCC), also develop small bowel adenocarcinomas at an increased frequency compared to nonaffected patients.6,34,35 The frequency of small bowel carcinomas in this patient population is far less than that of patients with FAP, but the prevalence of FAP in the general population is less than that of the Lynch syndrome, meaning that HNPCC makes a greater relative contribution to the total number of small bowel cancers.36 Juvenile polyposis patients may have numerous hamartomatous polyps, most of which are located in the colon, although small bowel involvement is also seen. These polyps have an increased risk of developing dysplasia and undergoing malignant transformation.37 In contrast, the hamartomatous polyps of Peutz –Jeghers syndrome are more likely to occur in the small bowel than in the colon. These polyps may as well progress to dysplasia and ultimately carcinoma. The risk of developing small bowel adenocarcinomas in these two genetic diseases is far less than in patients with FAP.38 Most carcinomas of the small bowel are adenocarcinomas, but adeno-squamous and pure squamous differentiation may be seen. These tumors are derived, like adenocarcinomas, from enterocytes. The mechanisms leading to the aberrant differentiation are not known. The aberrant differentiation
CANCER OF THE SMALL BOWEL
does not alter prognosis, which is strictly dependent on tumor stage.39 Approach to the Patient with Small Bowel Carcinoma
The most common presenting symptoms in patients with adenocarcinoma of the small bowel are pain, obstruction, and occult gastrointestinal tract bleeding. Symptoms are related to the size of the tumor as well as location in the small intestinal tract. The diagnosis of small bowel adenocarcinoma may be made preoperatively on the basis of an upper gastrointestinal series, small bowel follow-through examination, or enteroclysis. Enteroclysis is currently the best standard examination to diagnose or exclude a small bowel tumor,6 but capsule endoscopy is playing an increasing role when available. Location of the tumor and presenting symptoms can increase the likelihood of various histologies, but often the diagnosis is not secured until exploratory surgery and pathologic analysis. Radiographically, duodenal cancers appear as polypoid or carpetlike tumors near the ampulla of Vater, or polypoid, ulcerated, or annular lesions (see Figure 1) of the third and fourth portions of the duodenum.40,41 Jejunal adenocarcinomas usually appear as short, annular lesions with overhanging edges, nodular mucosa, and focal ulceration.42 During CT scan imaging, a jejunal cancer may demonstrate concentric or asymmetric soft tissue mass, causing luminal narrowing in the first two loops of the jejunum, within 25 cm of the ligament of Treitz. While metastases may have an annular configuration, these are usually located in the mid and distal small intestine, are eccentric with indistinct margins, and have preserved mucosal folds.42,43 Surgery is the mainstay of the treatment of small bowel adenocarcinomas when feasible.44 – 49 For patients with duodenal carcinomas, resection of the primary tumor with lymph
Figure 1 CT scan of a primary duodenal carcinoma narrowing the lumen of the bowel in concentric fashion.
393
node dissection is standard. Proximal tumors may require pancreaticoduodenectomy (Whipple procedure). More distal tumors of the small intestine are usually surgically approached with segmental resection followed by primary anastomosis.8 Most series have reported an approximate 20–30% survival at 5 years.44 – 49 The prognosis of patients who present with small bowel adenocarcinoma is related to the depth of penetration of the tumor into the muscular wall of the small bowel, as well as the presence or absence of nodal metastases.39 Some studies have suggested that histologic grade of the neoplasm also influences prognosis, while others have not.44,48 There are limited data with respect to the utilization of postoperative adjuvant therapy in patients with resected high-risk lesions. For the more proximal tumors, such as periampullary carcinomas, 5-fluorouracil (5-FU)-sensitized radiation therapy is often considered, although this approach did not statistically show benefit over observation alone in a phase III trial studying patients with pancreatic or periampullary cancers.50 Given the rarity of small bowl adenocarcinoma, there are no other prospective trials addressing the role of adjuvant therapy in this disease. Treatment discussions consequently need to be individualized, with the paucity of relevant data made clear to the patient. In general, a therapeutic approach similar to that for colon cancer or – less commonly – other upper gastrointestinal malignancies (e.g. gastric cancer) is undertaken. The management of patients with unresectable or metastatic disease focuses on palliation. Radiation therapy, usually in combination with 5-FU, may palliate locally advanced unresectable duodenal carcinomas. Case reports and small retrospective series comprise the majority of the literature on chemotherapy in the treatment of patients with metastatic adenocarcinoma of the small bowel. One case report describes a complete response to continuous infusion 5-FU, but no prospective clinical trials have been performed in the advanced disease setting.51 In a series reported out of MD Anderson, 14 patients with advanced disease received a number of drug regimens including 5-FU as a single agent or in combination with other agents. Nine patients had stable disease, two had minor responses, and one had a partial response. The median survival was 9 months.52 In another series, eight patients with metastatic small bowel cancer had been treated with continuous infusion 5-FU –based chemotherapy, most in combination with mitomycin C and epirubicin; there was one complete response and two partial responses, and the median survival was 13 months.53 In a series of seven patients treated with epirubicin, cisplatin, and 5-fluorouracil (ECF), four patients responded and survival of 11 months was reported.54 Ouriel reported a retrospective study of 65 patients with advanced disease with 5-FU –based chemotherapy; mean survival was 10.7 months in those treated, compared to 4 months in patients who were not treated with chemotherapy.48 Although this difference in survival may have been derived from selection bias that favored patients who were better candidates for chemotherapy, these results nonetheless suggest a role for
394
GASTROINTESTINAL TUMORS
further studying chemotherapy in the treatment of this disease. There is some literature describing good responses to combined 5-FU with radiation in the neoadjuvant setting, with some patients experiencing complete responses.55,56 However, many resected patients will ultimately develop distant metastases, suggesting that improved systemic therapy is needed57 . One recent publication of a phase II trial conducted through the Eastern Cooperative Oncology Group (ECOG) in the mid-1980s reported an 18.4% response rate in 39 patients treated with a regimen of 5-FU, mitomycin C, and doxorubicin. Six patients (15%) had grade 4 or higher hematologic toxicity.58 Although several of the aforementioned reports suggest that metastatic adenocarcinoma of the small bowel can respond to chemotherapy, there remains no well-accepted, standard therapy for the treatment of this disease. Moreover, publication bias may have selected the more successful case reports and series described above. There remains a clear need to pursue prospective clinical trials for patients with this malignancy and to test novel agents in this setting. In practice, outside of clinical trials, chemotherapy regimens developed for advanced colorectal and gastric cancer appear warranted and are commonly used.
Carcinoid Tumor Biology and Pathology
Approximately 25% of small bowel malignancies are carcinoid tumors.59 The majority of carcinoid tumors originate from the gut, predominately the appendix, followed by the small bowel and the rectum.13 Most of the small bowel carcinoids are located in the ileum. Carcinoids arise from enterochromaffin cells. These cells perform amine precursor uptake and decarboxylation (APUD), producing serotonin and other hormones. While carcinoid tumors are thought to be epithelial in origin, the genetics of their development in both the colon and the small bowel does not appear to involve APC gene mutations.60 These tumors usually grow as a mass in the submucosa with invasion into the gut wall rather than as an exophytic lesion. Their architecture may be trabecular or nested depending on the origin of the tumor in the midgut or foregut, respectively. Cytologically, the cells have nuclei with a “salt and pepper” chromatin pattern and a relatively low mitotic rate. Immunohistochemical stains for chromogranin A and synaptophysin are almost invariably positive in these tumors, confirming neuroendocrine lineage. Despite their seemingly bland histologic appearance, carcinoid tumors have the ability to locally invade and metastasize through lymphatic invasion. With invasion into the bowel wall, pronounced muscular hypertrophy, calcification, and fibrosis ensues,61 which may result in tethered segments of bowel. The underlying etiology for the fibrotic reaction is unknown.62 Encasement of mesenteric vessels may lead to ischemia or infarction,63 resulting in “elastic vascular sclerosis”.64 – 67 Approach to the Patient with Small Bowel Carcinoid
Most patients with small bowel carcinoid are asymptomatic, and the tumors that do not metastasize, are often discovered
incidentally at surgery or autopsy. Even many patients with extensive liver metastases from small bowel carcinoid may remain minimally symptomatic or asymptomatic.13 In the subset of patients with symptoms from their primary tumor, abdominal pain and obstructive symptoms are common. These symptoms may be due to the tumor itself or to the intense fibrotic reaction in the mesentery that frequently surrounds the primary tumor. In patients with metastases to the liver, the well-known carcinoid syndrome68 may occur, consisting of intermittent flushing, diarrhea, and, in advanced stages, right-sided cardiac valvular disease. Carcinoid syndrome is due to the excessive production of serotonin, most often measured clinically by urinary 5-HIAA, a biologically inactive metabolite of serotonin. While a minority of patients with small bowel carcinoid present with carcinoid syndrome, up to two-thirds may develop this later in the disease. The symptoms of carcinoid syndrome may be triggered by alcohol, emotional stress, or consumption of tyramine-containing foods. Bronchospasm mediated by histamine may also be demonstrated. Most small bowel carcinoid patients who develop carcinoid syndrome have hepatic metastases and an elevated urinary 5-HIAA level.13,52 If patients present with classic symptoms of the carcinoid syndrome, the diagnosis is often made easily. If patients do not have carcinoid syndrome, preoperative diagnosis of small bowel carcinoids may be made during radiographic studies performed for other reasons. Carcinoid tumors are usually at least 1–2 cm in size when detected by small bowel followthrough examination. The more modern studies mentioned earlier in this chapter may detect smaller tumors in patients with nonspecific symptoms. Carcinoid tumors of the small bowel, especially of the ileum, may be multiple in up to 30% of patients.69 Liver metastases may be demonstrated by CT scan, though they may be easier to detect by magnetic resonance imaging (MRI). (see Figure 2) CT scans, when used, should be performed with hepatic arterial phase imaging to detect the highly vascular lesions by contrast, whereas their relative isodensity compared with liver parenchyma can lead to falsely negative scans or underestimation of disease bulk.70 The finding of a low-grade neuroendocrine tumor metastatic to the liver raises the differential diagnosis of pulmonary carcinoid and pancreatic islet cell tumor. Immunohistochemical stains can help resolve the differential, with positive staining for thyroid transcription factor-1 (TTF-1) indicating pulmonary origin, and positive staining for caudal-type homeobox transcription factor 2 (CDX2) suggesting gastrointestinal origin. About one-third of pancreatic islet cell tumors are positive for CDX2. The surgical treatment of small bowel carcinoids includes wide resection of the primary tumor and mesentery. Distal ileal tumors may require a right hemicolectomy. Most surgeons perform a regional lymph node dissection to evaluate nodal disease. The risk of nodal metastasis is related to the size of the primary tumor. Search for synchronous multifocal tumors is important at the time of surgery. If the disease is not resectable, the focus of treatment is palliation. The growth rate of the disease, the extent of tumor
CANCER OF THE SMALL BOWEL
395
Figure 2 Comparison of a CT scan and contrast magnetic resonance imaging showing liver metastases from small bowel carcinoid. Often, liver metastases can be missed or underestimated by CT scan unless early-phase contrast imaging is employed (Acknowledgement to Patrick Burch, MD).
burden, and the severity of hormonal symptoms determine the best approach to treatment. Some patients may not need initial treatment because of the indolent nature of the tumor and the lack of symptoms. The 5-year survival for patients with distant disease at diagnosis exceeds 35%.62 The diarrhea associated with carcinoid syndrome may be treated with antidiarrheal agents. If these measures are not successful, or flushing is the debilitating symptom, the carcinoid syndrome may be controlled with the somatostatin analog octreotide acetate (SandostatinTM ). Octreotide not only is useful for controlling the symptoms related to carcinoid syndrome but it also can be utilized to prevent or treat carcinoid crisis,71 such as in patients undergoing surgical procedures.72 As such, injectable octreotide should be readily available when patients with carcinoid syndrome undergo surgery.73 While this agent in general is well tolerated, chronic utilization may be accompanied by the development of cholelithiasis, edema, nausea, and glucose intolerance. Rarely, objective antitumor responses have been reported in patients receiving octreotide.63 Patients with metastatic disease often have an indolent course and may be treated with a variety of approaches including chemotherapy, surgical resection of metastases, and hepatic artery embolization. Select patients with metastatic disease may simply be monitored if their disease is indolent. Resection of bulky hepatic metastases has resulted in prolonged remission of symptoms.74 Hepatic arterial embolization with or without intrahepatic chemotherapy will also be useful in patients with hepatic metastases who are suffering from symptoms of tumor bulk.75 Systemic chemotherapy including the utilization of single
agents such as 5-FU, doxorubicin, dacarbazine (DTIC), interferon, and streptozocin, and combinations such as 5FU plus streptozocin have been tested.76 Objective antitumor response rates are in the range of 20% with relatively short durations of response. For these reasons, chemotherapy is usually reserved for patients with progressive disease and symptoms despite other measures. Clinical trials incorporating novel agents targeting angiogenesis and tyrosine kinase pathways are currently under way for patients with metastatic neuroendocrine tumors.77,78 Additional information on the management of carcinoids may be found in Chapter 35, Unusual Tumors of the Colon, Rectum and Anus and Chapter 36, Cancer of the Appendix.
Lymphoma Biology and Pathology
Primary lymphoma of the small bowel is relatively rare, representing 9% of all non-Hodgkin’s lymphomas of the gastrointestinal tract.79 The pathologic classification of the lymphoma, essential for treatment, is dependent on morphologic, immunohistochemical, flow cytometric, and molecular biologic studies. Correct classification has important etiologic, prognostic, and therapeutic implications. The Western type of non-Hodgkin’s lymphoma is usually found in the distal small bowel. Most are classified as aggressive lymphomas under the World Health Organization (WHO) classification.80 Immuno-proliferative small intestinal disease (IPSID) is a lymphomatous disorder most commonly encountered in the Middle East and Africa, where small bowel lymphoma is more common than solid tumors of the small bowel. The opposite is true in Westernized countries.5 Patients with
396
GASTROINTESTINAL TUMORS
IPSID have lymphoid proliferation/neoplasm infiltrating the entire length of the small bowel and may present with severe malabsorption, abdominal pain, and nail clubbing.81 Prognosis of IPSID does not differ appreciably from non-IPSID small bowel lymphoma, with 5-year survivals slightly over 50%.82 Hodgkin’s disease of the small intestine is exceedingly rare.83 Most other lymphomas are B cell type of various histologies, though up to 25% are of T cell origin – most often in the setting of celiac disease.79 T cell lymphomas may arise several years after the initiation of the enteropathy and may present with exacerbation of diarrheal symptoms and/or ulcerated jejunal lesions. Mantle cell lymphoma of the small bowel, also termed “malignant lymphomatous polyposis”, typically involves the mucosa of the terminal ileum with subsequent involvement of other segments of the gut.84 Fluorescence in situ hybridization (FISH) assays for the Cyclin D1 (CCND1)/immunoglobulin H (IgH) translocation can distinguish mantle cell lymphoma from other lymphomas.85 The tumor is composed of small malignant lymphocytes that cluster in aggregates along the mucosa leading to polypoid lesions. Mantle cell lymphoma tends to behave more aggressively than other non-Hodgkin’s lymphomas. In contrast, MALT tumors may behave as very low-grade lesions.86 Small bowel involvement by this tumor is uncommon, with the more typical location being in the stomach. With further neoplastic transformation, such as p53 mutations, a more aggressive large B cell lymphoma may develop in the background of low-grade MALT lymphoma, and is treated similar to primary aggressive lymphomas. Posttransplant lymphoproliferative disorders (PTLD) usually present extranodally and may be classified as polymorphic or monomorphic depending on the cellular composition.87 Polymorphic lesions are composed of mixtures of small and large EBV (Epstein-Barr virus)-transformed B cells, and often respond to withdrawal of immune suppression. Monomorphic lesions are composed of sheets of large lymphoma cells. While these cells as well show evidence of EBV infection, withdrawal of immunosuppression is somewhat less likely to result in tumor regression. Burkitt’s lymphoma may also present with small bowel involvement. This lymphoma is highly aggressive, and treatment must be initiated quickly, as survival is measured in weeks without therapy.88 Treatment
The treatment of small bowel lymphoma is based on the histologic subtype of the disease. Patients have traditionally been treated with surgery, but because of poor survival even in early stage disease,89 systemic therapy is likely helpful. Low-grade MALT lymphoma may benefit from Helicobacter pylori eradication, but this is unproven, unlike in gastric MALT lymphoma. Since low-grade lymphomas are rarely curable, systemic therapy when patients are symptomatic would be a reasonable approach. Patients with aggressive intermediate and highly aggressive nonHodgkin’s lymphoma of the small bowel may be treated with combination chemotherapy (e.g. cyclophosphamide,
doxorubicin, vincristine, and prednisone (CHOP) ± rituximab) instead of surgical resection, as many patients treated surgically recur systemically.89 T cell histologies fare more poorly than B cell histologies.89 Mantle cell lymphoma patients also fare poorly. Do note that patients with transmural involvement probably have a small risk of bowel perforation as a result of chemotherapy. Because of the potential toxicity of abdominal radiotherapy, and the efficacy of chemotherapy, total abdominal radiation therapy is not frequently utilized. Management of Mediterranean lymphoma and enteropathy-associated T cell lymphoma frequently includes chemotherapy. Withdrawal of immune suppression, rituximab (anti-CD20 antibody therapy), and chemotherapy are among various options for treatment of PTLD of the small bowel.90 Burkitt’s and other highly aggressive lymphomas are managed with aggressive, high-intensity chemotherapy regimens including intrathecal therapy.91
Sarcomas and Gastrointestinal Stromal Tumors (GIST) Biology and Pathology
Sarcomas are malignant tumors derived from mesenchymal cells. They represent approximately 10% of all small bowel malignancies. The vast majority are gastrointestinal stromal tumors (GIST), but leiomyosarcomas and, rarely, other sarcomas arise in the small bowel. Tumors derived from the muscularis mucosa are the only true leiomyomas in the small bowel. Unlike the sarcomas, these are benign lesions, which are usually small (i.e. several millimeters in size) polypoid lesions. These lesions are composed of fascicles of spindle cells that stain strongly with desmin, indicative of their smooth muscle differentiation. GISTs probably arise from, or at least display differentiation toward, interstitial cells of Cajal. In acquired immunodeficiency syndrome (AIDS) patients, Kaposi sarcoma may develop in the small bowel, leading to gastrointestinal bleeding. These vascular tumors are derived from malignant transformation of endothelial cells. As such, they display a growth pattern and phenotype-resembling vessels. These tumors form ill-defined mucosal masses with anastomosing vascular spaces leading to fragility and easy bleeding. Two reports of clear cell sarcoma or melanoma of soft parts in the small bowel have been reported in the literature. This extremely rare tumor usually arises in association with tendons and aponeuroses of the distal extremities. The cells display melanocytic differentiation with occasional melanin pigment and immunohistochemical staining for S-100 and HMB-45. These tumors show a characteristic reciprocal translocation between chromosomes 12 and 17. The most important mesenchymal tumor of the small bowel is the GIST. On the basis of light microscopic findings, GISTs were initially thought to be neoplasms of smooth muscle origin, most often classified as leiomyosarcoma or leiomyoblastoma. After electron microscopic studies revealed ultrastructural inconsistencies with smooth muscle differentiation, a debate on the histopathologic origin of
CANCER OF THE SMALL BOWEL
GIST ensued. This debate was eventually settled by the realization that GIST represented a pathologically distinct entity with the common characteristic of CD34 and CD117 (c-kit) expression. Histologically, these tumors may display a surprising array of phenotypes from tight spindle cells with fasciculations (70% of cases) to epithelioid, loosely connected cells (20%) to a palisading paragangliomatous-like architecture. The overlap between smooth muscle and neural morphologic features led to the hypothesis that GIST are derived from intestinal cells of Cajal (ICC), the so-called pacemaker cells of gastrointestinal motility. Approximately 95% of GISTs stain positive for c-kit by immunohistochemistry. Staining for other markers is variable (Bcl-2 80%, CD34 70%, muscle-specific actin 50%, smooth muscle actin 35%, S-100 10%, desmin 5%). In a large series of 1004 patients with GIST,92 252 cases (25%) of primary small bowel GIST were identified, which made the small bowel the second most common primary location of GIST (see Table 1). In this series, size, mitotic index (>5 mitotic figures/50 high-power fields), patient age, and tumor location were the most important prognostic parameters in multivariate analysis. Small bowel and omentum/mesenteric GISTs had the poorest prognosis while patients with esophageal and stomach locations fared significantly better. It is important to note that these data were generated in the pre-imatinib era. Determining the malignant, metastatic potential of GIST is a clinical and pathological challenge. Up to 30% of newly diagnosed GISTs have overtly malignant characteristics. Tumor progression is mainly noted in the form of expansive growth or local recurrence after resection. Intra-abdominal spread along the serosal surface can occur and eventually liver metastases can develop. Lymph node metastases and metastatic spread outside the abdomen are uncommon. Importantly, tumor recurrence has been noted decades after curative resection even in low-risk GISTs. Approach to the Patient with Stromal Tumor of the Small Bowel
Patients with stromal tumors of the small bowel may present with bleeding, obstruction, pain, or perforation. A large, mobile mass may be palpated on examination. A small Table 1 Series from Emory, et al.92 describing anatomic locations and tumor sizes of 1004 patients with gastrointestinal stromal tumors (GIST).
Primary location of GIST Esophagus Stomach Small bowel Duodenum Jejunum Ileum Unspecified Large bowel Other (omentum/ peritoneum/ mesentery) Total Note: GIST, gastrointestinal stromal tumor.
Number of patients 53 524 252 45 68 33 106 108 67 1004
Mean size of tumor (cm) 3.9 6.7 7.1
5.5 11.3
397
bowel follow-through, enteroclysis, or abdominal CT/MRI scans may suggest the diagnosis of a small bowel tumor. Radiographically, GISTs are usually large, soft tissue masses that splay apart and extrinsically compress adjacent small bowel loops. The segment of small bowel from which the exophytic mass arises is abnormal. The flattened mucosa overlying the tumor may be ulcerated or preserved. CT may demonstrate a large soft tissue mass, often with areas of low attenuation where cavitation occurs. With the use of intravenous contrast, the tumor will show inhomogeneous but dense enhancement in the noncavitary areas. Frequently, the histologic diagnosis is not made until surgery. The primary treatment of stromal tumors of the small bowel is surgery. Wide resection of the primary tumor with an attempt at obtaining clear margins is mandatory. Because lymph node involvement is uncommon, regional lymphadenectomy is not necessary. While conventional chemotherapy has shown to be largely ineffective in GIST, the discovery that imatinib, a small-molecule tyrosine kinase inhibitor, has substantial activity against GIST has dramatically changed management and prognosis of patients with GIST since 2000. Imatinib (Gleevec/Glivec; Novartis Pharma, Basel, Switzerland) was initially specifically developed as an oral tyrosine kinase inhibitor of the bcr-abl gene product, the molecular hallmark of chronic myelogenous leukemia (CML). Imatinib, however, also inhibits other tyrosine kinase domains such as ARG (ABL-related gene), PDGFR-A/B (platelet-derived growth factor receptor), and KIT. The inhibition of wild-type and a mutant form of KIT commonly found in GIST made imatinib a promising agent in this disease. Subsequent clinical trials confirmed the substantial clinical activity of imatinib with partial responses in the range of 50–60% at a recommended dose of 400–800 mg daily and disease progression as best response only in about 15% of patients. On the basis of these findings, imatinib has emerged as FDA (Food and Drug Administration)-approved, standard palliative therapy for patients with unresectable or metastatic GIST. Data on long-term follow-up of patients on imatinib are inevitably premature at the time of this writing (September 2005), but median time to imatinib failure was described as 18–26 months. The role of imatinib as adjuvant, postoperative treatment after curative resection is currently being investigated in clinical trials. Most recently, initial and late resistance to imatinib has been correlated with mutations in PDGFRA in tumors lacking activating KIT mutations, which are commonly associated with an epithelioid phenotype. In tumors progressing on imatinib, the use of sunitinib (SU11248), an inhibitor of mutant PDGFRA tyrosine kinase, was found to be associated with significantly prolonged progression-free and overall survival compared with placebo in a phase III trial.93 For additional information on GIST, see Chapter 37, Gastrointestinal Stromal Tumors devoted to this topic. Approach to the Patient with non-GIST Sarcoma of the Small Bowel
Non-GIST sarcomas of the small intestines such as leiomyosarcomas, angiosarcomas, and malignant fibrous histocytomas are much less common than GIST so that
398
GASTROINTESTINAL TUMORS
only case reports or small case series exist and no specific validated treatment recommendations are available. Non-GIST sarcomas – with the exception of granulocytic sarcomas (see below) – should be treated according to guidelines developed for nonintestinal sarcomas.94 The mainstay of therapy is R0 resection of the primary tumor. The use of adjuvant therapy is debatable and should be considered if high-risk factors such as large tumors or histopathologic features suggesting aggressive tumor biology are present.95 Palliative chemotherapy for metastatic disease includes ifosfamide and doxorubicin-based regimens. For leiomyosarcomas, a gemcitabine or gemcitabine/docetaxel combination should be considered.96 Granulocytic sarcomas are rare extramedullary tumors composed of immature myeloid cells. Chemotherapy regimens analogous to the treatment of acute myelogenous leukemia have been successfully used in this rare entity.97
Metastatic Disease to the Small Bowel Biology and Pathology
The small bowel may be involved with metastatic disease. Metastases may arise by hematologic or lymphatic spread, intraperitoneal spread, or direct invasion into small bowel. The most common sites of hematogenous metastases are jejunum and ileum while duodenal small bowel metastases are very rare.98,99 Hematogenous metastases may cause bleeding because of ulceration overlying a submucosal lesion. Obstruction caused by intussusception may result in abdominal pain or distension. Intraperitoneal metastases or tumors that directly spread to the small intestine primarily cause obstruction. Gastrointestinal perforations secondary to small bowel metastases are infrequent, but have been observed in various advanced solid tumors.98,100 Malignant melanoma is the most common tumor that hematogenously metastasizes to the small bowel.101 It is estimated that over half of the patients who die of melanoma have intestinal metastases, but a clinical diagnosis is only made in 5% of melanoma patients.101,102 Other tumors that metastasize to the small bowel by the hematogenous route include the lung, breast, and renal cell carcinoma. Radiographically, hematogenous metastases appear as relatively flat or polypoid, well-circumscribed, smooth-surfaced submucosal masses, with abrupt angulation with the luminal contour. Central ulceration or umbilication is seen in about one-half of submucosal metastases, forming a so-called “target” or “bull’s-eye” lesion. Locally advanced tumors such as colorectal or pancreatic carcinomas may involve the small bowel by direct invasion or extension through mesenteric planes. The small intestine lies in the inframesocolic compartment of the peritoneal cavity, below the transverse mesocolon. Any malignancy arising in an organ bordering the peritoneal cavity can therefore secondarily seed the peritoneum and spread to the small intestine. The most common malignancies to spread to the small intestine via the intraperitoneal route are ovarian, colorectal, pancreatic, and gastric carcinomas. Tumors that spread to retroperitoneal and small bowel mesenteric lymph nodes, such as breast cancer, can also spread via the intraperitoneal route. Fluid in the peritoneal cavity flows into the pelvis and pools in the right
lower quadrant small bowel mesentery, the sigmoid mesentery, the right paracolic gutter, and the rectovesical space in males and the rectouterine space in females. Therefore, the most likely sites of intraperitoneal metastases are the anterior border of the rectosigmoid junction, the sigmoid colon, the distal small intestine, and the ascending colon. Radiographically, intraperitoneal metastases are manifested as an extrinsic mass effect pushing on the bowel lumen. The bowel contour appears spiculated or irregular because of a desmoplastic reaction related to direct invasion of the serosa or adjacent mesenteric fat. En face, the bowel folds may appear tethered toward the neoplastic process. The radiographic appearance is similar to peritonitis or endometriosis secondarily involving the small intestine. While CT is the best examination to detect the presence of ascites and implants on the surface of the peritoneum, barium studies are probably more sensitive in detecting small bowel involvement by intraperitoneal metastases. Video capsule endoscopy is a new diagnostic tool that can be useful in detecting small bowel lesions.103 Approach to the Patient with Small Bowel Metastasis
Metastatic tumors to the small bowel may result in symptoms, including bleeding, obstruction, and perforation. These lesions may be diagnosed radiographically or at exploratory surgery. Surgical resection of a small bowel metastasis that causes symptoms or diverting small bowel anastomosis may provide important palliation. However, this approach may be problematic in patients with extensive, multifocal metastatic disease. Angiographic embolization and endoscopic laser therapy may be utilized to decrease bleeding from small bowel metastases. In a woman presenting with peritoneal carcinomatosis of unknown primary, which frequently may involve the small bowel, following guidelines for the treatment of advanced ovarian carcinoma may be beneficial. Systemic and, in rare instances, intraperitoneal chemotherapy may provide palliative benefit in some patients with intra-abdominal carcinomatosis with small bowel involvement. Photodynamic therapy is an investigational approach being utilized to treat patients with intra-abdominal malignancies.
REFERENCES 1. Herbsman H, et al. Tumors of the small intestine. Curr Probl Surg 1980; 17(3): 121 – 82. 2. Norberg KA, Emas S. Primary tumors of the small intestine. Am J Surg 1981; 142(5): 569 – 73. 3. Weiss NS, Yang CP. Incidence of histologic types of cancer of the small intestine. J Natl Cancer Inst 1987; 78(4): 653 – 6. 4. Ryan JC. Premalignant conditions of the small intestine. Semin Gastrointest Dis 1996; 7(2): 88 – 93. 5. Neugut A, et al. The epidemiology of cancer of the small bowel. Cancer Epidemiol Biomarkers Prev 1998; 7(3): 243 – 51. 6. Lynch HT, et al. Genetics, natural history, tumor spectrum, and pathology of hereditary nonpolyposis colorectal cancer: an updated review. Gastroenterology 1993; 104(5): 1535 – 49. 7. Green PH, et al. Risk of malignancy in patients with celiac disease. Am J Med 2003; 115(3): 191 – 5. 8. Ito H, et al. Surgical treatment of small bowel cancer: a 20-year single institution experience. J Gastrointest Surg 2003; 7(7): 925 – 30.
CANCER OF THE SMALL BOWEL 9. Nordkild P, Kjaergard H. Primary tumours of the small intestine – the diagnostic problem. Ann Chir Gynaecol 1986; 75(1): 31 – 6. 10. Bessette JR, et al. Primary malignant tumors in the small bowel: a comparison of the small-bowel enema and conventional followthrough examination. AJR Am J Roentgenol 1989; 153(4): 741 – 4. 11. Hara AK, et al. Small bowel: preliminary comparison of capsule endoscopy with barium study and CT. Radiology 2004; 230(1): 260 – 5. 12. Saurin JC, et al. Clinical impact of capsule endoscopy compared to push enteroscopy: 1-year follow-up study. Endoscopy 2005; 37(4): 318 – 23. 13. Moertel CG. Karnofsky memorial lecture. An odyssey in the land of small tumors. J Clin Oncol 1987; 5(10): 1502 – 22. 14. Younes N, et al. The presence of K-12 ras mutations in duodenal adenocarcinomas and the absence of ras mutations in other small bowel adenocarcinomas and carcinoid tumors. Cancer 1997; 79(9): 1804 – 8. 15. Oshima M, et al. Evidence against dominant negative mechanisms of intestinal polyp formation by Apc gene mutations. Cancer Res 1995; 55(13): 2719 – 22. 16. Neugut AI, et al. An overview of adenocarcinoma of the small intestine. Oncology (Huntington) 1997; 11(4): 529 – 36, discussion 545. 17. Smits R, et al. Loss of Apc and the entire chromosome 18 but absence of mutations at the Ras and Tp53 genes in intestinal tumors from Apc1638N, a mouse model for Apc-driven carcinogenesis. Carcinogenesis 1997; 18(2): 321 – 7. 18. Perzin KH, Bridge MF. Adenomas of the small intestine: a clinicopathologic review of 51 cases and a study of their relationship to carcinoma. Cancer 1981; 48(3): 799 – 819. 19. Sutter T, et al. Frequent K-ras mutations in small bowel adenocarcinomas. Dig Dis Sci 1996; 41(1): 115 – 8. 20. Howe JR, et al. The American College of Surgeons Commission on Cancer and the American Cancer Society. Adenocarcinoma of the small bowel: review of the National Cancer Data Base, 1985 – 1995. Cancer 1999; 86(12): 2693 – 706. 21. Fresko D, et al. Early presentation of carcinoma of the small bowel in Crohn’s disease (“Crohn’s carcinoma”). Case reports and review of the literature. Gastroenterology 1982; 82(4): 783 – 9. 22. Collier PE, Turowski P, Diamond DL. Small intestinal adenocarcinoma complicating regional enteritis. Cancer 1985; 55(3): 516 – 21. 23. Cruz DN, Huot SJ. Metabolic complications of urinary diversions: an overview. Am J Med 1997; 102(5): 477 – 84. 24. Attanoos R, et al. Ileostomy polyps, adenomas, and adenocarcinomas. Gut 1995; 37(6): 840 – 4. 25. Church JM, et al. The relationship between fistulas in Crohn’s disease and associated carcinoma. Report of four cases and review of the literature. Dis Colon Rectum 1985; 28(5): 361 – 6. 26. Bernstein D, Rogers A. Malignancy in Crohn’s disease. AmJ Gastroenterol 1996; 91(3): 434 – 40. 27. Rhodes JM. Unifying hypothesis for inflammatory bowel disease and associated colon cancer: sticking the pieces together with sugar. Lancet 1996; 347(8993): 40 – 4. 28. O’Riordan BG, Vilor M, Herrera L. Small bowel tumors: an overview. Dig Dis 1996; 14(4): 245 – 57. 29. Straker RJ, Gunasekaran S, Brady PG. Adenocarcinoma of the jejunum in association with celiac sprue. J Clin Gastroenterol 1989; 11(3): 320 – 3. 30. Houlston RS, Ford D. Genetics of coeliac disease. QJM 1996; 89(10): 737 – 43. 31. Holmes GK, et al. Malignancy in coeliac disease – effect of a gluten free diet. Gut 1989; 30(3): 333 – 8. 32. Cooper BT, et al. Celiac disease and malignancy. Medicine 1980; 59(4): 249 – 61. 33. Potter DD, et al. The role of defective mismatch repair in small bowel adenocarcinoma in celiac disease. Cancer Res 2004; 64(19): 7073 – 7. 34. Soravia C, Bapat B, Cohen Z. Familial adenomatous polyposis (FAP) and Hereditary Nonpolyposis Colorectal Cancer (HNPCC): a review of clinical, genetic and therapeutic aspects. Schweizerische Medizinische Wochenschrift. J Suisse de Medecine 1997; 127(16): 682 – 90.
399
35. Vasen HF, et al. Cancer risk in families with hereditary nonpolyposis colorectal cancer diagnosed by mutation analysis. [erratum appears in Gastroenterology 1996; 111(5):1402]. Gastroenterology 1996; 110(4): 1020 – 7. 36. de la Chapelle A. Genetic predisposition to colorectal cancer. Nat Rev Cancer 2004; 4(10): 769 – 80. 37. Desai DC, et al. Juvenile polyposis. Br J Surg 1995; 82(1): 14 – 7. 38. Spigelman AD, Arese P, Phillips RK. Polyposis: the Peutz-Jeghers syndrome. [see comment]. Br J Surg 1995; 82(10): 1311 – 4. 39. Contant CM, et al. Prognostic value of the TNM-classification for small bowel cancer. Hepatogastroenterology 1997; 44(14): 430 – 4. 40. Barloon TJ, et al. Primary adenocarcinoma of the duodenal bulb: radiographic and pathologic findings in two cases. Gastrointest Radiol 1989; 14(3): 223 – 5. 41. Nix GA, Wilson JH, Dees J. Primary malignant tumours of the duodenum. Rofo: Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin 1985; 142(4): 385 – 90. 42. Levine MS, Drooz AT, Herlinger H. Annular malignancies of the small bowel. Gastrointest Radiol 1987; 12(1): 53 – 8. 43. Rubesin SE, et al. Non-Hodgkin lymphoma of the small intestine. Radiographics 1990; 10(6): 985 – 98. 44. Lai EC, et al. Primary adenocarcinoma of the duodenum: analysis of survival. World J Surg 1988; 12(5): 695 – 9. 45. Joesting DR, et al. Improving survival in adenocarcinoma of the duodenum. Am J Surg 1981; 141(2): 228 – 31. 46. Alwmark A, Andersson A, Lasson A. Primary carcinoma of the duodenum. Ann Surg 1980; 191(1): 13 – 8. 47. Michelassi F, et al. Experience with 647 consecutive tumors of the duodenum, ampulla, head of the pancreas, and distal common bile duct. Ann Surg 1989; 210(4): 544 – 54, discussion 554 – 6. 48. Ouriel K, Adams JT. Adenocarcinoma of the small intestine. Am J Surg 1984; 147(1): 66 – 71. 49. Dabaja BS, et al. Adenocarcinoma of the small bowel: presentation, prognostic factors, and outcome of 217 patients. Cancer 2004; 101(3): 518 – 26. 50. Klinkenbijl JH, et al. 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. [see comment]. Ann Surg 1999; 230(6): 776 – 82, discussion 782 – 4. 51. Witham M, Harnett PR. Adenocarcinoma of the duodenum with liver metastases. Complete remission and long-term survival with 5-fluorouracil chemotherapy – a case report. Am J Clin Oncol 1996; 19(3): 305 – 6. 52. Jigyasu D, Bedikian AY, Stroehlein JR. Chemotherapy for primary adenocarcinoma of the small bowel. Cancer 1984; 53(1): 23 – 5. 53. Crawley C, et al. The Royal Marsden experience of a small bowel adenocarcinoma treated with protracted venous infusion 5fluorouracil. Br J Cancer 1998; 78(4): 508 – 10. 54. Copson ER, et al. Epirubicin, cisplatin and infusional 5-fluorouracil chemotherapy for locally advanced and metastatic small bowel adenocarcinoma. Proc Am Soc Clin Oncol 2002; 21: 133b, abstract #2347. 55. Yeung RS, et al. Neoadjuvant chemoradiation in pancreatic and duodenal carcinoma. A Phase II Study. Cancer 1993; 72(7): 2124 – 33. 56. Coia L, et al. Preoperative chemoradiation for adenocarcinoma of the pancreas and duodenum. Int J Radiat Oncol Biol Phys 1994; 30(1): 161 – 7. 57. Barnes G, et al. Primary adenocarcinoma of the duodenum: management and survival in 67 patients. Ann Surg Oncol 1994; 1(1): 73 – 8. 58. Gibson MK, et al. Phase II study of 5-fluorouracil, doxorubicin, and mitomycin C for metastatic small bowel adenocarcinoma. Oncologist 2005; 10(2): 132 – 7. 59. Cunningham JD, et al. Malignant small bowel neoplasms: histopathologic determinants of recurrence and survival. Ann Surg 1997; 225(3): 300 – 6. 60. Vortmeyer AO, et al. Concordance of genetic alterations in poorly differentiated colorectal neuroendocrine carcinomas and associated adenocarcinomas. J Natl Cancer Inst 1997; 89(19): 1448 – 53.
400
GASTROINTESTINAL TUMORS
61. Pantongrag-Brown L, et al. Calcification and fibrosis in mesenteric carcinoid tumor: CT findings and pathologic correlation. AJR Am J Roentgenol 1995; 164(2): 387 – 91. 62. Modlin IM, Shapiro MD, Kidd M. Carcinoid tumors and fibrosis: an association with no explanation. Am J Gastroenterol 2004; 99(12): 2466 – 78. 63. Petrik PK. Fatal small intestinal infarction due to occlusion by mesenteric carcinoid tumor. Am J Forensic Med Pathol 1989; 10(2): 146 – 8. 64. Strobbe L, et al. Ileal carcinoid tumors and intestinal ischemia. Hepatogastroenterology 1994; 41(5): 499 – 502. 65. Harvey JN, Denyer ME, DaCosta P. Intestinal infarction caused by carcinoid associated elastic vascular sclerosis: early presentation of a small ileal carcinoid tumour. Gut 1989; 30(5): 691 – 4. 66. Iozzo RV. Case 16-1979 – elastic vascular sclerosis and carcinoid tumors. N Engl J Med 1979; 301(7): 385 – 6. 67. Anthony PP, Drury RA. Elastic vascular sclerosis of mesenteric blood vessels in argentaffin carcinoma. J Clin Pathol 1970; 23(2): 110 – 8. 68. Thorsen ABG, Bjorkman G, Waldenstrom J. Malignant carcinoid of the small intestine with metastases to the liver, valvular disease of the right side of the heart (pulmonary stenosis and tricuspid regurgitation without septal defects), peripheral vasomotor symptoms, bronchoconstriction and an Unusual Type of Cyanosis. Am Heart J 1954; 47: 795 – 817. 69. Jeffree MA, Nolan DJ. Multiple ileal carcinoid tumours. Br J Radiol 1987; 60(712): 402 – 3. 70. Gore RM, et al. GI carcinoid tumours: appearance of the primary and detecting metastases. Best Pract Res Clin Endocrinol Metab 2005; 19(2): 245 – 63. 71. Marsh HM, et al. Carcinoid crisis during anesthesia: successful treatment with a somatostatin analogue. Anesthesiology 1987; 66(1): 89 – 91. 72. Kvols LK, et al. Rapid reversal of carcinoid crisis with a somatostatin analogue. N Engl J Med 1985; 313(19): 1229 – 30. 73. Kinney MA, et al. Perianaesthetic risks and outcomes of abdominal surgery for metastatic carcinoid tumours. Br J Anaesth 2001; 87(3): 447 – 52. 74. Que FG, et al. Hepatic resection for metastatic neuroendocrine carcinomas. Am J Surg 1995; 169(1): 36 – 42, discussion 42 – 3. 75. Moertel CG, et al. The management of patients with advanced carcinoid tumors and islet cell carcinomas. Ann Intern Med 1994; 120(4): 302 – 9. 76. Moertel CG, Rubin J, Kvols LK. Therapy of metastatic carcinoid tumor and the malignant carcinoid syndrome with recombinant leukocyte A interferon. J Clin Oncol 1989; 7(7): 865 – 8. 77. Kulke M, et al. A phase 2 study to evaluate the efficacy and safety of SU11248 in patients (pts) with unresectable Neuroendocrine Tumors (NETs). J Clin Oncol (Meeting Abstracts) 2005; 23(Suppl 16): 4008. 78. Yao JC, et al. Improved Progression Free Survival (PFS), and rapid, sustained decrease in tumor perfusion among patients with advanced carcinoid treated with bevacizumab. J Clin Oncol (Meeting Abstracts) 2005; 23(Suppl 16): 4007. 79. Koch P, et al. Primary gastrointestinal non-Hodgkin’s lymphoma: I. Anatomic and histologic distribution, clinical features, and survival data of 371 patients registered in the German Multicenter Study GIT NHL 01/92. J Clin Oncol 2001; 19(18): 3861 – 73. 80. Harris NL, et al. World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: report of the Clinical Advisory Committee Meeting-Airlie House, Virginia, November 1997. [see comment]. J Clin Oncol 1999; 17(12): 3835 – 49.
81. Salem P, et al. Primary small intestinal lymphoma in adults. A comparative study of IPSID versus non-IPSID in the Middle East. Cancer 1987; 59(9): 1670 – 6. 82. Shih LY, et al. Primary small-intestinal lymphomas in Taiwan: immunoproliferative small-intestinal disease and nonimmunoproliferative small-intestinal disease. [see comment]. J Clin Oncol 1994; 12(7): 1375 – 82. 83. Morgan PB, et al. Uncommon presentations of Hodgkin’s Disease: CASE 1. Hodgkin’s disease of the jejunum. J Clin Oncol 2004; 22(1): 193 – 5. 84. Moynihan MJ, et al. Lymphomatous polyposis. A neoplasm of either follicular mantle or germinal center cell origin. Am J Surg Pathol 1996; 20(4): 442 – 52. 85. Remstein ED, et al. Diagnostic utility of fluorescence in situ hybridization in mantle-cell lymphoma. Br J Haematol 2000; 110(4): 856 – 62. 86. Wotherspoon AC. Gastric MALT lymphoma and Helicobacter pylori. Yale J Biol Med 1996; 69(1): 61 – 8. 87. Basgoz N, Preiksaitis JK. Post-transplant lymphoproliferative disorder. Infect Dis Clin North Am 1995; 9(4): 901 – 23. 88. Carbone PP. Non-Hodgkin’s lymphoma: recent observations on natural history and intensive treatment. Cancer 1972; 30(6): 1511 – 6. 89. Radaszkiewicz T, Dragosics B, Bauer P. Gastrointestinal malignant lymphomas of the mucosa-associated lymphoid tissue: factors relevant to prognosis. Gastroenterology 1992; 102(5): 1628 – 38. 90. Ghobrial IM, et al. Prognostic factors in patients with Post-transplant Lymphoproliferative Disorders (PTLD) in the rituximab era. Leuk Lymphoma 2005; 46(2): 191 – 6. 91. Rizzieri DA, et al. Intensive chemotherapy with and without cranial radiation for Burkitt leukemia and lymphoma: final results of Cancer and Leukemia Group B Study 9251. Cancer 2004; 100(7): 1438 – 48. 92. Emory TS, et al. Prognosis of gastrointestinal smooth-muscle (stromal) tumors: dependence on anatomic site. Am J Surg Path 1999; 23(1): 82 – 7. 93. Demetri GD, et al. Phase 3, multicenter, randomized, double-blind, placebo-controlled trial of SU11248 in patients (pts) following failure of imatinib for metastatic GIST. J Clin Oncol (Meeting Abstracts) 2005; 23(Suppl 16): 4000. 94. Clark MA, et al. Soft-tissue sarcomas in adults. N Engl J Med 2005; 353(7): 701 – 11. 95. Casali PG, Picci P. Adjuvant chemotherapy for soft tissue sarcoma. Curr Opin Oncol 2005; 17(4): 361 – 5. 96. Hensley ML, et al. Gemcitabine and docetaxel in patients with unresectable leiomyosarcoma: results of a phase II trial. J Clin Oncol 2002; 20(12): 2824 – 31. 97. Xavier SG, et al. Granulocytic sarcoma of the small intestine with CBFbeta/MYH11 fusion gene: report of an aleukaemic case and review of the literature. Leuk Res 2003; 27(11): 1063 – 6. 98. Garwood RA, et al. A case and review of bowel perforation secondary to metastatic lung cancer. Am Surg 2005; 71(2): 110 – 6. 99. Loualidi A, et al. Duodenal metastasis: an uncommon cause of occult small intestinal bleeding. Neth J Med 2004; 62(6): 201 – 5. 100. Shiraishi M, et al. Perforation due to metastatic tumors of the ileocecal region. World J Surg 1998; 22(10): 1065 – 8. 101. de la Monte SM, Moore GW, Hutchins GM. Patterned distribution of metastases from malignant melanoma in humans. Cancer Res 1983; 43(7): 3427 – 33. 102. Ihde JK, Coit DG. Melanoma metastatic to stomach, small bowel, or colon. Am J Surg 1991; 162(3): 208 – 11. 103. Raju GS, Nath SK. Capsule endoscopy. Curr Gastroenterol Rep 2005; 7(5): 358 – 64.
Section 6 : Gastrointestinal Tumors
35
Unusual Tumors of the Colon, Rectum and Anus William P. Tew and Leonard B. Saltz
INTRODUCTION Colorectal and anal cancers, with more than 150 000 new cases annually in the United States alone, are the fourth most common malignancy and the second leading cause of cancerrelated death in the United States.1 In one population-based study in the United States, adenocarcinoma accounted for 97% of all large bowel cancers, with the remaining 3% being cancers of histologic varieties other than adenocarcinoma (see Table 1).2 Of these 3%, about one-half were anal carcinomas (squamous and cloacogenic) and one-third were carcinoid, or neuroendocrine tumors. The rest were tumors rare to the large intestine, such as lymphoma, sarcoma, and melanoma. These nonadenocarcinoma tumors of the colorectal and anal region have distinctive epidemiologies, clinical traits, and pathologies. Treatment and prognoses of these tumors vary greatly, and will be discussed in this chapter.
CARCINOID TUMORS OF THE COLON AND RECTUM Carcinoid tumors, arising from mucosal neuroendocrine cells, are rare tumors with an overall incidence estimated to be fewer than two cases per 100 000 people in the United States and most commonly involve the lungs/bronchi (30%) and gastrointestinal (GI) tract (55–70%).3,4 In several series with over 10 000 cases identified from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute between 1973 and 1999, carcinoid tumors of the colorectum were extremely rare, accounting for less than a quarter of all carcinoid tumors.4 – 6 In the last decade, the incidence of all carcinoid tumors has increased by 3% every year, likely because of improved diagnostic imaging with a higher detection of these indolent tumors.6 GI carcinoid tumors are classified according to their presumed derivation from different embryonic divisions of the gut and their biologic characteristics vary according to location.3 While about half of all GI carcinoids are
functional (produce a hormone), most of these functional carcinoids arise in the small bowel or appendix (midgut). When carcinoid tumors do make a hormone, serotonin is by far the most common one, and this is best detected by measuring the serotonin breakdown product 5-hydroxyindoleacetic acid (5-HIAA), which is excreted in the urine. Other hormones, such as histamine, dopamine, substance P, prostaglandins, bradykinin, and corticotropin are also produced but are exceedingly rare.3 Rectal carcinoids may contain glucagon and glicentin-related peptides, rather than serotonin.7 However, the majority of the carcinoid tumors arising in the colon and rectum (hind-gut) are nonfunctional and do not produce a hormone. Thus, the carcinoid syndrome of episodic diarrhea, flushing, and wheezing, and eventual endomyocardial fibrosis leading to right-sided valvular heart disease is highly unusual with colorectal carcinoid.
Pathology Neuroendocrine tumors comprise a spectrum of diseases, from well-differentiated (classic carcinoid) to anaplastic neuroendocrine carcinomas (“oat cell”, small cell carcinoma). The typical carcinoid tumors are composed of relatively small, bland, homogeneous cells with regular, well-rounded homogeneous nuclei with few nucleoli. Consistent with the indolent nature of classic carcinoid tumors, they are typically mitotically inactive as illustrated in Figure 1. In contrast, tumors with increased nuclear atypia, high mitotic activity, or areas of necrosis have been called anaplastic or poorly differentiated. The distinguishing feature of carcinoid and other neuroendocrine tumors is the presence of neurosecretory granules, composed of a variety of hormones and biogenic amines. Histologically, these are demonstrated by silverbased stains and neuroendocrine tissue markers, such as chromogranin, neuron-specific enolase (NSE), and synaptophysin (see Figure 2). NSE is less specific, and by itself should not be accepted as proof of the diagnosis. In equivocal cases when the diagnosis remains in doubt, electron
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
402
GASTROINTESTINAL TUMORS
Table 1 Summary of large bowel cancer (nonadenocarcinoma).
Histology Squamous cell carcinoma Malignant carcinoid Transitional cell – like/cloacogenic Lymphoma Sarcoma Melanoma
Percentage (%) 33 33 16 11 4 1
Figure 1 Typical carcinoid – high-power under light microscopy showing typical neuroendocrine cytology without mitosis or necrosis (Courtesy of Dr Jinru Shia).
Figure 3 Electron microscopy of neurosecretory granules in a carcinoid tumor (Courtesy of Dr Jinru Shia).
can become intermixed with carcinoid tumor cells, simulating an adenocarcinoma. Carcinoembryonic antigen (CEA) is not helpful in distinguishing the two subtypes, since nearly a quarter of the large bowel carcinoid tumors can stain positively for CEA.8
Clinical Presentation
Figure 2 Neuroendocrine tissue markers include chromogranin, NSE, and synaptophysin. Synaptophysin-stained cells are seen below (Courtesy of Dr Jinru Shia).
microscopy may be useful in identifying neurosecretory granules as illustrated in Figure 3. Small cell carcinomas, being high grade and poorly differentiated, are defined, strictly morphologically, as neuroendocrine markers that are often stain negative. Neuroendocrine tumors are of epithelial origin, and thus stain positive for cytokeratins. In rare cases, colonic glands
Within the GI tract, the small intestine is the most common site (45%), followed by the rectum (20%), appendix (17%), colon (11%), and stomach (7%).4 Carcinoid tumors of the colorectum and appendix often have no symptoms until and unless they become quite large, a process which can take many years. As such, they are often incidental findings during an endoscopic, radiographic, or surgical evaluation of an unrelated symptom. Rectal carcinoids account for up to 2% of all rectal tumors, usually detected on routine endoscopy or rectal examination in the sixth decade of life. Patients who have symptoms may present with rectal bleeding, pain, or constipation; carcinoid syndrome is rare.9,10 Similarly, patients with colon carcinoids rarely (less than 5%) present with carcinoid syndrome. Patients with colon carcinoids are usually in their seventh decade of life and may have symptoms of abdominal pain, anorexia, heme occult positive stool, or weight loss, especially with larger tumors.11 Two-thirds of colon carcinoids are found in the right side of the colon, mostly in the cecum.11
UNUSUAL TUMORS OF THE COLON, RECTUM AND ANUS
Appendiceal carcinoid is the most common tumor of the appendix, usually detected as an incidental finding on appendectomy. In fact, older literature suggests that about one of every 200–300 appendectomy specimens is a carcinoid tumor.12 Since incidental appendectomies were performed more often on women and younger patients, these studies have also shown a higher incidence of appendiceal carcinoid in women and in those in the fourth or fifth decade of life.3 In a more recent review of the SEER database analysis, however, there is no longer a significant gender difference with appendiceal carcinoids.5 The majority of appendiceal carcinoids are asymptomatic with less than 10% causing obstructive symptoms, since approximately 75% are located in the distal third of the appendix.13,14 However, in a recent retrospective study, the rate of “incidental” carcinoid tumor was less common with a majority of patients (54%) presenting with signs and symptoms of an acute appendicitis.15 For more detail regarding appendiceal carcinoids, the reader is referred to Chapter 37, Gastrointestinal Stromal Tumors.
Management For carcinoid tumors that are local–regionally confined at the time of diagnosis, surgery is the treatment of choice. In appendiceal carcinoid tumors, over 95% are smaller than 2 cm and can be removed by simple appendectomy. Such surgeries are almost always curative, approaching 100% 10 years disease-free survival in one large series.14 Lesions greater than 2 cm in diameter carry a higher risk for nodal or distant metastasis. Since the lymphatic drainage of the appendix is along the right mesocolon, a formal right hemicolectomy is required to accomplish an adequate regional lymphadenectomy for these larger tumors. Lesions between 1 and 2 cm in size are more controversial, with no definitive data to support a better outcome with a hemicolectomy, and with some investigators favoring this more-aggressive approach. A similar surgical strategy is employed with rectal carcinoids. Local excision alone is adequate therapy for lesions less than 1 cm in diameter, which represent two-thirds of rectal carcinoids.3 In lesions greater than 2 cm, spread to local–regional lymph nodes is higher, and a formal lowanterior or abdominal-peritoneal resection should be considered. However, in several retrospective studies, these moreaggressive surgical procedures do not appear to improve survival rates.16 – 18 The treatment of tumors 1–2 cm in size is again somewhat controversial. Local excision may be more appropriate for tumors without muscular invasion or closer to 1 cm in size, for asymptomatic patients or with poor surgical outcomes patients, or for those who would require an abdominal-peritoneal resection, and thus, a permanent colostomy. Colon carcinoid tumors are treated similar to rectal tumors. However, carcinoid tumors of the colon are more likely to present with larger tumors, with the average colon tumor diameter of 5 cm at presentation. Over two-thirds of patients have either nodal or distant metastasis at presentation.11 Therefore, most patients require standard hemicolectomy. For the rare patient with a tumor smaller than 1 cm, local excision has been reported to be effective.19
403
After surgical resection of a carcinoid tumor, patients are monitored for symptomatic recurrence of disease. Specific follow-up guidelines are not well-described; however, given its indolent nature, routine imaging studies are often unnecessary. Computerized tomography (CT) or octreotide scans may be warranted if new symptoms arise, particularly with high-risk tumors. Lastly, there is no known role for adjuvant therapy, and no evidence that any such therapy reduces the risk of recurrence.
Metastatic Disease Patients with nonfunctional metastatic carcinoid often present with few symptoms and an excellent clinical performance status. Intra-abdominal or hepatic metastases may even present as an incidental finding of an asymptomatic mass or of hepatomegaly on routine physical examination. Because of the unusually slow tempo of the typical carcinoid tumor, metastatic disease per se does not necessarily constitute an indication for therapeutic intervention. Antineoplastic treatment need not be instituted unless there is pain or significant discomfort due to tumor bulk, uncontrollable hormonal symptoms caused by a functional tumor, or clear evidence of rapid progression of disease under observation. Carcinoid tumors typically express cell surface receptors for the regulatory hormone somatostatin. Six types of somatostatin receptors have been identified. Octreotide, which is a long-acting synthetic analogue of native somatostatin, inhibits the secretion of multiple hormones, including growth hormones, insulin, glucagon, and gastrin.20 It binds to predominantly type II somatostatin receptors, expressed on more than 80% of carcinoid tumors.21 Both short- and long-acting formulations of octreotide, along with a related peptide lanreotide, have been shown to be effective therapeutic agents in the symptomatic control of carcinoid syndrome in up to 90% of patients. Actual radiographic tumor responses to octreotide occur anecdotally, but are extremely rare.22 – 26 Radioactive-labeled somatostatin analogues are being developed as potential therapeutic agents to attempt to improve these responses, though early reports note increased hematologic and renal toxicities.27,28 At the time of writing this article, these agents are not commercially available in the United States. Diagnostically, indium 111-labeled octreotide (111 In Pentetreotide, OctreoScan) has been demonstrated to be useful for localization of somatostatin-receptor positive tumors, with a sensitivity of 80–90%, as well as a predictor to somatostatin treatment.29 This technique is used both to locate occult primary tumors in patients with suspected carcinoid tumor on the basis of symptoms and with no demonstrated metastatic disease, and to establish the extent of metastatic disease in patients with identified carcinoid or other neuroendocrine primary tumors. 111 In-labeled octreotide scans are particularly appropriate when considering use of octreotide in patients with a nonfunctional carcinoid tumor, since a negative scan would suggest that octreotide therapy is without rational basis and is highly unlikely to be beneficial. Other radiographic techniques such as metaiodobenzylguanidine (MIBG) scan, bone scintigraphy, and CT scan can be utilized for tumor localization. Positron-emission tomography (PET)
404
GASTROINTESTINAL TUMORS
scan with 18-FDG is limited, because of the low proliferative activity and high differentiation rate in carcinoid tumors, thus routine FDG PET should not be routinely used in the management of carcinoid tumors.28 However, from an experimental perspective, PET scans with various tracers such as 5-hydroxytryptophan (HTP) are under investigation.30 In addition to somatostatin analogues, interferon α (IFN-α) has been reported to be effective in controlling hormonal symptoms, and may have some limited usefulness in producing objective regressions of bulky disease. However, there was no clear synergy effect in combining IFN with somatostatin analogues or chemotherapy.3,31 Cytotoxic chemotherapy is generally of minimal usefulness, and is reserved as a palliative treatment. Drugs such as doxorubicin, 5fluorouracil (FU) with leucovorin, dacarbazine, and docetaxel have limited single-agent activity with short-lasting responses in less than 20% of patients.32 – 34 It is noteworthy that many of these older studies accepted clinical assessment (physical examination) as a basis of declaring a “response”, and as such, the true activity of these agents is likely to be lower than what has been reported. More recently, extrapolation from the 5-FU experience has led to the use of capecitabine, with some anecdotal activity. In a small study, capecitabine combined with temozolomide had high activity.35 Streptozocin, in combination with either cyclophosphamide or 5-FU, has been reported to have activity in older literature, which often did not routinely utilize CT or magnetic resonance imaging (MRI) scan for assessment of response. Furthermore, in these trials, many of the “carcinoid” tumors arose from pancreatic islet cell tumors, biologically different from colorectal carcinoids. In addition, substantial toxicity and no survival advantage was noted in the streptozocin arms.3 Thus, streptozocin does not appear to have a role in the routine management of carcinoid tumors. Systemic chemotherapy, such as cisplatin and etoposide, offers the most benefit for the more aggressive small cell neuroendocrines variants, but is relatively inactive in welldifferentiated carcinoid tumors.36 Because of the limited utility of chemotherapy, symptomatic hepatic carcinoid lesions may be better palliated by regional ablative therapies such as hepatic arterial embolization, which takes advantage of the dual blood supply to the liver. While the normal liver obtains approximately 75% of its blood supply from the portal vein, tumors greater than 1 cm in size obtain the majority for their blood supply through the hepatic artery. Injection of polyvinyl alcohol or other small particulate materials into the angiographically located hepatic arterial blood supply to the metastasis can cause substantial tumor regressions. Toxicities such as pain, infection, fever, nausea, and hepatorenal syndrome have been described, and treatment-related deaths have been reported.37 Patients undergoing hepatic arterial embolization should have it done by experienced physicians, and octreotide premedication should be given in patients with functional tumors to reduce the risk of carcinoid crisis. Since carcinoid is a vascular tumor that expresses vascular endothelial growth factor (VEGF), targeted antiangiogenic drugs such as bevacizumab (AvastinTM ) and SU11248 are being actively investigated. In a recent abstract, bevacizumab
was associated with higher response rates, greater reduction in 5-HIAA levels and longer 18-week progression-free survival (95 vs 67%) compared to IFN alone.38 In addition, preliminary data from a phase II trial with single agent SU11248, an oral targeted tyrosine kinase inhibitor to vascular endothelial growth factor receptor (VEGFR), illustrated low toxicities and a high disease stabilization rate (>90%) in carcinoid and pancreatic islet cell tumors.39 However, actual shrinkage of the tumors was uncommon (partial response: 13.5% islet cell, 5.1% carcinoid). Nonetheless, these novel targeted agents appear promising. Surgery is generally not indicated in the management of metastatic carcinoid. However, in a retrospective analysis, an aggressive surgical intervention in a very selected patient cohort with small volume bowel carcinoid and metastatic liver lesions had a sizable number of patients with potential cures.40 In addition, palliative surgery may be considered in selected patients with symptomatic bulky liver metastases to try delay medical therapy for the endocrinopathies of a functional tumor, or to alleviate pain due to tumor bulk.
Prognosis The aggressiveness of a carcinoid tumor is often related to the degree of tumor atypia. The classic, well-differentiated carcinoid tumors have long survival rates, although they have poor response to chemotherapy. The patient’s age, size of the tumor, site of primary carcinoid tumor, and presence of metastatic disease are the best predictors of prognosis.41 According to the primary site, the 5-year survival is 71–100% (appendix), 33–75% (colon), and 62–100% (rectum).42 Appendiceal and rectal carcinoids are usually smaller than 1 cm when detected and rarely metastasize. In contrast, 85–90% of colonic carcinoids are 2 cm or larger at presentation, with a high metastatic rate (60%).42 Other indicators of a poorer prognosis are overexpression of the proliferation protein Ki-67 or the p53 tumor suppressor protein, carcinoid syndrome, carcinoid heart disease, and high concentrations of urinary 5-HIAA or plasma chromogranin A.42 The high-grade neuroendocrine tumors and undifferentiated small cell carcinomas are the other end of the spectrum. These tumors are very aggressive with short survival, despite high initial response rates to chemotherapy. In a retrospective analysis of 38 patients with colorectal high-grade neuroendocrine tumors or small cell carcinomas, metastatic disease was detected in up to 70% of them at presentation. Like small cell lung cancer, these tumors had a very poor prognosis, with a median survival of 10.4 months and 3-year survival of 13%.43
SARCOMA OF THE COLON AND RECTUM Primary colorectal sarcomas are uncommon. In a retrospective review from Memorial Sloan-Kettering Cancer Center from 1982 to 1991, 38 adult patients were admitted with primary GI sarcomas, accounting for less than 2% of all adult sarcomas admissions. Only 9 of these 38 patients had large bowel sarcomas (two colon, seven rectum), mostly leiomyosarcomas (LMSs) (>90%).44 In recent years, the
UNUSUAL TUMORS OF THE COLON, RECTUM AND ANUS
classification and treatment options for the most common histologic subtypes of the bowel, LMS, have changed dramatically. This section will focus on the recent updates on treatment and biology of intestinal LMS and gastrointestinal stromal cell tumors (GISTs). Readers are also referred to Chapter 37, Gastrointestinal Stromal Tumors for further discussion.
Intestinal LMS and GIST On the basis of light microscopic descriptions in the 1930s to 1950s, stromal tumors of the GI tract were thought to be neoplasms of smooth muscle origin, and were thus classified as LMSs (leiomyosarcomas). However, through electron microscopic and immunohistochemistry studies, the expression of muscle markers (actin and desmin) was more variable than those observed in smooth muscle tumors of the uterus or vessel walls. Moreover, a subset of GI stromal tumors stained positive for neural crest markers (NSE, PGP9.5) and CD34.45 By the early 1990s, there was a growing consensus for the recognition of a clinicopathologically distinct tumor with both smooth muscle and neural features. Thus, this subtype was reclassified as a gastrointestinal stromal tumor (GIST). Due to the strong expression (>95%) of the receptor tyrosine kinase KIT (CD117) and CD34, it is believed that GISTs arise from the gut pacemaker cells, or interstitial cell of Cajal.45 GISTs arise predominately in the stomach (40–60%) and small intestine (25–32%), but also in the rectum (5–10%), large intestine (5%), esophagus (2%), and other locations (5%), including the appendix, mesentery, and retroperitoneum.45,46 Patients range in age from the teens to the 90s, with a median age around 60 years old. In a recent Swedish retrospective analysis of all intestinal tumors that were potential GISTs, the annual incidence was estimated at 14.5 cases per million, with 69% of patients presenting with symptoms.47 In the United States, this would translate to about 4300 new cases per year with fewer than 1000 cases involving the large bowel or rectum. The vast majority of these cases are sporadic events. However, a small series of families with a high incidence of GISTs and skin abnormalities has been described.48 These patients have similar KIT germline mutations as seen in the sporadic cases. In addition, patients with neurofibromatosis type 1 (von Recklinghausen’s neurofibromatosis) have a higher incidence of GISTs (7%), though the underlying reason is still unknown.45 Most patients present with local pain, weight loss, bowel obstruction, tenesmus, or lower GI bleeding; however, about one-third of GISTs are found incidentally. Since these tumors can grow quite large before symptoms develop, GISTs can often be palpated on abdominal or digital rectal examination. Hematogenous spread to the liver or lung has been reported in over 40% cases at presentation. Surgical resection is the mainstay of therapy for localized GISTs. Overall prognosis is best predicted by the size of the tumor and the mitotic count.49 In a retrospective analysis of 200 patients with surgically resected GISTs, the 5-year disease-free survival was 20% for tumors greater than 10 cm, and 60% for tumors less than 5 cm.46 In recurrent or
405
metastatic disease, secondary surgery or standard cytotoxic chemotherapy offer little benefit. For example, single agent doxorubicin, an active chemotherapy in various soft tissue sarcomas, produced partial response rates less than 5%.45 The median duration of survival for patients with metastatic GISTs historically had been less than 20 months, and for patients with local recurrence 9 to 12 months.46 Treatment options have changed dramatically in the last decade with the development of imatinib (Gleevec), a selective tyrosine kinase inhibitor. This drug targets the intracellular Abelson murine leukemia (ABL) kinase, the chimeric breakpoint cluster region (BCR) –ABL fusion oncoprotein of chronic myeloid leukemia, the transmembrane receptor KIT, and the platelet-derived growth factor receptor (PDGFRA).50,51 With single agent imatinib, Demetri and colleagues reported a sustained objective response in more than half the patients with unresectable or metastatic GISTs.52 These results led to its Food and Drug Administration (FDA) approval in 2002. Activating mutations are found in most GISTs, including 88% with KIT (exon 9 or 11) and 8% PDGFRA mutations, and these mutated oncoproteins were predictive of clinical response to imatinib. Exon 11 kit mutations yield higher response rates (84%) and overall survival (approaching 2 years).53 Current research is focusing on the development of new agents for imatinib-resistant tumors, expanding the role of imatinib into the adjuvant setting and in combining such therapy with reoperation for responsive disease. In addition to GISTs, pure smooth muscle tumors (leiomyomas and LMSs) occur in the colon and rectum, but are rare. Older literature is often limited in delineating GISTs from bowel LMSs, given that they were previously classified as one entity.54 In two recent retrospective analyses by the Armed Forces Institute of Pathology, the authors reviewed over 200 colorectal cases and described the clinical and pathologic differences between colorectal GISTs and LMS.55,56 The median age (60 years) was similar. However, GISTs were more common (7 : 1) in men (>70%), and typically the tumors were transmural with frequent intramural and outward-bulging components. GISTs were usually negative for actin, desmin, and S-100. On the other hand, LMS were intraluminally bulging, polypoid masses that showed a histologic likeness to differentiated smooth muscle cells, with negative c-kit and CD34. Despite high mitotic counts, fewer LMS patients died of disease (pre-Imatinib era). Both GISTs and LMS present as an incidental mass or with obstruction or bleeding symptoms. The mainstay of treatment of large bowel LMS is proper surgical resection. Adjuvant radiation therapy, either with external beam or brachytherapy to tumor bed, particularly after resection with close margins, may improve local control in rectal sarcomas but does not appear to impact on the rate of distant metastasis or overall survival.57 Imatinib does not appear to be active in non-GIST soft tissue sarcomas.58 Colorectal LMS is usually treated with similar chemotherapy regimens as other advanced soft tissue sarcomas. Modest response rates have been described with doxorubicin (25%), ifosfamide (17%), etoposide (11%), and a newer combination of gemcitabine and docetaxel (53%); however, the pattern
406
GASTROINTESTINAL TUMORS
and response for GI LMS may be different from that of other sarcomas.59
LYMPHOMA OF THE COLON AND RECTUM Colorectal lymphomas account for 6–14% of all GI lymphomas, with the cecum (73%) and the rectum being the most frequently involved sites.60 Clinical presentation of colorectal lymphoma is typically indistinguishable from colorectal adenocarcinoma. Most present with a mass, with signs of intestinal obstruction, or with GI bleeding. Some patients present with pyrexia of unknown origin, or “B” symptoms with night sweats, fevers, and weight loss. Multiple polypoidal lymphoma affecting the entire colonic mucosa is an extremely rare but well-documented entity. Primary lymphomas of the colon and rectum are believed to arise from lymphocytes associated with the bowel mucosa and may be associated with immunosuppressed states and inflammatory bowel disease.61,62 Aggressive type histology is more frequent, with diffuse large B-cell lymphoma (DLBCL) and Burkitt’s lymphoma accounting for up to 60% of the primary intestinal lymphomas in older retrospective reports.63 Recent studies have reported an increase in the detection of other subtypes, including T-cell lymphomas, Hodgkin disease (HD), mantle cell lymphoma (MCL) and indolent mucosa-associated lymphoid tissue (MALT) lymphoma, also known as marginal zone lymphoma (MZL). For example, investigators reported 88% of colon involvement with MCL, often underreported in prior studies.64 The use of aggressive staging with colonoscopy was found to have little impact on patient management decisions. However, for MCL and MZL, some recommend an initial evaluation of the bowel and if positive, more aggressive treatment and endoscopic follow-up.60 In an attempt to avoid frequent repeat endoscopies, 18F-FDG PET scans are showing promise to determine response and predict recurrence.65 Although the role of surgery in the management of these lymphomas is debated, surgical exploration remains the initial intervention in the majority of cases, and may be life saving in patients who present with perforation, intussusception, or obstruction. Surgery also provides detailed staging and pathology. Chemotherapy, either as primary therapy or following surgery, can be curative. The choice of chemotherapy depends on the grade or aggressiveness of the lymphoma. In general, higher-grade tumors require doxorubicin-containing combination chemotherapy regimens. In low-grade lymphomas, single agent chlorambucil or fludarabine may effect adequate tumor control. In gastric MALT, antibiotics to treat Helicobacter pylori infection can offer a high response rate. An infectious cause for colonic MALT has not yet been discovered, although reports of response to antibiotics have also been described.60
ANAL CARCINOMA Carcinoma of the anal canal will account for approximately 4000 new cases in 2005, although these only represent 1.5% of all digestive system cancers.1 The anal canal extends proximally from the anal verge to the rectal mucosa, with most of
the anal canal lined with squamous mucosa. The dentate line represents a border between the distal squamous mucosa and nonsquamous mucosa, either transitional (urothelium-like) or rectal glandular mucosa. Tumors arising within the anal canal distal to the dentate line are most often keratinizing squamous cell carcinoma, whereas those appearing in the transitional mucosa above the dentate lines are nonkeratinizing squamous cell carcinomas (once referred to as transitional cell and cloacogenic cancers).66 Various cancers can arise in the anal canal including squamous cell carcinoma (74%), adenocarcinoma (19%), melanoma (4%), or other types (3%), such as neuroendocrine, Kaposi sarcoma, LMS, and lymphoma.67 In general, anal adenocarcinomas are treated like rectal adenocarcinoma with concurrent preoperative chemoradiation, followed by surgery and further adjuvant chemotherapy in higher risk patients. This section will focus on the most common subtype of anal cancers, squamous cell carcinoma.
Predisposing Factors For much of the last century, anal cancer was believed to develop due to chronic, local inflammation as a result of conditions such as hemorrhoids, fissures, fistulae, and inflammatory bowel. However, several case-control studies have proven this theory incorrect.68 Instead, genital viral infection and sexual practices are the major predisposing factors.69,70 Risk factors include: human papillomavirus (HPV) infection and anogenital warts (HPV 16, 18), history of receptive anal intercourse before age 30, history of sexually transmitted disease, more than 10 sexual partners, history of cervical, vulvar, or vaginal cancer, and history of cigarette smoking or immunosuppression (chronic steroids, solid-organ transplant or human immunodeficiency virus [HIV]).66
Clinical Presentation and Diagnosis Rectal bleeding is the most common symptom (45%), and patients often delay seeking medical care because symptoms simulate hemorrhoidal bleeding. Some patients (30%) present with local discomfort or pain aggravated by defecation. Change in bowel habits, presence of an anal mass, tenesmus, pruritus, and foul-smelling discharge are less common presentations. Approximately 20% have no rectal symptoms at all.66 Staging workup should include a physical examination with careful attention to the inguinal lymph nodes (since these are frequent sites of local–regional spread), CT scan of the abdomen and pelvis, chest radiography, and routine blood tests. The primary tumor can be measured clinically with anoscopy and with endoscopic ultrasonography. Mobile lesion smaller than 2 cm can be cured in approximately 80% of cases, whereas tumors larger than 5 cm can only be cured in less than 50% of cases.66 Tumor size and lymph node status have important prognostic implications, and are the basis for the clinical staging systems currently in use (American Joint Committee on Cancer [AJCC]).
Treatment Historically, definitive surgery with an abdominal perineal resection and permanent colostomy was the primary therapy
UNUSUAL TUMORS OF THE COLON, RECTUM AND ANUS
for carcinoma of anus. The overall 5-year survival with this surgery ranged from 40–70%.66 However, over the past several decades, the use of combined modality therapy with concurrent radiation therapy and systemic chemotherapy has become widely accepted as the standard of care for this disease. Chemoradiation was initiated by Nigro and colleagues in the early 1970s and was initially intended as a preoperative regimen to convert inoperable cases to candidates for surgery and to improve the cure rate in resectable cases. This has subsequently evolved into the current, nonsurgical, definitive treatment with significant improvement in survival rates and preservation of organ function. The standard chemotherapeutic agents employed have been FU and mitomycin C. Cisplatin has been explored as well. Three large, randomized trials have been reported and provide an objective assessment of the role of chemoradiation over radiation alone.71 – 73 Studies conducted by the European Organization for Research and Treatment of Cancer (EORTC), which included 110 patients, and the United Kingdom Coordinating Committee on Cancer Research (UKCCCR), which involved 585 patients, compared pelvic radiation alone (total dose of 45 Gy) with the same radiation administered with concurrent 5-FU and mitomycin. In both studies, concurrent chemoradiation resulted in a significant increase in local control (39 vs 58%, EORTC; 39 vs 61%, UK-CCCR) and a lower probability of a colostomy. The overall 3-year survival tended toward the chemoradiation group (65 vs 72%, EORTC; 58 vs 65%, UK-CCCR), however, it was not significantly different.71,72 This was likely due to the fact that the patients could still be salvaged with abdominoperineal resection (APR). The Radiation Therapy Oncology Group (RTOG) and Eastern Cooperative Oncology Group (ECOG) studies were designed to test the need for inclusion of mitomycin C as a component in the 5-FU –based chemoradiation, and to test the effectiveness of 5-FU infusion plus cisplatin as a salvage measure for persistent disease.73 In total, 310 patients were enrolled. Participants in one arm of the study were treated with concurrent pelvic radiation plus intravenous 5-FU given by 96-hour continuous infusion beginning on days 1 and 28 of radiation. Patients in the second arm received the same radiation and FU infusions, plus a bolus injection of mitomycin on days 1 and 28. If the postchemoradiation biopsy was negative for tumor, no further treatment was given unless a relapse occurred. The results showed substantially greater toxicity with the use of mitomycin, but a significant decline in local recurrence (36 vs 17%). Besides, the addition of mitomycin led to a significantly improved colostomy-free survival at 5 years (58 vs 64%) and disease-free survival (50 vs 67%), though a significant difference in overall survival was not reached (65 vs 67%). This study also evaluated the usefulness of salvage, nonsurgical therapy in the event of local relapse or failure to respond fully to primary therapy. If the biopsy was positive for tumor, then salvage chemotherapy was initiated with cisplatin plus FU with a concomitant dose of salvage radiation therapy. Patients who failed to achieve a complete
407
response to this salvage therapy and who did not have evidence of metastatic disease were treated with a salvage abdominal-peritoneal resection. Fifty percent of the patients who underwent this salvage chemoradiation with cisplatinbased therapy were rendered disease-free at 4 years. Given these promising results, combined modality therapy with cisplatin and 5-FU is being studied as an initial treatment regimen. In one retrospective study, this regimen was well tolerated and resulted in local control, overall survival, and sphincter preservation rates comparable to the best results reported with mitomycin C and 5-FU.74 To attempt to improve outcomes in poorer prognosis patients (T3/T4 or nodal metastasis), the Cancer and Leukemia Group B (CALGB) performed a phase II study of induction 5-FU/cisplatin followed by combined radiation plus 5-FU/mitomycin.75 The induction chemotherapy led to an impressive 18% clinical complete response and 47% partial response rate. Moreover, when followed by radiation, 5-FU plus/mitomycin, most patients were alive (68%) and diseasefree (61%) at 4-years of follow-up. Phase III trials are under way to confirm the benefit of this platinum-containing regimen. Chemoradiation is considered to be the current standard treatment. In debilitated patients or patients in whom chemotherapy is contraindicated, radiation therapy alone can be offered as an alternative, but it should be recognized that this is substandard care, and is being given only due to limitations imposed by a patient’s medical comorbidities. HIV infection alone (CD4 count >200) does not appear to constitute a substantial contraindication to chemotherapy; however, patients with clinical debilitation, clinical acquired immune deficiency syndrome (AIDS), and/or significant bone marrow compromise may not be appropriate candidates for full combined modality therapy because of the greatly increased risk of serious toxicities. Lastly, in the EORTC and UKCCCR trials, distant metastases developed in 17 and 10%, respectively, with the most frequent site being the liver.66 Chemotherapy options include cisplatin and 5-FU with minimal evidence of activity for other regimens or agents. Such treatment is considered palliative.
MALIGNANT MELANOMA Primary malignant melanoma is an extremely rare and aggressive tumor of the anal region. Since melanocytes are absent above the pectinate line, melanoma in the colon or rectum is considered to be metastatic. As in cutaneous melanoma, depth of penetration is the most important prognostic indicator, but this can be difficult to assess in melanoma of the anal canal. Anal melanoma accounts for 24% of mucosal melanomas and less than 1% of all melanomas. Patients often present in the fifth to sixth decade of life with vague symptoms, including rectal bleeding and anal discomfort, and diagnosis is often delayed.76 Typically, treatment for localized disease has been complete surgical resection, ranging from a radical APR with elective lymph node dissection to conservative sphinctersparing local excision alone. Although APR has been reported to have better local control rates (70% with APR
408
GASTROINTESTINAL TUMORS
vs 35% with local excision), this surgery has high comorbidity with no improvement in overall survival (<25% 5-year survival).77,78 Another approach is to combine sphinctersparing local excision with adjuvant radiation (30 Gy). In a retrospective analysis of 23 patients, this combined approach appeared to be reasonably effective and well tolerated.77 Adjuvant systemic therapy with either cytotoxic therapy or immunotherapy has been investigated and is favored by some clinicians, but its benefit has not been clearly defined. Although the majority of patients (70%) present with localized disease, most will eventually succumb to brain, lung, or liver metastases. Treatment of metastatic anal melanoma is based on available drugs for advanced cutaneous melanoma and includes combinations of cisplatin, vinblastine, dacarbazine, IFN-α, and interleukin II. Other reported options include liposomal doxorubicin and oral temozolomide.76 However, there is no improvement in survival rates with any of these regimens and thus, enrollment in logical clinical trials should be encouraged.
PAGET’S DISEASE OF THE PERIANAL REGION Paget’s disease of the perianal skin is a rare intraepithelial adenocarcinoma of the dermal apocrine gland, with fewer than 200 cases reported in the literature between 1966 and 1995.79 Extramammary Paget’s disease can occur in any region of skin-bearing apocrine glands such as anus, axilla, scrotum, groin, eyelids, and external auditory canal. Like Paget’s disease of the nipple, Paget’s disease of the perianal skin may be associated with an underlying malignancy of the GI tract and these should be actively looked for. Paget’s disease of the perianal region usually presents as a well-demarcated, erythematous, scaly, and pruritic lesion that bleeds occasionally.80 Diagnosis is often delayed since the lesion is confused with common conditions such as pruritus anni or bleeding hemorrhoids. Diagnosis is made with a biopsy which illustrates the classic histologic finding of Paget cells: large round cells with pale vacuolated cytoplasm and reticular nucleus that stain positive for periodic acid-Schiff (PAS), CAM5.2, CK7 and CEA.80,81 Paget’s disease of the anus appears to run one of two clinical courses: to develop malignancy; or to recur locally, often on repeated occasions. Treatment with wide local excision is recommended in the absence of underlying malignancy. Small case reports have described radiation with or without chemotherapy as a nonsurgical alternative.80 Local recurrence with in situ disease is common (up to 60%); however, this can be managed with repeat wide excisions with excellent long-term prognosis. In patients with an invasive component or an associated anorectal adenocarcinoma, APR is usually recommended.
REFERENCES 1. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55(1): 10 – 30. 2. DiSario JA, et al. Colorectal cancers of rare histologic types compared with adenocarcinomas. Dis Colon Rectum 1994; 37(12): 1277 – 80. 3. Kulke MH, Mayer RJ. Carcinoid tumors. N Engl J Med 1999; 340(11): 858 – 68.
4. Maggard MA, O’Connell JB, Ko CY. Updated population-based review of carcinoid tumors. Ann Surg 2004; 240(1): 117 – 22. 5. Modlin IM, Sandor A. An analysis of 8305 cases of carcinoid tumors. Cancer 1997; 79(4): 813 – 29. 6. Crocetti E, Paci E. Malignant carcinoids in the USA, SEER 1992 – 1999. An epidemiological study with 6830 cases. Eur J Cancer Prev 2003; 12(3): 191 – 4. 7. Capella C, et al. Revised classification of neuroendocrine tumours of the lung, pancreas and gut. Virchows Arch 1995; 425(6): 547 – 60. 8. Federspiel BH, et al. Rectal and colonic carcinoids. A clinicopathologic study of 84 cases. Cancer 1990; 65(1): 135 – 40. 9. Soga J, Yakuwa Y, Osaka M. Carcinoid syndrome: a statistical evaluation of 748 reported cases. J Exp Clin Cancer Res 1999; 18(2): 133 – 41. 10. Jetmore AB, et al. Rectal carcinoids: the most frequent carcinoid tumor. Dis Colon Rectum 1992; 35(8): 717 – 25. 11. Rosenberg JM, Welch JP. Carcinoid tumors of the colon. A study of 72 patients. Am J Surg 1985; 149(6): 775 – 9. 12. Moertel CG. Karnofsky memorial lecture. An odyssey in the land of small tumors. J Clin Oncol 1987; 5(10): 1502 – 22. 13. Moertel CG, Dockerty MB, Judd ES. Carcinoid tumors of the vermiform appendix. Cancer 1968; 21(2): 270 – 8. 14. Moertel CG, et al. Carcinoid tumor of the appendix: treatment and prognosis. N Engl J Med 1987; 317(27): 1699 – 701. 15. Roggo A, Wood WC, Ottinger LW. Carcinoid tumors of the appendix. Ann Surg 1993; 217(4): 385 – 90. 16. Koura AN, et al. Carcinoid tumors of the rectum: effect of size, histopathology, and surgical treatment on metastasis free survival. Cancer 1997; 79(7): 1294 – 8. 17. Burke M, Shepherd N, Mann CV. Carcinoid tumours of the rectum and anus. Br J Surg 1987; 74(5): 358 – 61. 18. Sauven P, et al. Anorectal carcinoid tumors. Is aggressive surgery warranted? Ann Surg 1990; 211(1): 67 – 71. 19. Ballantyne GH, et al. Incidence and mortality of carcinoids of the colon. Data from the Connecticut tumor registry. Cancer 1992; 69(10): 2400 – 5. 20. Reichlin S. Somatostatin. N Engl J Med 1983; 309(24): 1495 – 501. 21. Reubi JC, et al. Detection of somatostatin receptors in surgical and percutaneous needle biopsy samples of carcinoids and islet cell carcinomas. Cancer Res 1990; 50(18): 5969 – 77. 22. Saltz L, et al. Octreotide as an antineoplastic agent in the treatment of functional and nonfunctional neuroendocrine tumors. Cancer 1993; 72(1): 244 – 8. 23. Kvols LK, et al. Treatment of the malignant carcinoid syndrome. Evaluation of a long-acting somatostatin analogue. N Engl J Med 1986; 315(11): 663 – 6. 24. de Herder WW, Lamberts SW. Somatostatin and somatostatin analogues: diagnostic and therapeutic uses. Curr Opin Oncol 2002; 14(1): 53 – 7. 25. di Bartolomeo M, et al. Clinical efficacy of octreotide in the treatment of metastatic neuroendocrine tumors. A study by the Italian Trials in Medical Oncology Group. Cancer 1996; 77(2): 402 – 8. 26. Rubin J, et al. Octreotide acetate long-acting formulation versus openlabel subcutaneous octreotide acetate in malignant carcinoid syndrome. J Clin Oncol 1999; 17(2): 600 – 6. 27. Otte A, et al. Yttrium-90-labelled somatostatin-analogue for cancer treatment. Lancet 1998; 351(9100): 417 – 8. 28. Zuetenhorst JM, Taal BG. Metastatic carcinoid tumors: a clinical review. Oncologist 2005; 10(2): 123 – 31. 29. Lamberts SW, et al. Somatostatin-receptor imaging in the localization of endocrine tumors. N Engl J Med 1990; 323(18): 1246 – 9. 30. Oberg K. Neuroendocrine tumors of the gastrointestinal tract: recent advances in molecular genetics, diagnosis, and treatment. Curr Opin Oncol 2005; 17(4): 386 – 91. 31. Saltz L, et al. A phase II trial of alpha-interferon and 5-fluorouracil in patients with advanced carcinoid and islet cell tumors. Cancer 1994; 74(3): 958 – 61. 32. Bukowski RM, et al. Phase II trial of dimethyltriazenoimidazole carboxamide in patients with metastatic carcinoid. A Southwest Oncology Group study. Cancer 1994; 73(5): 1505 – 8. 33. Kulke MH, et al. A phase II study of docetaxel in patients with metastatic carcinoid tumors. Cancer Invest 2004; 22(3): 353 – 9.
UNUSUAL TUMORS OF THE COLON, RECTUM AND ANUS 34. Moertel CG. Treatment of the carcinoid tumor and the malignant carcinoid syndrome. J Clin Oncol 1983; 1(11): 727 – 40. 35. Fine RL, Fogelman DR, Schreibman SM. Effective treatment of neuroendocrine tumors with temozolomide and capecitabine. Proc Am Soc Clin Oncol 2005; 23(16s): 361s (abstr 4216). 36. Moertel CG, et al. Treatment of neuroendocrine carcinomas with combined etoposide and cisplatin. Evidence of major therapeutic activity in the anaplastic variants of these neoplasms. Cancer 1991; 68(2): 227 – 32. 37. Gupta S, et al. Hepatic artery embolization and chemoembolization for treatment of patients with metastatic carcinoid tumors: the M.D. Anderson experience. Cancer J 2003; 9(4): 261 – 7. 38. Yao JC, et al. Improved progression free survival (PFS), and rapid, sustained decrease in tumor perfusion among patients with advanced carcinoid treated with bevacizumab. Proc Am Soc Clin Oncol 2005; 23(16s): 309s (abstr. 4007). 39. Kulke M, et al. A phase 2 study to evaluate the efficacy and safety of SU11248 in patients (pts) with unresectable neuroendocrine tumors (NETs). Proc Am Soc Clin Oncol 2005; 23(16s): 310s (abstr. 4008). 40. McEntee GP, et al. Cytoreductive hepatic surgery for neuroendocrine tumors. Surgery 1990; 108(6): 1091 – 6. 41. Shebani KO, et al. Prognosis and survival in patients with gastrointestinal tract carcinoid tumors. Ann Surg 1999; 229(6): 815 – 21; discussion 822 – 3. 42. Rorstad O. Prognostic indicators for carcinoid neuroendocrine tumors of the gastrointestinal tract. J Surg Oncol 2005; 89(3): 151 – 60. 43. Bernick PE, et al. Neuroendocrine carcinomas of the colon and rectum. Dis Colon Rectum 2004; 47(2): 163 – 9. 44. Conlon KC, Casper ES, Brennan MF. Primary gastrointestinal sarcomas: analysis of prognostic variables. Ann Surg Oncol 1995; 2(1): 26 – 31. 45. Corless CL, Fletcher JA, Heinrich MC. Biology of gastrointestinal stromal tumors. J Clin Oncol 2004; 22(18): 3813 – 25. 46. DeMatteo RP, et al. Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 2000; 231(1): 51 – 8. 47. Nilsson B, et al. Gastrointestinal stromal tumors: the incidence, prevalence, clinical course, and prognostication in the preimatinib mesylate era – a population-based study in western Sweden. Cancer 2005; 103(4): 821 – 9. 48. Li FP, et al. Familial gastrointestinal stromal tumor syndrome: phenotypic and molecular features in a kindred. J Clin Oncol 2005; 23(12): 2735 – 43. 49. Fletcher CD, et al. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Hum Pathol 2002; 33(5): 459 – 65. 50. Druker BJ, et al. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med 1996; 2(5): 561 – 6. 51. Buchdunger E, et al. Abl protein-tyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-kit and plateletderived growth factor receptors. J Pharmacol Exp Ther 2000; 295(1): 139 – 45. 52. Demetri GD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med 2002; 347(7): 472 – 80. 53. Heinrich MC, et al. Kinase mutations and imatinib response in patients with metastatic gastrointestinal stromal tumor. J Clin Oncol 2003; 21(23): 4342 – 9. 54. Hatch KF, et al. Tumors of the appendix and colon. World J Surg 2000; 24(4): 430 – 6. 55. Miettinen M, et al. Gastrointestinal stromal tumors and leiomyosarcomas in the colon: a clinicopathologic, immunohistochemical, and molecular genetic study of 44 cases. Am J Surg Pathol 2000; 24(10): 1339 – 52. 56. Miettinen M, et al. Gastrointestinal stromal tumors, intramural leiomyomas, and leiomyosarcomas in the rectum and anus: a clinicopathologic, immunohistochemical, and molecular genetic study of 144 cases. Am J Surg Pathol 2001; 25(9): 1121 – 33. 57. Grann A, et al. Sphincter preservation of leiomyosarcoma of the rectum and anus with local excision and brachytherapy. Dis Colon Rectum 1999; 42(10): 1296 – 9.
409
58. Demetri G, George S, Heinrich M. Clinical activity and tolerability of the multi-targeted tyrosine kinase inhibitor SU 11248 in patients with metastatic gastrointestinal stromal tumor refractory to imatinib mesylate (abstract). Proc Am Soc Clin Oncol 2003; 22: 814a. 59. Hensley ML, et al. Gemcitabine and docetaxel in patients with unresectable leiomyosarcoma: results of a phase II trial. J Clin Oncol 2002; 20(12): 2824 – 31. 60. Romaguera J, Hagemeister FB. Lymphoma of the colon. Curr Opin Gastroenterol 2005; 21(1): 80 – 4. 61. Doolabh N, et al. Primary colonic lymphoma. J Surg Oncol 2000; 74(4): 257 – 62. 62. Watanabe N, et al. Association of intestinal malignant lymphoma and ulcerative colitis. Intern Med 2003; 42(12): 1183 – 7. 63. Zighelboim J, Larson MV. Primary colonic lymphoma. Clinical presentation, histopathologic features, and outcome with combination chemotherapy. J Clin Gastroenterol 1994; 18(4): 291 – 7. 64. Romaguera JE, et al. Frequency of gastrointestinal involvement and its clinical significance in mantle cell lymphoma. Cancer 2003; 97(3): 586 – 91. 65. Kumar R, et al. 18F-FDG PET for evaluation of the treatment response in patients with gastrointestinal tract lymphomas. J Nucl Med 2004; 45(11): 1796 – 803. 66. Ryan DP, Compton CC, Mayer RJ. Carcinoma of the anal canal. N Engl J Med 2000; 342(11): 792 – 800. 67. Klas JV, et al. Malignant tumors of the anal canal: the spectrum of disease, treatment, and outcomes. Cancer 1999; 85(8): 1686 – 93. 68. Frisch M, et al. Benign anal lesions and the risk of anal cancer. N Engl J Med 1994; 331(5): 300 – 2. 69. Daling JR, et al. Sexual practices, sexually transmitted diseases, and the incidence of anal cancer. N Engl J Med 1987; 317(16): 973 – 7. 70. Frisch M, et al. Sexually transmitted infection as a cause of anal cancer. N Engl J Med 1997; 337(19): 1350 – 8. 71. UKCCCR Anal Cancer Trial Working Party. UK Co-ordinating Committee on Cancer Research. Epidermoid anal cancer: results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and mitomycin. Lancet 1996; 348(9034): 1049 – 54. 72. Bartelink H, et al. 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 1997; 15(5): 2040 – 9. 73. Flam M, et al. 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 1996; 14(9): 2527 – 39. 74. Hung A, et al. Cisplatin-based combined modality therapy for anal carcinoma: a wider therapeutic index. Cancer 2003; 97(5): 1195 – 202. 75. Meropol NJ, et al. Combined-modality therapy of poor prognosis anal canal carcinoma: a phase II study of the Cancer and Leukemia Group B (CALGB). Proc Am Soc Clin Oncol-Gastrointest Symp 2005; 203 (abstr. 238). 76. Yeh JJ, et al. Response of stage IV anal mucosal melanoma to chemotherapy. Lancet Oncol 2005; 6(6): 438 – 9. 77. Ballo MT, et al. Sphincter-sparing local excision and adjuvant radiation for anal-rectal melanoma. J Clin Oncol 2002; 20(23): 4555 – 8. 78. Brady MS, Kavolius JP, Quan SH. Anorectal melanoma. A 64-year experience at memorial Sloan-Kettering cancer center. Dis Colon Rectum 1995; 38(2): 146 – 51. 79. Beck D. Paget’s disease and Bowen’s disease of the anus. Semin Colon Rectal Surg 1995; 6: 143 – 9. 80. Moore H, Guillem J. Anal neoplasms. Surg Clin N Am 2002; 82: 1233 – 51. 81. Armitage NC, et al. Paget’s disease of the anus: a clinicopathological study. Br J Surg 1989; 76(1): 60 – 3.
Section 6 : Gastrointestinal Tumors
36
Cancer of the Appendix Matthew H. Kulke and Charles S. Fuchs
INTRODUCTION Appendiceal tumors comprise a broad range of benign and malignant lesions. Benign mucinous neoplasms are discovered in 0.3% of all appendectomy specimens, and appendiceal malignancies are diagnosed in 0.9–1.4% of cases.1,2 Carcinoid tumors have historically been the most common appendiceal malignancy, and are followed in frequency by mucinous adenocarcinomas (including signet ring cell carcinomas) and colonic-type adenocarcinomas. Adenocarcinoid (goblet cell) tumors, which share features of both carcinoids and adenocarcinomas are among the least common.3,4 Although numerous case series of appendiceal tumors exist, few large-scale, population-based studies have examined these neoplasms. McCusker and colleagues analyzed all cases of appendiceal neoplasms in the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) program. Between 1973 and 1998, 2117 appendiceal malignancies were reported, and the overall age-adjusted incidence of appendiceal malignancies was 0.12 cases per million people per year.5 Mucinous adenocarcinomas and signet ring carcinomas comprised 41% of all cases, and were followed in incidence by colonic-type adenocarcinomas (25%), “malignant” carcinoid tumors (20%), and adenocarcinoid tumors (14%). Of note, the authors did not include smaller, “benign” appearing carcinoid tumors, which may represent a substantial proportion of all appendiceal neoplasms. In fact, other case series that have sought to capture all appendiceal tumors found that appendiceal carcinoids represented over 50% of all appendiceal malignancies.2,6,7 The prognosis for patients with appendiceal malignancies is strongly related to histologic subtype. Among the patients identified through the SEER registry, those with carcinoid tumors experienced the longest overall survival, whereas survival progressively decreased in patients with goblet cell carcinoid, “colonic-type” adenocarcinoma, mucinous adenocarcinoma, and signet ring carcinoma (see Figure 1). Nonetheless, when adjustment for age and extent of disease were incorporated in the analysis, differences in survival were only significant for malignant carcinoid tumor, which had the best prognosis, and signet ring cell carcinoma, which conferred the worst outcome.
BENIGN MUCINOUS NEOPLASMS OF THE APPENDIX Epithelial neoplasms of the appendix comprise a wide spectrum of tumors, ranging from benign mucinous hyperplasia or mucoceles to adenocarcinoma. Although earlier studies tended to draw a strict distinction between mucinous and nonmucinous tumors, investigators have more recently classified each histology separately, according to its malignant potential. Benign appendiceal mucoceles are usually asymptomatic and discovered incidentally during surgery for other procedures. They are more common in women than in men, and are discovered in the fifth or sixth decade of life during surgery for other procedures.1,8,9 Occasionally, these tumors may present as acute appendicitis or as a right lower quadrant mass. Appendiceal mucoceles are presumably caused by obstruction of the appendiceal lumen. While the appendix may appear grossly dilated, the epithelium itself remains either normal or demonstrates only mild degenerative changes. Other mucoceles may be associated with neoplastic changes analogous to those seen in hyperplastic or adenomatous polyps of the colon. The dilated mucinfilled appendix may contain tall, columnar cells with papillary projections into the lumen; such lesions are typically classified as mucinous cystadenomas. Both mucoceles and mucinous cystadenomas are benign lesions that are cured surgically; in the absence of perforation, these lesions are handled by simple appendectomy with no reported cases of recurrence. Nonetheless, within the resection specimen, any evidence of tumor invasion into the muscularis propria or extra-appendiceal spread should raise concerns for potential malignant behavior.10
PSEUDOMYXOMA PERITONEI Pseudomyxoma peritonei (PMP) is a rare condition characterized by the development of abundant mucin on peritoneal surfaces and within the abdominal cavity. PMP occurs with equal incidence in males and females, and typically presents in the sixth decade of life.11 – 13 Patients generally describe symptoms of abdominal pain or distension; the duration of
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
CANCER OF THE APPENDIX
411
100 “Malignant carcinoid”
Percent survivng
80 Goblet c
ell carcin
60
“Colonic
-type” a deonca rcinoma Mucin ous ad enoca rcinom a
40
Signet rin
20
0
oid
g cell carc
0
1
2
3
4 5 6 7 Years after diagnosis
inoma
8
9
10
Figure 1 Observed survival for appendiceal malignancies by histology (Adapted from McKusker et al., Cancer 2002; 94: 3307 – 12, by permission of John Wiley & Sons, Inc.).
Figure 2 Computed tomography demonstrating loculated mucinous ascites consistent with pseudomyxoma peritonei.
these symptoms can vary from as little as 2 months to, in one case, 16 years.12,14 Despite its often benign appearance, PMP generally is fatal and is therefore most accurately classified as a low-grade malignancy. PMP has been reported with less than 10% of mucinous appendiceal neoplasms, and is thought to occur when a mucinous appendiceal tumor ruptures and spills mucin into the peritoneal cavity. PMP has also been reported in association with mucinous neoplasms of other sites, including the gallbladder, stomach, ovary, and pancreas.13,15 – 18 While an older body of literature describes the relatively frequent development of PMP in association with ovarian neoplasms, more recent reports suggest that the vast majority of cases are appendiceal in origin.19,20 The development of PMP is thought to require the presence of extra-appendiceal mucin-secreting epithelial cells.1,9,21 The rupture of benign mucoceles without intramucinous epithelial cells does not appear to result in PMP, whereas patients with PMP often have implants of mucin-secreting cells in the peritoneum. Epithelial cells have, in some cases, been detected in the mucin itself; in contrast to mucinous adenocarcinoma, the epithelial cells in classic PMP have a benign appearance. As mucin is distributed to the peritoneum, epithelial cell implants most commonly develop in the omentum, the undersurface of the right diaphragm, and the pelvis. The surfaces of the small and large intestines may be spared, presumably because peristaltic activity prevents tumor seeding in these areas.22 While PMP can occasionally be diagnosed with paracentesis, the gelatinous nature of the mucin often makes such attempts difficult.14 Radiologic imaging with computed tomography (CT) typically reveals the appearance of multiple mucin-filled cystic lesions (see Figure 2); scalloping of the hepatic margin due to compression from the mucinous ascites is another classic finding.23 Surgical debulking procedures are commonly used as the primary treatment for patients with PMP, although the optimal extent of debulking remains controversial. The objective
evaluation of treatment approaches in PMP is complicated by varying definitions of PMP, with some series reporting only more indolent forms of the disease and others reporting results from patients with frank peritoneal carcinomatosis. Repeated debulking procedures with evacuation of all free mucus have historically been the most common treatment approach; occasionally patients have also received intraperitoneal (see Table 1) and systemic chemotherapy. This approach has been associated with 5-year survival rates of 15–75% and 10-year survival rates of 32–60%.12 – 14,24 – 26 Sugarbaker and colleagues have advocated an aggressive, multimodality approach incorporating even more extensive cytoreductive surgery with intraperitoneal chemotherapy.27 With this approach, patients have undergone omentectomy–splenectomy, left and right subdiaphragmatic peritonectomy, pelvic peritonectomy-sleeve resection of the sigmoid colon, and cholecystectomy-lesser omentectomy. Early postoperative intraperitoneal chemotherapy is then administered with mitomycin-C and 5-fluorouracil (5-FU).28 In an initial study of 69 patients with peritoneal seeding, the 3-year survival rate was 70%.29 The procedure, however, was associated with a 35% complication rate. In a subsequent follow-up study of 385 patients, patients with complete cytoreduction experienced a 5-year survival rate of 86%, and those with incomplete cytoreduction a 5-year survival rate of 20%.26 Postoperative morbidity occurred in 27% and postoperative mortality in 2.7% of patients. Because of the lack of randomized trials of extensive cytoreductive surgery in PMP and the relatively indolent nature of the malignancy, it is difficult to assess the true impact of this intervention on patient outcome. While the ability to achieve a complete cytoreduction may represent a valuable prognostic factor in multiple studies, it is not clear whether achieving maximum surgical cytoreduction truly alters the natural history of the disease or rather simply identifies patients with an underlying superior prognosis. Nevertheless, the potential for morbidity and mortality associated with this approach has led some investigators to
412
GASTROINTESTINAL TUMORS
Table 1 Selected studies of surgical debulking and intraperitoneal therapy for pseudomyxoma peritonei.
Study Long et al.14 Gough et al.13 Smith et al.12 Culliford et al.24
Number of patients
5-year survival (%)
17 56 17 64
45 53 75 54 (Complete cytoreduction) 16 (Incomplete cytoreduction) 15 (Incomplete cytoreduction) 86 (Complete cytoreduction) 20 (Incomplete cytoreduction)
Glehen et al.25
174
Sugarbaker et al.26
385
∗
10-year survival (%) ∗
32 60 ∗
∗
∗
Not reported.
discourage its use, particularly in patients in whom complete cytoreduction is not felt to be feasible.30 At present, it remains difficult to assess the relative impact of aggressive surgical debulking and intraperitoneal chemotherapy on survival in patients with PMP. The value of systemic chemotherapy in PMP is even less clear. While small studies have suggested that systemic chemotherapy has little or no impact on this disease, there are anecdotal reports of responses to systemic 5-FU-based regimens.31 While surgical debulking approaches should clearly be considered first, systemic chemotherapy with agents typically utilized for metastatic colorectal cancer could therefore be considered, particularly in patients who have evidence of mucinous adenocarcinoma of appendiceal origin.
APPENDICEAL ADENOCARCINOMA Adenocarcinomas are generally grouped into two histologic types: mucinous adenocarcinomas, characterized by abundant extra- and intracellular mucin production (of which signet cell carcinomas can be considered a further subtype), and colonic-type adenocarcinomas, which resemble the typical malignancy found elsewhere in the large bowel.5,8,9 Both histologic subtypes are characterized by the presence of tumor invasion, a feature that separates them from benign appendiceal mucoceles. In contrast to colorectal adenocarcinoma, both mucinous and nonmucinous appendiceal adenocarcinomas are commonly associated with perforation, which occurs in up to half of all cases.3,4,32,33 This high frequency of perforation has been attributed to the thin muscular wall of the appendix and the tendency for the appendix to become obstructed, resulting in lumenal distension. Interestingly, in retrospective studies, the presence of perforation does not appear to have an adverse effect on prognosis.4,34 However, in analyzing the effects of perforation on prognosis, no study has yet adequately controlled for tumor stage, and it is possible that perforation leads to earlier diagnosis. Appendiceal adenocarcinomas are more frequent in males than in females, with ratios ranging from 3 : 2 to 3 : 1.3 – 5,32,35 The majority of patients present with symptoms of acute appendicitis; in contrast to younger patients with benign
appendicitis, however, the median age of patients with appendiceal adenocarcinoma is 60.5 Less commonly, patients may present with a palpable right lower quadrant mass, ascites, periappendiceal abscess, or generalized abdominal pain.36 Occult bleeding is rare. The pattern of metastatic spread for appendiceal adenocarcinomas varies according to histologic subtype. Whereas colonic-type adenocarcinomas tend to spread via lymphatic or hematogenous dissemination, mucinous adenocarcinomas more commonly disseminate directly to the peritoneum.35,36 These different patterns of metastatic spread have led to varying recommendations regarding surgical management. Consistent with standard adenocarcinoma of the colon, surgical management of colonic-type appendiceal adenocarcinomas has routinely involved a right hemicolectomy, except for patients with widespread metastatic disease. In selected, nonrandomized series of patients with nonmetastatic colonictype appendiceal adenocarcinoma, right colectomy has been consistently associated with improved overall survival rates when compared to simple appendectomy alone.4,33 Conversely, in patients with mucinous adenocarcinomas of the appendix, some investigators have recommended simple appendectomy alone, given the high incidence of peritoneal, rather than lymph node metastases.35 Nonetheless, other series that include patients with nonmetastatic mucinous cancers demonstrate a survival benefit associated with right colectomy when compared to simple appendectomy.4 Similarly, while appendectomy has been advocated for patients with early stage lesions, subsequent studies have documented the presence of more extensive disease in up to 40% of cases of presumed early stage adenocarcinoma.4 The longitudinal and circular muscle layers are deficient in certain areas of the appendix, causing the submucosa and serosa to lie in close proximity, theoretically increasing the possibility of metastatic spread. Given this potential risk of more extensive disease, right colectomy would appear to be the appropriate management for the majority of patients with localized appendiceal adenocarcinomas, regardless of histology. Whether right colectomy truly reduces the longer-term risk of recurrence in such patients, however, has not been formally studied.
CANCER OF THE APPENDIX
Metastases to the ovaries are common in women with appendiceal adenocarcinomas, a finding that has led some investigators to recommend routine prophylactic oophorectomy in addition to right colectomy in these cases.3,4 While there is little evidence to either refute or support such an approach, several investigators argue that oophorectomy for women with metastatic disease limited to the ovaries may improve survival. In one series, patients who underwent resection of isolated ovarian metastases experienced a 5-year survival rate of 31%.4 The role of systemic chemotherapy in the treatment of patients with appendiceal adenocarcinoma has not been formally studied, and, given the rarity of this disease, it is unlikely that such studies will be performed. Given the anatomic location and similarities in histology between appendiceal adenocarcinomas and more typical colon cancers, treatment recommendations for appendiceal adenocarcinoma generally parallel those for colon cancer. The administration of adjuvant 5-FU-based chemotherapy has clearly been shown to be associated with survival in patients with stage III (lymph node positive) colon cancer, and may also be administered to patients with stage II disease and high-risk features such as perforation.37,38 Treating patients with similarly staged appendiceal adenocarcinoma with such adjuvant regimens would thus seem to be reasonable. External beam radiation therapy appears to reduce the risk of local recurrence in resected patients with evidence of perforation or tumor adherence to other structures, and should be considered in selected cases.39 Patients with metastatic appendiceal adenocarcinoma are also generally treated with systemic chemotherapy regimens that have been established for the treatment of advanced colorectal cancer. Anecdotal reports of tumor responses to 5FU in patients with metastatic appendiceal adenocarcinoma are common.3,9,33,34,36,40 – 43 The more recent development of chemotherapy regimens for advanced colorectal adenocarcinoma incorporating oxaliplatin, irinotecan, and the “targeted” agents bevacizumab and cetuximab, may ultimately expand future treatment options for patients with advanced appendiceal adenocarcinoma.
ADENOCARCINOID TUMORS OF THE APPENDIX Adenocarcinoid tumors are tumors with characteristics of both carcinoid tumors and adenocarcinoma. They have also been referred to as goblet cell carcinoids, crypt cell carcinomas, and amphicrine carcinomas.44 – 46 Patients with adenocarcinoid tumors generally present in the sixth decade of life, with symptoms of acute appendicitis or chronic abdominal pain. They have a roughly equal gender ratio, and only 20% are found during an appendectomy performed for other reasons. The presence of cells with neuroendocrine features together with mucin-secreting cells has raised uncertainty about their histologic derivation. While some studies have postulated distinct derivations for the two cell populations, others have documented neuroendocrine and mucinous features in the same cell, supporting a common cell of origin for these tumors.47 – 49 The histologic features of
413
adenocarcinoid tumors may vary from a benign appearance to an appearance that is malignant. Adenocarcinoid tumors have, in fact, been further subclassified into “tubular carcinoids”, which comprise predominantly welldifferentiated carcinoid cells and small amounts of mucin, “goblet cell carcinoids” containing greater amounts of mucin and a higher mitotic rate, and mixed carcinoidadenocarcinomas, which have features more suggestive of a frank appendiceal adenocarcinoma.50 Adenocarcinoids are often infiltrative in nature, making precise measurement of their size and location within the appendix difficult.51 The specific histologic subtype of adenocarcinoid tumors may significantly affect their clinical behavior and may be the most accurate predictor of prognosis. Patients whose tumors have features of frank adenocarcinoma are at highest risk of recurrence. In one series of 10 patients with mixed carcinoid-adenocarcinoma, 8 patients died of metastatic disease.50 Peritoneal spread of disease is a far more common finding in this setting than are lymph node metastases.9,47,50 – 56 Adenocarcinoids also commonly metastasize to the ovaries, presenting as Krukenberg tumors.54,55,57 – 62 In one series of 810 patients presenting with peritoneal carcinomatosis, 22 had adenocarcinoid as their primary site of origin.63 The surgical management of adenocarcinoid tumors remains controversial. While appendectomy has been considered by some investigators to be adequate in the absence of positive margins, others have recommended routine right hemicolectomy.9,44,53,57 Commonly, right colectomy is performed in patients whose tumors measure more than 2 cm in diameter, have involvement of the base of the appendix, or contain histologic features suggestive of an adenocarcinoma. Unfortunately, recurrences have been reported following both procedures. A 100-patient meta-analysis of retrospective surgical series found a 7% failure rate associated with appendectomy alone and a 10% failure rate with more extended resection of adenocarcinoid tumors.64 Several investigators have noted that metastases in patients with adenocarcinoid primaries closely resemble adenocarcinomas, and have lost their neuroendocrine features. Hirschfield and colleagues found no carcinoid features in a Krukenberg tumor that had apparently arisen from a goblet cell primary.55 Two other series also report adenocarcinomas arising in the ovaries of patients with appendiceal adenocarcinoid primary tumors.53,57 On the basis of these findings, patients with metastatic adenocarcinoid tumors are most commonly treated with chemotherapy regimens used for advanced colorectal cancer; there has been one report of complete and persistent remission of a metastatic adenocarcinoid tumor following treatment with 5-FU, leucovorin, and oxaliplatin.65 An approach incorporating surgical debulking and intraperitoneal chemotherapy has also been utilized in patients with metastatic disease. In one series, the overall median survival for patients with carcinomatosis from appendiceal adenocarcinoid tumors who had undergone surgery and intraperitoneal chemotherapy was 18.5 months.63
414
GASTROINTESTINAL TUMORS
CARCINOID TUMORS OF THE APPENDIX Carcinoid tumors have been categorized based on their site of origin, and are classified as foregut (lungs and bronchi), midgut (small bowel and appendix), or hindgut tumors (rectum). While in the past, the often indolent nature of carcinoid tumors has resulted in their being considered benign lesions, their malignant potential is now clearly evident. Historically, carcinoid tumors have been considered the most common tumor of the appendix, accounting for over 50% of all appendiceal malignancies and discovered in up to 7 of every 1000 appendectomy specimens.2,6,7 Data from an early tumor registry, the End Results Group, which collected data in the United States from 1950 to 1969, reported that appendiceal carcinoid tumors comprised 44% of all carcinoid tumors. A subsequent registry, the Third National Cancer Survey (1969–1971) reported that appendiceal carcinoids comprised 35.5% of all carcinoids.66 More recent data from the SEER program of the National Cancer Institute indicate that appendiceal carcinoids comprise only 2.4% of all carcinoid tumors.67 This apparent decline in the incidence of appendiceal carcinoids, however, should be viewed with caution, and may substantially reflect changes in reporting methodology. Specifically, while the earlier registries reported carcinoids with both “benign” and “malignant” features, the current SEER database captures only “malignant” carcinoid tumors. A decreasing incidence of incidental appendectomy during surgery for other indications may have also influenced the recent decrease in the diagnosis of appendiceal carcinoid tumors. In older surgical series, appendiceal carcinoids were most commonly discovered during incidental appendectomy for other reasons. The most common indication for appendectomy in a 1968 series of 137 appendiceal carcinoid tumors was benign pelvic disease (43%) or gallbladder disease (35%).67 In recent series, however, the majority of appendiceal carcinoids are discovered during surgery for acute appendicitis.68 The carcinoid tumor itself is thought to be the cause of appendicitis in only a minority of these cases. Approximately two-thirds of appendiceal carcinoid tumors arise in the tip of the appendix, where they are unlikely to cause symptoms of obstruction. Less than 10% arise in the base, where they are more prone to obstruct the appendix, giving rise to acute appendicitis.7,69 Patients with appendiceal carcinoid tumors generally present at a relatively young age; the mean age at presentation in the recent SEER database analysis was 49 years, and in older series is even younger.7,67 While the early age at presentation may in part reflect the mean age at appendectomy, other explanations have also been proposed. Some investigators have speculated that the pattern of appendiceal carcinoid tumors parallels the biologic behavior of subepithelial neuroendocrine cells, from which appendiceal carcinoid tumors are thought to arise. The density of these cells tends to peak in the third decade of life and subsequently decreases throughout the remainder of life.70 – 72 Appendiceal carcinoids are more common in women than in men; in the SEER database, the male –female ratio was 0.82.67 The higher incidence of appendiceal carcinoids in
women, like the relatively young age at presentation, may be in part attributable to appendectomy patterns. Specifically, women are more likely to undergo incidental appendectomy as a result of gynecologic procedures. However, a female preponderance of appendiceal carcinoids has also been reported in children, an observation that cannot be explained by differences in appendectomy rates.73 – 75 The clinical behavior of appendiceal carcinoid tumors can be predicted based on the size of the tumor. Over 95% of appendiceal carcinoid tumors are less than 2 cm in diameter.7,76 The incidence of metastatic disease in such patients is extraordinarily low, although rare cases have been reported in the literature.69,76 – 80 In contrast, approximately one-third of patients with appendiceal carcinoid tumors measuring more than 2 cm in diameter have either nodal or distant metastases.81 Other criteria such as nodal invasion, perineural invasion, tumor location, and histologic pattern have not consistently correlated with the presence of metastatic disease. The surgical management of appendiceal carcinoids derives from these historical data, and surgical recommendations are based on tumor size. On the basis of the low incidence of metastases in patients whose tumors measure less than 2 cm, simple appendectomy is felt to be sufficient in such cases and is nearly always curative. In contrast, the higher incidence of metastases in patients whose tumors measure more than 2 cm in diameter has led to the recommendation for complete right hemicolectomy in this instance. These recommendations are supported by data from surgical series, one of the largest of which was published by Moertel and colleagues.81 Among 122 patients undergoing simple appendectomy for tumors measuring less than 2 cm, none developed disease recurrence, whereas one of 12 patients undergoing simple appendectomy for a tumor measuring more than 2 cm developed a local recurrence. Whether right colectomy decreases the risk of future distant recurrence has not been formally evaluated, and simple appendectomy for tumors more than 2 cm may still be considered in older patients or those with other major comorbidities. In the SEER registries, dating from 1973 to 1999, the 5-year survival rate for patients with regional metastases from appendiceal carcinoid tumors was 81–88%. Patients with distant metastatic spread had a 5-year survival rate of 10–30%.67 Patients with metastatic neuroendocrine tumors have a highly variable and sometimes unpredictable clinical course. Some patients may live for many years, and asymptomatic patients with metastatic neuroendocrine tumors can occasionally be followed without treatment. Such patients may be followed with serial CT scans as well as tumor markers. In patients with metastatic carcinoid tumors, 24hour collections of the serotonin metabolite 5HIAA may be useful in patient monitoring.82 Another tumor marker, chromogranin A (CgA) is a 49-kD protein that is contained in the neurosecretory vesicles of neuroendocrine tumor cells, and is detectable in the plasma of patients with endocrine neoplasms. Because it does not rely on serotonin secretion, serum CgA is a more broadly applicable marker than urinary 5HIAA, and is commonly used in the monitoring of patients with carcinoid tumors.83
CANCER OF THE APPENDIX
Aside from local lymph nodes, the liver is the most common site for metastases from appendiceal carcinoid tumors. In patients with a limited number of hepatic metastases, surgical resection has resulted in both long-term survival and in significant palliation of symptoms.84 – 86 Newer techniques, such as cryoablation and radiofrequency ablation have also been successful, both alone and in combination with surgery, in appropriately selected patients. The utility of hepatic transplantation in patients with metastatic neuroendocrine tumors remains uncertain. While high rates of 5-year survival have been reported, the majority of patients undergoing transplantation for neuroendocrine tumors in reported series have ultimately developed tumor recurrences.87 – 89 Hepatic arterial embolization may be used as a palliative technique in patients with hepatic metastases who are not candidates for surgical resection. Hepatic artery embolization is based on the principle that tumors in the liver derive most of their blood supply from the hepatic artery, whereas healthy hepatocytes derive most of their blood supply from the portal vein. Embolization of the hepatic arterial blood supply can be performed with or without the concurrent injection of chemotherapy, and is associated with tumor response rates that generally exceed 60%.90 Early studies reported a significant incidence of postembolization complications, which included renal failure, hepatic necrosis, and sepsis. Improved techniques have, in recent years, reduced the incidence of such complications, making embolization an important and generally safe treatment option for patients with neuroendocrine tumors.91 Patients with metastatic appendiceal carcinoids may first become symptomatic from symptoms of hormonal hypersecretion rather than from symptoms related to tumor bulk. In patients with appendiceal carcinoid tumors, the secretion of serotonin and other vasoactive substances leads to the development of the carcinoid syndrome.7 This syndrome is characterized by episodic flushing, secretory diarrhea, and, less commonly, the development of wheezing and right-sided valvular heart disease. The carcinoid syndrome, as well as other hormonal syndromes associated with neuroendocrine tumors, can often be well controlled with somatostatin analogs. In an initial study, the subcutaneous administration of the somatostatin analog octreotide, administered at a dosage of 150 µg three times a day, improved the symptoms of carcinoid syndrome in 88% of patients.92 More recently, the use of a long-acting depot form of octreotide, which can be administered on a monthly basis, has largely obviated the need for patients to inject themselves on a daily basis.93 Therapy with low-dose alpha interferon has been reported to result in biochemical responses in approximately 40% of patients with metastatic neuroendocrine tumors.94 The addition of alpha interferon to therapy with somatostatin analogs has also been reported to be effective in controlling symptoms in patients who may be resistant to somatostatin analogs alone.95 The more widespread acceptance of alpha interferon in the treatment of metastatic neuroendocrine tumors has been limited, however, by studies challenging its efficacy as well as the potential for side effects, which may include myelosuppression, fatigue, and depression.96
415
A number of chemotherapeutic regimens have been utilized in the treatment of patients with metastatic carcinoid tumors. In an initial study, performed by the Eastern Cooperative Oncology Group, patients with metastatic carcinoid tumors were randomized to receive streptozocin in combination with either 5-FU or cyclophosphamide.97 Tumor responses, defined as either radiologic regression or decreases in biochemical markers, occurred in 33% of the patients treated with streptozocin/5FU and in 26% of the patients treated with streptozocin/cyclophosphamide. The toxicity associated with streptozocin/5FU was felt to be prohibitive, prompting a second trial in which the dosing interval between cycles was lengthened. In this second randomized trial, the response rate associated with the streptozocin/5FU combination decreased to 22%, as compared to 21% for patients treated with doxorubicin alone.98 Dacarbazine (DTIC) has been evaluated as a potential alternative to streptozocin-based therapy, and in one phase II study, treatment with DTIC was associated with an objective radiologic response rate of 16% in 56 patients with metastatic carcinoid tumors.99 As with the streptozocin-based regimens, the relatively limited response rates, together with concerns about toxicity, have limited the use of DTIC in the treatment of carcinoid tumors. Newer chemotherapeutic agents have, to date, proved relatively inactive in neuroendocrine tumors. High dose paclitaxel, administered with granulocyte-colony stimulating factor, was evaluated in 24 patients with metastatic carcinoid and islet cell tumors.100 Significant hematologic toxicity was observed, and the objective radiologic response rate was only 8%. Treatment with docetaxel was associated with biochemical responses but no radiologic responses in a recent phase II trial of 21 patients with carcinoid tumors.101 No responses were observed in 19 neuroendocrine tumor patients treated with gemcitabine.102
REFERENCES 1. Aho A, Heinonen R, Lauren P. Benign and malignant mucocele of the appendix: histologic types and prognosis. Acta Chir Scand 1973; 139: 254 – 6. 2. Collins D. 71,000 appendectomy specimens: a final report summarizing 40 years of study. Am J Proctol 1963; 14: 365 – 81. 3. Cortina R, et al. Management and prognosis of adenocarcinoma of the appendix. Dis Colon Rectum 1995; 38: 848 – 52. 4. Nitecki S, et al. The natural history of surgically treated primary adenocarcinoma of the appendix. Ann Surg 1994; 219: 51 – 7. 5. McCusker M, et al. Primary malignant neoplasms of the appendix: a population based study from the Surveillance, Epidemiology and End Results program 1973 – 1998. Cancer 2002; 94(12): 3307 – 12. 6. Lyss A. Appendiceal malignancies. Semin Oncol 1988; 15: 129 – 37. 7. Moertel C, Dockerty M, Judd E. Carcinoid tumors of the vermiform appendix. Cancer 1968; 21: 270 – 8. 8. Carr N, McCarthy W, Sobin L. Epithelial noncarcinoid tumors and tumor-like lesions of the appendix. Cancer 1995; 75: 757 – 68. 9. Wolff M, Ahmed N. Epithelial neoplasms of the vermiform appendix (exclusive of carcinoid) I: adenocarcinoma of the appendix. Cancer 1976; 37: 2493 – 510. 10. Misdraji J, et al. Appendiceal mucinous neoplasms:a clinicopathologic analysis of 107 cases. Am J Surg Pathol 2003; 27(8): 1089 – 103. 11. Fernandez R, Daly J. Pseudomyxoma peritonei. Arch Surg 1980; 115: 409 – 14. 12. Smith J, et al. Pseudomyxoma of appendiceal origin: the Memorial Sloan-Kettering Cancer Center experience. Cancer 1992; 70: 396 – 401.
416
GASTROINTESTINAL TUMORS
13. Gough D, Donohue J, Schutt A. Pseudomyxoma peritonei: long term patient survival with an aggressive surgical approach. Ann Surg 1994; 219(112 – 19). 14. Long R, Spratt J, Dowling E. Pseudomyxoma peritonei: new concepts in management with a report of seventeen patients. Am J Surg 1969; 117: 162 – 9. 15. Hinson F, Ambrose N. Pseudomyxoma peritoneii. Br J Surg 1998; 85: 1332 – 9. 16. Costa M. Pseudomyxoma peritonei: histologic predictors of patient survival. Arch Pathol Lab Med 1994; 118: 1215 – 9. 17. McCarthy J, Aga A. A fallopian tube lesion of borderline malignancy associated with pseudomyxoma peritonei. Histopathology 1998; 13: 223 – 5. 18. Kurita M, Komatsu H, Hata Y. Pseudomyxoma peritonei due to adenocarcinoma of the lung: a case report. J Gastroenterol 1994; 29: 344 – 8. 19. Prayson R, Hart W, Petras R. Pseudomyxoma peritonei: a clinicopathologic study of 19 cases with emphasis on site of origin and nature of associated ovarian tumors. Am J Surg Pathol 1994; 18: 591 – 603. 20. Mukherjee A, et al. Pseudomyxoma peritonei usually originates from the appendix: a review of the evidence. Eur J Gynaecol Oncol 2004; 25(4): 411 – 4. 21. Ronnett B, et al. Disseminated peritoneal adenomucinosis and peritoneal mucinous carcinomatosis: a clinicopathologic analysis of 109 cases with emphasis on distinguishing pathologic features, site of origin, prognosis and relationship to “pseudomyxoma peritonei”. Am J Surg Pathol 1995; 19: 1390 – 408. 22. Sugarbaker P. Pseudomyxoma peritonei: a cancer whose biology is characterized by a redistribution phenomenon. Ann Surg 1994; 219: 109 – 11. 23. Seshul M, Coulam C. Pseudomyxoma peritonei: computed tomography and sonography. AJR Am J Roentgenol 1981; 136: 803 – 6. 24. Culliford A, et al. Surgical Debulking and intraperitoneal chemotherapy for established peritoneal metastases from colon cancer and appendix cancer. Ann Surg Oncol 2001; 8(10): 787 – 95. 25. Glehen O, Mohamed F, Sugarbaker P. Incomplete cytoreduction in 174 patients with peritoneal carcinomatosis from appendiceal malignancy. Ann Surg 2004; 240(2): 278 – 85. 26. Sugarbaker P, Chang D. Results of treatment of 385 patients with peritoneal surface spread of appendiceal malignancy. Ann Surg Oncol 1999; 6: 727 – 31. 27. Sugarbaker P. Managing the peritoneal surface component of gastrointestinal cancer. Part 1: patterns of dissemination and treatment options. Oncology (Williston Park) 2004; 18: 51 – 9. 28. Sugarbaker P. Managing the peritoneal surface component of gastrointestinal cancer. Part 2. Perioperative intraperitoneal chemotherapy. Oncology (Williston Park) 2004; 18: 207 – 19. 29. Sugarbaker P, et al. Peritoneal carcinomatosis from appendiceal cancer: results in 69 patients treated by cytoreductive surgery and intraperitoneal chemotherapy. Dis Colon Rectum 1993; 36: 323 – 9. 30. Verwaal V, et al. Toxicity of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. J Surg Oncol 2004; 85(2): 61 – 7. 31. Levitz J, et al. Unusual abdominal tumors, case 1. Pseudomyxoma peritonei: response to capecitabine. J Clin Oncol 2004; 22: 1518 – 20. 32. Cerame M. A 25-year review of adenocarcinoma of the appendix. Dis Colon Rectum 1988; 31: 145 – 50. 33. Lenriot J, Huguier M. Adenocarcinoma of the appendix. Am J Surg 1988; 155: 470 – 5. 34. Didolkar M, Fanous N. Adenocarcinoma of the appendix: a clinicopathologic study. Dis Colon Rectum 1977; 20: 130 – 4. 35. Schlatter M, et al. Primary appendiceal adenocarcinoma. Am Surg 1987; 53: 434 – 7. 36. Harris G, Urdaneta L, Mitros F. Adenocarcinoma of the vermiform appendix. J Surg Oncol 1990; 44: 218 – 24. 37. Andre T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med 2004; 350: 2343 – 51. 38. Twelves C, Wong A, Nowacki M, et al. Capecitabine as adjuvant treatment for stage III colon cancer. N Engl J Med 2005; 352: 2696 – 704.
39. Willett C, et al. Postoperative radiation therapy for high-risk colon carcinoma. J Clin Oncol 1993; 11: 1112 – 7. 40. Lin J, et al. Superficial spreading adenocarcinoma of appendix, cecum and terminal ileum. Dis Colon Rectum 1980; 23: 587 – 9. 41. Flint F, Kahn A, Passaro E. Adenocarcinoma of the appendix. Surgery 1970; 120: 707 – 9. 42. Pugeda F, Hishaw J. Primary adenocarcinoma of the appendix. Dis Colon Rectum 1969; 12: 457 – 61. 43. Reichle F, et al. Adenocarcinoma of the vermiform appendix. Am Surg 1969; 37: 344 – 50. 44. Subbuswamy G, et al. Goblet cell carcinoid of the appendix. Cancer 1974; 34: 338 – 44. 45. Isaacson P. Crypt cell carcinoma of the appendix. Am J Surg Pathol 1981; 5: 213 – 24. 46. Chefjeck G, et al. Amphicrine cells, dysplasias and neoplasias. Cancer 1985; 56: 2683 – 90. 47. Warkel R, Cooper P, Helwig E. Adenocarcinoid, a mucin producing carcinoid tumor of the appendix: a study of 39 cases. Cancer 1978; 42: 2781 – 93. 48. Cooper P, Warkel R. Ultrastructure of the goblet cell type of adenocarcinoid of the appendix. Cancer 1978; 42: 2687 – 95. 49. Hofler H, Kloppel G, Heitz P. Combined production of mucus, amines, and peptides by goblet-cell carcinoids of the appendix. Pathol Res Pract 1984; 178: 555 – 61. 50. Burke A, et al. Goblet cell carcinoids and related tumors of the vermiform appendix. Am J Clin Pathol 1990; 94: 27 – 35. 51. Edmonds P, et al. Adenocarcinoid (mucinous carcinoid) of the appendix. Gastroenterology 1984; 86(302 – 9). 52. Bak M, Jorgenson L. Adenocarcinoma of the appendix presenting with metastases to the liver. Dis Colon Rectum 1987; 30: 112 – 5. 53. Haqqani M, Williams G. Mucin producing carcinoid tumors of the vermiform appendix. J Clin Pathol 1977; 30: 473 – 80. 54. Heisterberg L, Wahlin A, Nielson K. Two cases of goblet cell carcinoid of the appendix with bilateral ovarian metastases. Acta Obstet Gynecol Scand 1982; 61: 153 – 6. 55. Hirschfield L, et al. Adenocarcinoid of the appendix presenting as bilateral Krukenberg’s tumors of the ovaries. Immunohistochemical and ultrastructural studies and literature review. Arch Pathol Lab Med 1985; 109: 930 – 3. 56. Jones R, McFarlane A. Carcinomas and carcinoid tumors of the appendix in a district general hospital. J Clin Pathol 1976; 29: 687 – 92. 57. Miller R, Sarikaya H, Jenison E. Adenocarcinoid tumor of the appendix presenting as unilateral Krukenberg tumor. J Surg Oncol 1988; 37: 65 – 71. 58. Merino M, et al. Appendiceal carcinoma metastatic to the ovaries am mimicking primary ovarian tumors. Int J Gynecol Pathol 1985; 4: 110 – 20. 59. Thomas R, et al. Krukenberg tumor of the ovary from an occult appendiceal primary: case report and review of the literature. Obstet Gynecol 1985; 65: 95S – 8S. 60. Zirkin R, Brown S, Hertz M. Adenocarcinoid of the appendix presenting as bilateral ovarian tumors: a case report with histochemical and ultrastructural studies. Diagn Gynecol Obstet 1980; 2: 269 – 74. 61. Hood I, Jones B, Watt J. Mucinous carcinoid tumor of the appendix presenting as bilateral ovarian tumors. Arch Pathol Lab Med 1986; 110: 336 – 40. 62. Mandai M, et al. Krukenberg tumor from an occult appendiceal adenocarcinoid: a case report and review of the literature. Eur J Obstet Gynecol Reprod Biol 2001; 97: 90 – 5. 63. Mahteme H, Sugarbaker P. Treatment of peritoneal carcinomatosis from adenocarcinoid of appendiceal origin. Br J Surg Soc 2004; 91(9): 1168 – 73. 64. Varisco B, et al. Adenocarcinoid of the appendix: is right hemicolectomy necessary? A meta-analysis of retrospective chart reviews. Am Surg 2004; 70(7): 593 – 9. 65. Garin L, et al. Adenocarcinoid of the appendix veriformis: complete and persistent remission after chemotherapy (folfox) of a metastatic case. Dig Dis Sci 2002; 47(12): 2760 – 2. 66. Godwin J. Carcinoid tumors: an analysis of 2837 cases. Cancer 1975; 36: 560 – 9.
CANCER OF THE APPENDIX 67. Modlin I, Lye K, Kidd M. A 5 Decade Analysis of 13,715 Carcinoid Tumors. Cancer 2003; 97(4): 934 – 59. 68. Roggo A, Wood W, Ottinger L. Carcinoid tumors of the appendix. Ann Surg 1993; 217: 385 – 90. 69. Syracuse D, et al. Carcinoid tumors of the appendix: mesoappendiceal extension and nodal metastases. Ann Surg 1979; 190: 58 – 63. 70. Lundqvist M, Wilander E. Subepithelial neuroendocrine cells and carcinoid tumors of the human small intestine and appendix. A comparative immunohistochemical study with regard to serotonin, neuron specific enolase and S-100 protein reactivity. J Pathol 1986; 148: 141 – 7. 71. Lundqvist M, Wilander E. A study of the histopathogenesis of carcinoid tumors of the small intestine and the appendix. Cancer 1987; 60: 201 – 6. 72. Shaw P. The topographical and age distributions of neuroendocrine cells in the normal human appendix. J Pathol 1991; 164: 235 – 9. 73. Moertel C, Weiland L, Elander R. Carcinoid tumor of the appendix in the first two decades of life. J Pediatr Surg 1990; 25: 1073 – 5. 74. Jonsson T, Johannsson J, Hallgrimsson J. Carcinoid tumors of the appendix in children younger than 16 years. Acta Chir Scand 1989; 155: 113 – 6. 75. Parks S, Muir K, Alsheyyab M. Carcinoid tumors of the appendix in children: 1957 – 1985. Br J Surg 1993; 80: 502 – 4. 76. Anderson J, Wilson B. Carcinoid tumors of the appendix. Br J Surg 1985; 72: 545 – 6. 77. Bowman G, Rosenthal D. Carcinoid tumors of the appendix. Am J Surg 1983; 146: 700 – 3. 78. Thirlby R, Kasper C, Jones R. Metastatic carcinoid tumor of the appendix: report of a case and review of the literature. Dis Colon Rectum 1984; 27: 42 – 6. 79. MacGillivray D, et al. Distant metastases from a carcinoid tumor of the appendix less than one centimeter in size. Surgery 1992; 111: 466 – 71. 80. Pearlman D, Srinivasan K. Malignant carcinoid of the appendix: metastases from a small primary tumor which appeared as appendiceal intussusception. N Y State J Med 1971; 71: 1529 – 31. 81. Moertel C, et al. Carcinoid tumor of the appendix: treatment and prognosis. N Engl J Med 1987; 317: 1699 – 701. 82. Feldman J, O’Dirisio T. Role of neuropeptides and serotonin in the diagnosis of carcinoid tumors. Am J Med 1986; 81(suppl 6B): 41 – 8. 83. Eriksson B, Oberg K, Stridsberg M. Tumor markers in neuroendocrine tumors. Digestion 2000; 62(suppl 1): 33 – 8. 84. Que F, et al. Hepatic resection for metastatic neuroendocrine carcinomas. Am J Surg 1995; 169: 36 – 43.
417
85. McEntee G, et al. Cytoreductive hepatic surgery for neuroendocrine tumors. Surgery 1990; 108: 1091 – 6. 86. Norton J, et al. Aggressive surgery for metastatic liver neuroendocrine tumors. Surgery 2003; 134: 1057 – 65. 87. Lang H, Oldhafer K, Weimann A, et al. Liver transplantation for metastatic neuroendocrine tumors. Ann Surg 1997; 225: 347 – 54. 88. LeTreut Y, et al. Results of liver transplantation in the treatment of metastatic neuroendocrine tumors: a 31-case French multicentric report. Ann Surg 1997; 225: 355 – 64. 89. Olausson M, et al. Indications and results of liver transplantation in patients with neuroendocrine tumors. World J Surg 2002; 26: 998 – 1004. 90. Venook A. Embolization and chemoembolization therapy for neuroendocrine tumors. Curr Opin Oncol 1999; 11: 38 – 41. 91. Gupta S, et al. Hepatic artery embolization and chemoembolization for treatment of patients with metastatic carcinoid tumors: the MD Anderson experience. Cancer J 2003; 9: 261 – 7. 92. Kvols L, et al. Treatment of the malignant carcinoid syndrome: evaluation of a long-acting somatostatin analog. N Engl J Med 1986; 315: 663 – 6. 93. Rubin J, et al. Octreotide acetate long-acting formulation versus open-label subcutaneous octreotide acetate in malignant carcinoid syndrome. J Clin Oncol 1999; 17: 600 – 6. 94. Oberg K, Eriksson B. The role of interferons in the management of carcinoid tumors. Acta Oncol 1991; 30: 519 – 22. 95. Janson E, Oberg K. Long term management of the carcinoid syndrome: treatment with octreotide alone and in combination with alpha-interferon. Acta Oncol 1993; 32: 225 – 9. 96. Valimaki M, et al. Is the treatment of metastatic carcinoid tumor with interferon not as successful as suggested? Cancer 1991; 67: 547 – 9. 97. Moertel C, Hanley J. Combination chemotherapy trials in metastatic carcinoid tumor and the malignant carcinoid syndrome. Cancer Clin Trials 1979; 2: 327 – 34. 98. Engstrom P, et al. Streptozocin plus fluorouracil versus doxorubicin therapy for metastatic carcinoid tumor. J Clin Oncol 1984; 2: 1255 – 9. 99. Bukowski R et al., A Southwest Oncology Group Study. Phase II trial of dimethyltriazenoimidazole carboxamide in patients with metastatic carcinoid. Cancer 1994; 73: 1505 – 8. 100. Ansell S, et al. A phase II study of high-dose paclitaxel in patients with advanced neuroendocrine tumors. Cancer 2001; 91: 1543 – 8. 101. Kulke M, et al. Phase II study of docetaxel in patients with metastatic carcinoid tumors. Cancer Invest 2004; 22(3): 353 – 9. 102. Kulke M, et al. A Phase II trial of gemcitabine for metastatic neuroendocrine tumors. Cancer 2004; 101(5): 934 – 9.
Section 6 : Gastrointestinal Tumors
37
Gastrointestinal Stromal Tumors Margaret von Mehren and Douglas Flieder
HISTORICAL BACKGROUND Gastrointestinal Stromal Tumor (GIST) are uncommon tumors of the intestinal tract. These tumors are of mesenchymal origin. Prior to their recognition as a distinct biologic subtype, they were termed leiomyosarcomas, leiomyomas, or leiomyoblastomas. Mazur and Clark determined that these tumors contained both smooth muscle and neural features and reclassified them as GIST.1 Subsequently, these tumors were found to express CD34 and KIT, further aiding in their classification.2,3 Importantly, Hirota and colleagues determined that a majority of tumors contained mutations of KIT leading to constitutive activation of the molecule.4 These three findings aided pathologists in making the correct diagnosis and the selection of appropriate patients for KIT-targeted therapy. Importantly, KIT-targeted therapy has significantly changed the management and prognosis of GIST patients.
ANATOMY GISTs arise typically along the intestinal tract. The most common site of primary tumors is the stomach (39–70%), followed by the small intestine (31–45%), colon, rectum, and anus (10–16%), mesentery and peritoneum (8%), with rare cases arising in the esophagus.5 – 9 Case reports in the literature also describe primary tumors of the duodenal ampulla,10 appendix,11 gallbladder, and urinary bladder.12 Metastatic disease is most commonly found in the liver as well as the peritoneum and omentum. Less common sites of disease include lung and bone.
EPIDEMIOLOGY The incidence of GIST is being clarified as pathologists, surgeons, and medical oncologists become increasingly familiar with this disease entity. A population-based study of GIST analyzing tumors identified in Iceland from 1990 to 2003 estimated the annual incidence to be 1.1 per 100 000.13 A population-based study of western Sweden, with a population of 1.3–1.6 million, identified 288 cases of primary GIST between 1983 and 2000, with an annual incidence of 14.5
per million.14 It estimated the overall prevalence to be 129 per million. Another study evaluating the SEER database of cases diagnosed between 1992 and 2000 determined the age-adjusted yearly incidence rate as 0.68 per 100 000.15 The median age at diagnosis in these cohorts was 65.8 and 63, with some studies suggesting an increased incidence among men. The SEER database analysis also suggested a higher incidence in blacks. Prognostic factors associated with a poor outcome were older age, black race, advanced stage, and lack of therapy. The majority of tumors are sporadic in nature, although there have been reports of familial tumors associated with a germline KIT mutation.16 – 20 These reports suggest an earlier age of onset than is seen in sporadic disease, some as early as late teens. In some cases, there have been multiple GISTs or GIST associated with hyperplasia of the interstitial cells of Cajal, the pacemaker cells of the gut, which are thought to be the cell of origin of GIST. In addition, familial GIST patients have been reported to have skin hyperpigmentation, pigmented macules, and/or skin mastocytosis. Another family was characterized by multiple autonomic nerve tumors as well as neuronal hyperplasia within the small intestine; tumors proved to be KIT positive and associated with KIT mutations. Carney’s triad, a syndrome of pulmonary chondromas, GIST, and functioning extra adrenal-paraganglioma,21 has been reported. Tumors again appear to occur in a younger age-group and have an indolent course; however, the syndrome does not appear to be familial. Patients with neurofibromatosis-1 also have been noted to have GIST.22
PATHOLOGY Gross Features GIST can develop in any portion of the tubular gastrointestinal tract within any portion of the gut wall. They are most often centered within the submucosa or muscularis propria. Some tumors are mostly extramural, while very large neoplasms can extend to or invade adjacent organs. GISTs are well circumscribed, yet may be multinodular. Overlying mucosa may be intact or ulcerated regardless of
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
GASTROINTESTINAL STROMAL TUMORS
419
the aggressiveness of the tumor. Cut surfaces are usually tan, smooth, and lobulated with either whorled or granular appearances. Tumors may feature hemorrhage, necrosis, calcification, and/or cystic change.
carcinoid-like growth pattern, signet-ring cell features, granular cell changes, oncocytic cytoplasmic changes, crystalloid formation, osteoclast-like giant cells, tumor giant cells, and, although quite rare, notable cytologic pleomorphism.
Microscopic Features
Immunohistochemical Features
A diagnosis of GIST is often suspected histologically since the majority of cases have uniform appearances falling into one of three categories: spindle cell type, epithelioid type, or mixed type. Interestingly, the majority of gastric GISTs feature an epithelioid phenotype, while small intestinal, anorectal, colonic, and esophageal tumors more often feature spindle cell morphology. GISTs with spindle cell morphology are composed of relatively uniform cells arranged in whorls or short fascicles. Individual cells have light pink (eosinophilic) cytoplasm with indistinct cellular borders. Nuclei are uniform and spindled with open evenly distributed chromatin and inconspicuous nucleoli. Perinuclear cytoplasmic vacuoles that appear to indent the nuclei are seen in up to 5% of cases and are a characteristic feature of gastric GIST (see Figure 1a). These vacuoles are an artifact of fixation since they are not present in frozen section specimens. Stromal hyalinization and/or calcification are not uncommon features, and thin-walled vessels may be abundant resulting in stromal hemorrhage. Hypercellularity, tumor cell necrosis, and brisk mitotic activity may be seen. GISTs of epithelioid type feature rounded cells with either pink (eosinophilic) or clear cytoplasm arranged in sheets or nests (see Figure 1b). In tumors with clear cytoplasm, a condensed perinuclear rim of pink cytoplasm is often observed. Nuclei are usually eccentrically placed and are round with small nucleoli and evenly distributed chromatin. While diffuse cellular pleomorphism is associated with malignant behavior, scattered bizarre cells are more commonly seen in indolent neoplasms. Mitotic counts may not be particularly helpful in discerning the aggressiveness of epithelioid GIST. They often have an indolent morphology with morphologic heterogeneity, which upon extensive sampling will reveal malignant components within a largely “benign” tumor. Tumors with mixed cell types may feature a commingling of spindle cell and epithelioid cells or an abrupt transition between the patterns. The recognition of this pattern simply emphasizes the unique nature of GIST and supports the distinction from smooth-muscle tumors as well as neurogenic lesions of the gastrointestinal tract. Up to 20% of GISTs, whether spindle cell or epithelioid, feature stromal hyaline or fibrillary pink material referred to as skeinoid fibers. These globular and elongated nodular tangles of collagen fibers are seen in a large percentage of small intestinal GIST and, though once thought to indicate neural differentiation, are currently believed to have no histogenetic significance.23 Rare morphologic findings also seen in GIST include prominent myxoid stroma, nested so-called paragangliomalike growth (especially in the small intestinal neoplasms),
Discussing the immunohistochemical profile of GIST requires a basic compromise on defining pathologic criteria for an entity. While some authorities require CD117 positivity of tumor cells in order to diagnose a tumor as a GIST, this is unreasonable as are nearly all proposals to define an entity on the basis of a single marker. Rare tumors with all the morphologic and phenotypic features of GIST, which do not stain for CD117, should still be diagnosed as GIST, while tumors that share few if any morphologic features of GIST should not be designated as GIST simply on account of CD117 positivity.24 Against this background, 90–95% of GIST from all sites, regardless of their morphologic features and degree of malignancy, show strong cytoplasmic CD117 staining25 (see Figure 1c). Prominent membranous reactivity is the most convincing pattern of staining since other mesenchymal tumors such as fibromatosis (desmoid tumor) may stain with CD117 in a coarse granular cytoplasmic pattern. Up to 50% of GIST show cytoplasmic dotlike (golgi pattern) staining that often coexists with the more diffuse cytoplasmic pattern. While most GISTs show CD117 positivity in 90% or more of tumor cells, a small percentage of tumors show focal staining, that is, only 5 to 20% of tumor cells react with anti-CD117 antibodies. The therapeutic relevance of limited immunoreactivity appears to be unimportant; however, antigen retrieval techniques should not be performed as this increases the false positive rate for CD117 staining. Also of note, the presence of CD117 immunoreactivity in a tumor (GIST or otherwise) does not necessarily correlate with a ckit gene mutation. Conversely, a CD117 mutation can exist in the absence of CD117 immunohistochemical staining. The ubiquitous immunohistochemical marker CD34 stains up to 70% of spindle cell and epithelioid GIST, and at one time was considered a reproducible marker of GIST. Staining is most consistently seen in colorectal and esophageal lesions. However, Schwann cell neoplasms and a proportion of true smooth-muscle tumors also show CD34 positivity. Myoid marker reactivity is also seen in up to 30% of GIST and most often in small intestinal tumors. Most of these positive reactions have been reported with smooth-muscle actin, either focally or diffusely, while desmin is found in less than 5% of GIST with staining usually limited to epithelioid tumors or only scattered spindle tumor cells. Caldesmon is also detected in a significant subset of GIST. Neural antigens can also stain GIST. S-100 protein may be seen focally in up to 10% of tumors; yet these lesions are usually negative for neurofilament protein and glial fibrillary acidic protein. Lastly, malignant epithelioid GIST can feature scattered immunoreactivity with cytokeratins; but the presence of CD117 positivity strongly mitigates against a diagnosis of carcinoma. Novel markers that have been identified, which appear to have specificity for GIST, are DOG126 and
420
GASTROINTESTINAL TUMORS
(a)
(b)
(c) Figure 1 Histologic sections of GIST tumors. (a) Spindle cell GIST. Spindle cells with inconspicuous cytoplasmic borders and bland nuclei feature scattered perinuclear vacuoles (H&E, 60× original magnification). (b) Epithelioid GIST. Clusters of neoplastic cells with round nuclei, small nucleoli, and abundant pink (eosinophilic) cytoplasm cluster in small groups (H&E, 60× original magnification). (c) CD117 immunohistochemical staining of GIST. All neoplastic cells feature cytoplasmic staining as well as membranous reactivity (H&E, 60× original magnification).
nPKCθ .27 At present, these are not used in routine pathologic evaluation of GIST.
Microscopic Differential Diagnosis Given the wide morphologic spectrum of GIST, the morphologic differential diagnosis includes solitary fibrous tumor, fibromatosis, inflammatory fibroid tumor, glomus tumor, schwannoma/malignant peripheral nerve sheath tumor, leiomyoma/leiomyosarcoma, and even carcinomas
and lymphomas. Fibromatosis is probably the most difficult differential since it can involve the gastrointestinal wall and, as mentioned above, stains for CD117. In addition, fibromatosis can develop as a postsurgical complication of GIST resection. The distinction between GIST and leiomyoma/leiomyosarcoma along with schwannoma/malignant peripheral nerve sheath tumor is quite difficult given the smooth muscle and/or neural features of GIST. The practical approach is to diagnose a tumor as either a smooth-muscle tumor or neural
GASTROINTESTINAL STROMAL TUMORS
tumor if the typical morphologic and immunohistochemical features are seen and the neoplasm lacks CD117 immunoreactivity.
Criteria for Malignancy Unlike many other mesenchymal tumors that are diagnosed as either benign or malignant, GIST tumors often defy such prognostication. Instead, surgical pathologists comment on the risk of aggressive behavior on the basis of reproducible pathologic features. A recent National Institutes of Health workshop on GIST proposed that tumors could have very low risk, low risk, intermediate risk, or high risk on the basis of tumor size (single largest dimension) and mitotic count (number of mitotic figures per 50 high-power fields [hpfs])24,28 (see Table 1). Though easy to apply and fairly reproducible, this approach is not universally accepted. Critics find this method overly simplistic since it ignores other morphologic features thought to be of prognostic significance and does not account for differences among GISTs in different anatomic sites. For example, as a group, a higher percentage of small intestinal tumors are malignant compared to gastric neoplasms.29
421
Table 1 National Institutes of Health Gastrointestinal Stromal Tumor Workshop: proposed approach for defining risk of aggressive behavior.
Risk
Size (cm)
Mitotic count (per 50 hpfs)
Very low risk Low risk Intermediate risk
<2 2–5 <5 5 – 10 >5 >10 Any size
<5 <5 6 – 10 <5 >5 Any mitotic rate >10
High risk
KIT
Exon 9 Exon 11 Exon 13
Juxtamembrane
PDGFRA
Exon 12
TK1 domain Kinase
Molecular Pathology GISTs typically express KIT, and commonly contain mutations in the KIT gene. KIT is a member of the tyrosine kinase type III family that also includes platelet-derived growth factor receptor (PDGFR). In the minority of tumors that are KIT negative, approximately 30% have PDGFRα mutations.30 The protein structures of KIT and PDGFR consist of an extracellular domain with five immunoglobulin-like domains, a transmembrane domain followed by two tyrosine kinase domains. Mutations in KIT have been diagnosed most commonly in exon 11, followed in frequency by exons 9, 13, and 17 (see Figure 2). In PDGFRα, mutations are found primarily in exons 12 and 18. These mutations occur in the transmembrane domain (KIT exons 9 and 11 and PDGFRα exon 12), and tyrosine kinase domains (KIT exons 13 and 17 and PDGFRα exon 18). Mutations lead to constitutive activation of the protein and tumor formation. Recently, tumors arising outside of the stomach have been shown to have a higher proportion of KIT exon 9 mutations.31 Interestingly, GISTs in the pediatric population are very rare, and do not commonly contain mutations in KIT or PDGFRα,32 with only one recent report describing a mutation in exon 9.33 In addition, Carney’s triad, a syndrome encompassing pulmonary chondromas, GIST, and functioning extra adrenalparaganglioma,21 as well as neurofibromatosis-1 associated GIST do not contain KIT mutations.22
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS GIST patients can present asymptomatically. Some tumors are identified on EGD as submucosal lesions. Ultrasonic criteria have been identified, which increase the risk for a submucosal lesion being malignant GIST.34 Occasionally, patients who have had abdominal pelvic CT scans for
Exon 17
TK2 domain
Exon 18
Figure 2 Depiction of KIT and PDGFRα tyrosine kinase receptors. The area of the receptor where the most common exon mutations occur is shown.
other reasons have been identified with gastric, intestinal, or peritoneal masses, which subsequently prove to be GISTs. Patients can present with various symptoms, some of which are dependent on the location of the primary tumor.5 – 8 The most common symptoms are pain, a palpable mass, and bleeding, with the latter two being somewhat more common in tumors that arise in the small or large intestine. Obstructive symptoms are not seen with gastric primaries but can be seen in patients with tumors arising in the more distal portions of the intestinal tract. Other symptoms include nausea, vomiting, early satiety, and fever. Patients presenting symptomatically typically undergo endoscopic evaluation and/or CT scans. Any abdominal mass is considered for resection. As GISTs are typically vascular and friable, percutaneous biopsies have the theoretic risk of intra-abdominal metastases. Endoscopic biopsies, however, can be performed safely to obtain a diagnosis because they do not transect the gut lumen into the peritoneal cavity. Patients with lesions not amenable to endoscopic biopsies should undergo resection without a definitive diagnosis. Routine laboratory studies including a complete blood count and a complete metabolic panel are performed for staging. Radiographic studies, typically CT or MRI, should assess for liver and mesenteric, peritoneal, or omental metastases. Lung metastases are uncommon, and a chest xray is sufficient for evaluation. Lastly, bone metastases are seen uncommonly. Thus, only patients with a history of bone pain or unexplained elevations in alkaline phosphatase or transaminases require bone scan evaluations. FDG PET scans
422
GASTROINTESTINAL TUMORS
have also been shown to be highly informative as GISTs are typically very PET avid; there are, however, tumors that have minimal glucose uptake on PET scan and therefore scans done following diagnosis when patients are on therapy should be interpreted with caution if no baseline study is available.
TREATMENT Surgery Surgery is the mainstay of therapy for GIST. Although there are reports on the use of laparoscopic surgery for the resection of GISTs,35,36 at the present time, such an approach is not considered a standard of care because tumors are highly vascular and friable, again potentially increasing the risk for intra-abdominal spreading. Complete en bloc surgical resection of the primary tumor is the goal of surgery. A lymph node dissection is not required given the infrequency of metastases to regional lymph nodes. Examination of the peritoneal surfaces and the liver is warranted, as they are common sites of metastases. Neoadjuvant therapy with imatinib mesylate in patients with large primary tumors that will require a procedure with excess morbidity should be considered. This setting is one in which a PET scan may be of particular utility because of its ability to give an early indication of disease response or progression. Patients with lack of response should be considered for early resection before progressive disease makes such a procedure no longer feasible.37 The data on the efficacy of surgery is retrospective in nature, having been collated following the identification of GIST as a distinct sarcoma from leiomyosarcomas.6,7,38,39 However, studies indicate that complete resections in all stages of disease, that is, primary presentation, locally recurrent disease and metastatic disease, lead to prolonged survival compared to patients who have incomplete resections. They also suggest that in up to 50% of patients presenting with primary disease the tumor will eventually recur. However, when the current risk stratification system utilizing size and mitotic rate of the tumor has been applied, the risk of death is increased in patients with high-risk and metastatic disease, but not in the other risk groups (see Table 1).14
Radiation Therapy Data on the role and benefit of radiation therapy is difficult to elucidate. Many intra-abdominal sarcomas have not traditionally been treated with radiation because of local tissue toxicities, and older series do not identify GIST patients. At the present time, radiation therapy is not considered effective in the primary setting. It has been utilized in the palliative setting to aid in the control of pain and bleeding. Some have suggested that there may be a role for radiation therapy for rectal GIST, particularly if such an approach would eliminate the need for an abdominal perineal resection and colostomy; a clinical trial will be needed to assess the benefits of such an approach.
Local Therapies Liver metastases have been managed with embolization with or without chemotherapy,40,41 and more recently with
cryotherapy or radiation frequency ablation. Approaches such as this have often been utilized in the palliative setting prior to the advent of imatinib mesylate, or in patients who have become refractory to imatinib mesylate therapy. Rajan and colleagues treated a group of sarcomas metastatic to the liver, 11 of which were characterized as gastrointestinal leiomyosarcomas and likely GISTs. Treatment led to stable disease (SD) in 69%, with a median survival of 13 months. Progression within the liver occurred at a median of 8 months, whereas disease outside of the liver at a median of 6 months. Patients now diagnosed with metastatic disease who are considered for resection of primary tumors may undergo localized therapies for limited liver metastases. Most of these patients are subsequently treated with imatinib therapy. The benefit of such local procedures is unclear at this time.
Chemotherapy Standard therapy for soft tissue sarcomas utilizes doxorubicin alone or in combination with other agents, most commonly ifosfamide or dacarbazine. Review of past clinical trials in soft tissue sarcomas to determine the response rate to standard chemotherapies is confounded by the inclusion of GIST into the leiomyosarcoma subgroup or no breakdown of the histologies included. A recent study by Edmonson and colleagues of dacarbazine, mitomycin, doxorubicin, and cisplatin enrolled two cohorts of patients: leiomyosarcomas and GIST.42 There was a 54% response rate in leiomyosarcomas compared with 4.9% response rate in GIST. In addition, doxorubicin and ifosfamide containing regimens have reported response rates of 0–27%, paclitaxel 7%, and gemcitabine 0% in GIST.43 GISTs have been found to have enhanced expression of multidrug resistance proteins compared with leiomyosarcomas that may explain the difference in chemotherapy sensitivity.44 The limited response rate to these therapies was associated with poor survival in patients with metastatic disease.
Targeted Therapy: Imatinib Mesylate The understanding of the oncogenesis of GIST has led to a dramatically different approach to treatment. KIT and PDGFRα provided biologically meaningful targets for therapy (reviewed in Refs 22, 28), which was aided by the development of imatinib mesylate (imatinib), an oral tyrosine kinase inhibitor with activity against ABL, BCRABL, KIT, and PDGFR.45,46 Preclinical data demonstrated activity against wild type and mutant forms of KIT,46,47 and more recently to some mutant forms of PDGFRα.48 One patient with metastatic GIST refractory to multiple types of therapies was treated with imatinib and provided proof of principle that inhibition of KIT by drug therapy was associated with improvement in disease.49 The phase I trial of imatinib in GIST patients tested 400 mg, 300 mg b.i.d., 400 mg b.i.d., and 500 mg b.i.d with the latter dose identified as dose limiting.50,51 The maximum tolerated dose was 400 mg twice daily, with dose limiting toxicities of nausea, vomiting, edema, and rash. Of note, in chronic myelogenous leukemia (CML), another disease
GASTROINTESTINAL STROMAL TUMORS
in which imatinib has significant clinical activity, there is a higher frequency of hematologic toxicities likely due to leukemic cell involvement of the bone marrow in CML.52,53 When bleeding is noted in GIST patients it is primarily due to responding tumors, typically bulky masses, rather than to thrombocytopenia. The factors impacting toxicity are low hemoglobin correlating with hematologic toxicity, and low albumin correlating with development of edema and fatigue. Higher dose therapy had greater toxicity with edema, fatigue, rash, and dyspnea seen with greater frequency.54 Intriguingly, patients on imatinib for a prolonged period have a decline in symptoms correlated with an improvement in imatinib clearance.55 In addition, patients requiring an increase in the therapeutic dose of imatinib, have fewer side effects than those whose therapy was initiated at a high dose.56 Enrollment and completion of phase II and III clinical trials testing imatinib were rapid due to early promising data and no other effective therapeutic options for patients with metastatic GIST being available. The US Finland phase II trial, designed and initiated prior to the completion of the phase I trial, tested 400 and 600 mg.57,58 The study included 147 patients yet was not powered to determine superiority of one dose level over the other. The EORTC performed a phase II trial using 400 mg b.i.d. in GIST and non-GIST soft tissue sarcomas.59 Lastly, the two large international phase III trials compared 400 mg daily and 400 mg b.i.d. since there was no standard therapeutic option with any activity to use as the comparator arm. The response rates in phase I and II trials of imatinib in GIST patients were 54–71% partial response (PR), with an additional 17–37% with SD, with only 1% of patients achieving a complete response (CR). Within days of initiating therapy, patients noted an improvement in symptoms as exemplified by a decrease in narcotic utilization.60 Although responses using standardized response criteria were not evident on CT scans at early time points, this rapid improvement in clinical symptoms correlated with the loss of metabolic activity seen by FDG PET scanning.61,62 Early indications of response by computer tomography can be detected by decreases in tumor nodule density.63 Objective responses by computer tomography scanning using standard size-based criteria can be seen to occur up to 1 year after starting imatinib. A majority of patients do benefit, with 79.5–91% of patients obtaining objective responses or prolonged SD. The median progression-free survival (PFS) is 84 weeks.64 Survival of patients with SD as their best response to treatment is equivalent to survival of patients who achieve partial or CRs.65 The phase III trials conducted in North America, Europe, and Australia evaluated toxicities, response rates, time to tumor progression and overall survival in patients with metastatic GIST treated with low dose, 400 mg daily, compared to high dose, 400 mg twice daily, imatinib. Both studies documented an increase in grade 3 and 4 toxicities in patients initially treated on the high dose arm.56,66 Toxicities were mitigated in patients who began on the low dose arm, and then had their dose escalated to the high dose arm with tumor progression. Neutropenia decreased in frequency, and only
423
fatigue and anemia increased in frequency compared to what was observed at the low dose.56 The multicenter phase III trials demonstrated a CR rate of 3–6%, a PR rate of 45–48%, and a SD rate of 26–32%, for a total clinical benefit of 76%.56,67 However, the two studies arrived at slightly different conclusions. The North American trial, S0033, was powered to determine if one dose was superior to the other in terms of overall survival (OS) and enrolled 746 patients.66 In contrast, the EORTC-led trial had as its primary endpoint, PFS, and enrolled 946 patients. The EORTC-led trial documented an advantage to the initiation of imatinib at 400 mg b.i.d. over 400 mg daily in terms of PFS, without any difference in OS.56 The North American trial found no statistical difference in the OS and PFS between 400 mg daily and 400 mg b.i.d.66,67 The reasons for the differences in conclusions are not clear. One explanation is that S0033 was underpowered, and if a larger number of patients had been studied, the same difference in PFS would have been seen as well. An alternative explanation is that differences in the rate of dose reductions and delays occurred between the two studies affecting the amount of the drug that patients actually received in each study. More importantly, as response to imatinib is correlated with mutation site in KIT and PDGFRα (see the following text), differences in the distribution of tumors with different mutation sites in the low and high dose cohorts could impact on the response rate and PFS seen. The phase III studies treated patients until there was evidence of disease progression, at which time point patients on low doses of imatinib were allowed to increase their dose up to 400 mg twice daily. Results of such an increase led to SD in 30% and PR in 2.5–6%.56,67 Grades 3, 4, and 5 toxicities were noted in 23, 7, and 2%, with gastrointestinal and hematologic toxicities being the most common.67 Interestingly, when compared to initial toxicities on the low dose arm, the rate of toxicity was equivalent, with the exception of fatigue and anemia which were increased and neutropenia which was decreased.56 This data suggests that there may be a development of tachyphalaxis to imatinib over time. Alternatively, as was suggested in an analysis of the phase I and II trials of imatinib done by the EORTC, there is approximately a 33% enhanced clearing of the drug when comparing samples drawn on day 1 and 29 to those drawn after 1 year of therapy.68 Although this did not meet statistical significance, the increase in clearance is intriguing as an explanation both for the restabilization of the disease as well as the apparent improvement in toxicity. The French Sarcoma Group designed a phase III trial to ask if imatinib can be stopped after 1 year in patients with surgically unresectable disease when the disease was stabilized on imatinib. The study was designed to assess the PFS with secondary endpoints to assess OS and response to the reinitiation of imatinib in patients who discontinued the drug.69,70 This strategy, if successful, would decrease toxicities and also potentially decrease the rate of resistance to imatinib. This study was the first to enroll patients who were not only KIT positive by immunohistochemistry, but also KIT negative GIST with evidence of PDGFRα mutations. Following 1 year of imatinib therapy, 58 patients
424
GASTROINTESTINAL TUMORS
free from tumor progression were randomized to interrupt imatinib or to continue therapy. All patients were evaluated every 3 months to determine disease status. Patients in the interruption arm were allowed to reinstitute imatinib at the time of disease progression. On the basis of the initial interim analysis done in May 2004, the rate of progression in the interruption arm was greater than six in the initial 14 patients randomized, a predetermined criteria for stopping further patient accrual and randomization. With a median follow-up time of 21 months following randomization, 66% of patients whose imatinib was interrupted experienced disease progression compared with only 15% of patients who continued on imatinib. PFS in the first group is 6 months but has not been reached in the continuous therapy group. Of the 14 patients in whom imatinib was reintroduced following disease progression, 11 of them have had their disease recontrolled (79%). There was no significant difference in 1-year survival between the treatment groups.
Determinants of Imatinib Therapeutic Response Imatinib is a competitive inhibitor of ATP binding to KIT4,71 – 75 and PDGFRα,30,76 which leads to inactivation of tyrosine kinase activity. Studies evaluating the structure of imatinib binding to KIT have determined important sites within the KIT protein structure for its activity.20,71 Wildtype KIT is present on the cellular membrane in an inactive form with the juxtamembrane portion of the molecule inserted into the kinase active site. Normally, in the presence of its ligand steel factor, KIT homodimerizes. This leads to a conformational change of the juxtamembrane domain and activation of the receptor. Imatinib binds to conserved sequences in the kinase domain, competing with ATP, thus inactivating the kinase activity of the molecule. Mutations in KIT lead to activation of the kinase in the absence of steel factor. Mutations in exon 11, in particular, disrupt the normal interactions between the juxtamembrane and kinase active site, and thus favor the active form of the kinase domain. However, these mutations do not affect the binding of imatinib, and thus the agent is effective in tumors with KIT exon 11 mutations. In vitro most KIT mutations appear to be sensitive to imatinib, with the exception of mutations in exon 17.77 – 79 Mutations in exon 12 but not exon 18 of PDGFRα are sensitive.77 The largest reported series of clinical samples correlating tumor mutations with response comes from the North American phase III trial of imatinib (Table 2).80 Of 344 samples analyzed, immunohistochemistry found that 94.2% were KIT positive and 2.3% were KIT negative GIST, with the remainder being non-GIST sarcomas. Tumors were screened for mutations in the sites known to commonly contain mutations: KIT exons 9, 11, 13, and 17 and PDGFRα exons 12 and 18. In KIT expressing tumors, 87% were found to have mutations, with 1% containing PDGFR mutations. In tumors without KIT protein expression, 50% were found to contain a KIT mutation, with an additional 38% with PDGFRα mutations. Tumors with KIT exon 11 mutations had the highest objective response rates (68%), while tumors with KIT exon 9 mutations or a wild-type KIT had a 40% response rate. In addition, PFS was also longer in the exon 11
group of tumors. There were too few patients with PGDFRα, KIT exon 13 or 17 mutations to analyze. The analysis did suggest that higher dose treatment maybe of benefit to patients whose tumors contained an exon 9 mutation; however, this was not statistically significant.
GIST Progression on Imatinib The initial studies of imatinib pointed to two patterns of resistance to imatinib in GIST patients. The first represents about 9–17% of patients who progress rapidly and have PD as their best response.51,56,57,66 Although it is conceivable that these patients have another sarcoma with KIT expression, most trials have had expert pathologic review as well as genotyping of these tumors. Thus, it is more likely that these patients are those with tumors with mutations of KIT or PDGFR that are less sensitive to imatinib. Resistance is also seen to develop in a second cohort of patients who have been maintained on imatinib for many months. These patients typically have had stable or responding disease for greater than 3 months.51,56,66,82 The median time to progressive disease is 18–24 months. Clinically, this second group contrasts with the first in that progression is more typically focal rather than involving all sites of known disease. Mechanisms hypothesized to be of importance are the loss of KIT inhibition as a consequence of increased drug efflux or other pharmacokinetic factors, KIT amplification/KIT deletion, or additional KIT or PGFRα mutations. Alternatively, KIT inhibition may still be present with a second genetic mutation or activating pathway as a cause of drug resistance. Table 2 Correlation between progression-free survival (PFS) and overall survival (OS) and site of KIT mutation with imatinib mesylate80 and sunitinib malate81 therapy.
Therapy Imatinib Mesylate
Sunitinib malate
NA = not available.
KIT mutation status KIT exon 11 (n = 211) KIT exon 9 (n = 25) Wild-type KIT (n = 33) Exon 9, original tumor (n = 9) Exon 9, following imatinib therapy (n = 13) Wild type, original tumor (n = 9) Wild type, following imatinib therapy (n = 8) Exon 11, original tumor (n = 42) Exon 11, following imatinib therapy (n = 7) Exon 11 + secondary mutations following Imatinib therapy (n = 25)
PFS
OS
19.2
NA
10.3 8.4
NA NA
14.3
29.2
31.8
NA
13.8
NA
13.8
NA
5.1
12.7
3.3
NA
5.1
NA
GASTROINTESTINAL STROMAL TUMORS
Evaluation of tumors from patients with recurrence has identified new mutations in KIT or PDGFRα in addition to the primary tumor mutation.83 – 87 The sites of secondary mutations include KIT exons 1, 13, 14, and 17 or PDGFRα exon 18. Microdissection of tumor metastases from patients following progression on imatinib have shown that different sites may contain different secondary mutations.84 This has raised the question of whether these areas represent outgrowth of preexisting clones, or develop as a consequence of imatinib therapy. These additional mutations appear to overcome the KIT/PDGFR inhibition induced by imatinib. Of interest is tumor stabilization/response seen in patients who progress on low dose imatinib and whose dose is increased. Approximately 32–36% of patients will have restabilization of their disease, with a minority of these patients having PRs.56,67 Explanations for the effectiveness of dose escalation are that it can overcome drug efflux mechanisms, increase therapeutic levels that over time have decreased with increased clearance of the agent, or the increased drug levels in a therapeutic range can inhibit the new KIT or PDGFR activation in areas with secondary mutations. To date, there are no reports of gene amplification of KIT or PGDFRα as an alternate reason for the efficacy of increasing drug dosage. Loss of KIT gene expression without a secondary PDGFR mutation is a rare event. Clearly understanding the biologic mechanisms of resistance is important, as these patients will increasingly provide therapeutic challenges.
Management of GIST with Imatinib Resistance Surgery should first be considered in patients with progressive disease. Patients should be carefully selected for surgery, as patients with diffuse disease progression are likely to experience significant surgical morbidity compared to those who only have isolated sites of progression.88 Alternative palliative approaches include chemoembolization or radiofrequency ablation of liver metastases. Increasing the dose of imatinib in patients who have progressed on 400 mg daily is also an appropriate option, as discussed above. In addition, there are anecdotal reports of doses of imatinib escalated above 400 mg b.i.d. without clear data on its benefits. Lastly, for patients who are not candidates for the above measures and who do not qualify for experimental approaches, continuation on imatinib at a dose that is well tolerated is of benefit in spite of progression.37 Clinical trials that have stopped imatinib therapy prior to the initiation of alternate therapies have demonstrated increases in clinical symptoms and tumor flare by PET scan.89,90 Thus, the use of imatinib until oral intake is no longer feasible is recommended in those patients without alternative options.
Sunitinib Malate (SU11248, Sutent) Sunitinib malate, a multitargeted tyrosine kinase inhibitor with activity against KIT, PDGFR, VEGFR, and FLT1/KDR, is expected to receive FDA approval for the management of patients who are imatinib refractory or intolerant. Phase I studies tested various doses and schedules including: 25, 50, or 75 mg orally once daily for 14 days, followed
425
by a 14-day rest period per cycle, 50 mg orally for 14 days with 7 days rest, and 50 mg orally for 28 days with 14 days rest.81,91 The latter schedule was chosen for the subsequent phase II and III trials in GIST. The most common grade 3 and 4 toxicities noted included fatigue, asymptomatic lipase and amylase increases, and hypertension. Other side effects included nausea, diarrhea, stomatitis, hand foot syndrome, anemia, and skin discoloration. Of interest, the stomatitis is described by patients as a burning sensation with eating and usually is not associated with visible oral lesions. Bleeding has also been described at sites of tumor biopsies when patients were on drug. In addition, patients with a history of coronary artery disease were found to have asymptomatic cardiac enzyme elevations. The phase I and II trials of sunitinib malate treated 97 patients, 96% of whom had progressed on a dose of 600 mg or higher of imatinib.81,91,92 These patients had extensive metastases. PET scan again documented decreased metabolic activity rapidly, with CT scan responses evolving more slowly.93 PET scans were obtained at baseline, day 7 of therapy, following the first period off of therapy, and at the end of the second treatment cycle with sunitinib malate. These studies demonstrated an increase in the glucose uptake during the period off of therapy that was statistically significant compared to the values obtained when patients were on therapy. This data supports the notion that continued anti-KIT or PDGFR targeted therapy is required to suppress disease activity. Response data in the phase I and II studies reveals a PR rate of 8%. An additional 33% of patients had SD lasting between 6 weeks and <6 months, as well as 35% patients who had SD for longer than six months. The median time to tumor progression was 7.8 months, with a median survival of 19.8 months. What is of particular interest is that responses and clinical benefit were observed in patients with mutations that were less sensitive to imatinib such as exon 9, wildtype KIT and PDGFRα, and those with acquired mutations identified with the development of resistance (Table 2).91 When analyzing time to progression, it was the longest for patients whose tumors contained a KIT exon 9 mutation, followed by wild type, exon 11, and worst for a tumor with both an exon 11 mutation and a new mutation (Table 2).81 Patients with KIT exon 9 and wild-type mutations had the best OS. The number of patients with PGDFRα mutations was too few to make statistical comparisons. The recently completed phase III trial of sutent was a double-blind placebo-controlled trial in patients with imatinib refractory GIST or patients who were intolerant to imatinib. Sunitinib malate therapy was available to patients on placebo at the time of disease progression. Three hundred and twelve patients were accrued over 12 months.94 Patients had previously been on imatinib for a median duration of 107 weeks at a median dose of 800 mg. At a planned interim analysis, it was found that the median time to progression was 6.3 months in the patients receiving sunitinib malate compared to 1.5 months in the patients receiving placebo (hazard ratio 0.335, p = 0.0001). The OS was also improved in the patients who initiated therapy on sunitinib malate, in spite of the cross over design of the study (hazard
426
GASTROINTESTINAL TUMORS
ration 0.491, p = 0.00674); median OS of patients initially randomized to sunitinib malate has not yet been reached. The response rate for patients receiving sunitinib malate confirmed the findings in the phase I/II trials: PR 8% and SD 58%. Fifty percent of the patients on the placebo group had SD, which was short lived, with only 1% lasting greater than 6 months compared with 19% of the patients receiving sunitinib malate. Fifty-nine patients crossed over from the placebo to the active agent. Of these, 10% had a PR.
Other Tyrosine Kinase Inhibitors AMG706 is a tyrosine kinase inhibitor with specificity against KIT and VEGFR. An ongoing phase II trial is testing its efficacy in patients who have progressed on imatinib, and the results will be available in 2006. BMS 354825 [N -(2-chloro-6-methylphenyl)-2-(6-(4-(2hydroxyethyl)piperazin-1-yl)-2-methylpyrimidin-4-ylamino) thiazole-5-carboxamide] is a SRC-family kinase inhibitor. It is being evaluated for treatment of CML refractory to imatinib because of the structural similarities between SRC and the mutated BCR-ABL protein in patients who have become refractory to imatinib.95,96 Preclinical testing in a mouse model of CML reveals that animals treated with BMS 354825 have a prolonged survival compared to untreated animals. A phase I trial of the agent has revealed responses in 31 of 36 patients with imatinib refractory CML or imatinib intolerance.96 An ongoing phase I trial is evaluating the efficacy in GIST patients and other solid tumors.
Combination Therapies GIST tumors are very vascular. They express VEGFR and patients with metastatic tumors have been shown to have elevated serum VEGF levels.97 At the baseline, patients with metastatic and unresectable disease have elevated levels of circulating KIT and VEGF levels compared to controls, but lower levels of the KIT ligand, steel factor.98 Following therapy with imatinib, KIT serum levels decreased, whereas levels of steel factor increased. Interestingly, VEGF levels decreased in patients responding to imatinib. Also, patients treated with imatinib have been shown to have a decrease in blood flow by perfusion MRI associated with decreased microvessel density and CD31 expression on biopsy.99 The underlying mechanism for these findings are not clear, but may relate to the effects on PDGFR, known to be involved in angiogenesis. In addition, some have hypothesized that part of the clinical activity of sunitinib malate is anti-VEGFR inhibition. On the basis of these data, a phase III trial will test the combination of imatinib with Bevacizumab, a fully humanized monoclonal antibody that binds VEGF, compared to single agent imatinib in patients with previously untreated metastatic disease. The pathways that KIT and PDGFR signal through have multiple downstream proteins that are additional therapeutic targets. For example, RAD001, an inhibitor of the mammalian target of rapamycin (mTOR), is being added to imatinib.89 mTOR is a member of the phosphatidylinositol kinase (PIK)-related kinase family in which a lipid kinase homology domain functions as a serine/threonine kinase to
regulate protein translation, cell cycle progression, and cellular proliferation. The initial results utilizing a weekly dose of RAD001 with daily imatinib demonstrated significant pharmacokinetic interactions between the two agents, with increases in the serum concentration of RAD001 when given concurrently with imatinib. When the schedule was modified to give RAD001 2.5 mg with imatinib 600 mg daily, 7 of 18 patients have derived clinical benefit, with two documented PR.100 PKC412, an oral staurosporine derivative that has activity against multiple kinases including protein kinase C isotypes a,b, and g, mutated and wild-type KIT, PDGFRα and β, VEGFR2, FGFR, and FLT3, is also being tested.90 Preclinical data have shown that PKC412 is active against activated KIT mutations, including those in exon 17 which are refractory to imatinib.85,101 Pharmacokinetic studies of the combination of PKC412 when added to imatinib revealed up to 70% decreases in serum concentrations of imatinib. In contrast, when imatinib was added to PKC412, serum levels of PKC412 increased as well as toxicity. To date, 3 of 17 evaluable patients have SD. Refinements in the dose and schedule of these drugs are being addressed to define the phase II dose, ensuring imatinib levels can be maintained at therapeutic levels when these agents are given together. A novel inhibitor of KIT, AMN107 is also being combined with imatinib. AMN107 is a synthetic second-generation inhibitor of the BCR-ABL tyrosine kinase that competes for the ATP-binding site of BCR-ABL.102,103 It has been shown to have in vitro activity in imatinib resistant cell lines. An upcoming trial will assess activity of this agent alone and in combination with imatinib in patients who have progressed on imatinib.
PROGNOSIS The prognosis of patients diagnosed with GIST is changing in the era of targeted therapy. Prior to the development of imatinib, patients’ primary treatment modality was surgery. Although patients with completely resected tumors had a median survival of 96 months, those with completely resected locally recurrent disease or metastatic disease had survivals of only 49 and 39 months respectively.7 Patients in whom complete resection was not feasible fared worse with survival rates of 26, 8, and 11 months following resection of primary, locally recurrent, and metastatic disease. Unfortunately, the role of other therapeutic modalities was very limited with no data to suggest an impact on survival. Imatinib has clearly increased OS in patients with metastatic and unresectable disease. Prior to imatinib, patients with metastatic disease lived approximately 12 months. With imatinib, survival has clearly been improved, with the median time to progression on therapy being 18–24 months. The role of adjuvant imatinib is being studied. It is not known if patients with high-risk and/or lower risk tumors will have an increased survival with adjuvant therapy. Therapy with imatinib has the potential to delay recurrent disease. Ongoing trials are determining if a benefit exists, and also asking what the optimal length of therapy in this setting is.
GASTROINTESTINAL STROMAL TUMORS
It is not clear if imatinib will have a role in increasing cure rates of surgically resected patients given the limited numbers of CRs noted in the metastatic disease setting. In addition, neoadjuvant imatinib with surgical resection is being studied to determine how to best cure and palliate patients with primary as well as recurrent disease. We are in an era that is very hopeful for GIST patients and their physicians. GIST is a disease that now has a very effective treatment for frontline metastatic disease. The challenge for physicians and patients now is the management of GIST that is refractory to imatinib therapy. A second agent, sunitinib malate, has shown efficacy in patients refractory to imatinib and will likely be approved for use in the early part of 2006. In addition, multiple new agents that target KIT/PDGFR and/or downstream mediators of the disease, as well as VEGFR inhibitors are being tested. Many of these will be tested first in the metastatic disease setting in tumors refractory to imatinib. However, the biologic rationale for many of these approaches warrants testing earlier in the disease process. In addition, unfolding preclinical data is adding to our biologic understanding, leading to the discovery of new therapeutic targets as well as to a better understanding of resistant disease.
14.
15.
16. 17.
18.
19.
20.
21.
22. 23.
REFERENCES 1. Mazur MT, Clark HB. Gastric stromal tumors: reappraisal of histogenesis. Am J Surg Pathol 1983; 7: 507 – 19. 2. Sarlomo-Rikala M, et al. CD117: a sensitive marker for gastrointestinal stromal tumors that is more specific than CD34. Mod Pathol 1998; 11(8): 728 – 34. 3. Miettinen M, Virolainen M, Maarit Sarlomo R. Gastrointestinal stromal tumors – value of CD34 antigen in their identification and separation from true leiomyomas and schwannomas. Am J Surg Pathol 1995; 19(2): 207 – 16. 4. Hirota S, et al. Gain-of-function mutations of c-kit in human gastrointestinal stromal tumors. Science 1998; 279: 577 – 80. 5. Crosby JA, et al. Malignant gastrointestinal stromal tumors of the small intestine: a review of 50 cases from a prospective database. Ann Surg Oncol 2001; 8(1): 50 – 9. 6. Pidhorecky I, et al. Gastrointestinal stromal tumors: current diagnosis, biologic behavior, and management. Ann Surg Oncol 2000; 7(9): 705 – 12. 7. DeMatteo RP, et al. Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 2000; 231(1): 51 – 8. 8. Miettinen M, et al. Gastrointestinal stromal tumors/smooth muscle tumors (GISTs) primary in the omentum and mesentery: clinicopathologic and immunohistochemical study of 26 cases. Am J Surg Pathol 1999; 23(9): 1109 – 18. 9. Miettinen M, et al. Esophageal stromal tumors: a clinicopathologic, immunohistochemical, and molecular genetic study of 17 cases and comparison with esophageal leiomyomas and leiomyosarcomas. Am J Surg Pathol 2000; 24(2): 211 – 22. 10. Takahashi Y, et al. Gastrointestinal stromal tumor of the duodenal ampulla: report of a case. Surg Today 2001; 31(8): 722 – 6. 11. Miettinen M, Sobin LH. Gastrointestinal stromal tumors in the appendix: a clinicopathologic and immunohistochemical study of four cases. Am J Surg Pathol 2001; 25(11): 1433 – 7. 12. Lasota J, Carlson JA, Miettinen M. Spindle cell tumor of urinary bladder serosa with phenotypic and genotypic features of gastrointestinal stromal tumor. Arch Pathol Lab Med 2000; 124(6): 894 – 7. 13. Tryggvason G, et al. Gastrointestinal stromal tumors in Iceland, 1990 – 2003: the Icelandic GIST study, a population-based incidence
24. 25.
26.
27.
28.
29.
30. 31.
32.
33.
34.
35. 36. 37.
427
and pathologic risk stratification study. Int J Cancer 2005; 117: 289 – 93 [Epub ahead of print]. Nilsson B, et al. Gastrointestinal stromal tumors: the incidence, prevalence, clinical course, and prognostication in the preimatinib mesylate era – a population-based study in western Sweden. Cancer 2005; 103(4): 821 – 9. Tran T, Davila J, El-Serag H. The epidemiology of malignant gastrointestinal stromal tumors: an analysis of 1,458 cases from 1992 to 2000. Am J Gastroenterol 2005; 100: 162 – 8. Nishida T, et al. Familial gastrointestinal stromal tumors with germline mutation of the KIT gene. Nat Genet 1998; 19: 323 – 4. Maeyama H, et al. Familial gastrointestinal stromal tumor with hyperpigmentation: association with a germline mutation of the c-kit gene. Gastroenterology 2001; 120(1): 210 – 5. Isozaki K, et al. Germline-activating mutation in the kinase domain of KIT gene in familial gastrointestinal stromal tumors. Am J Pathol 2000; 157(5): 1581 – 5. Hirota S, et al. Familial gastrointestinal stromal tumors associated with dysphagia and novel type germline mutation of KIT gene. Gastroenterology 2002; 122(5): 1493 – 9. Tarn C, et al. Analysis of KIT mutations in sporadic and familial gastrointestinal stromal tumors: therapeutic implications through protein modeling. Clin Cancer Res 2005; 11(10): 3668 – 77. Carney JA. Gastric stromal sarcoma, pulmonary chondroma, and extra-adrenal paraganglioma (Carney Triad): natural history, adrenocortical component, and possible familial occurrence. Mayo Clin Proc 1999; 74(6): 543 – 52. Coreless CL, Fletcher JA, Heinrich MC. Biology of gastrointestinal stromal tumors. J Clin Oncol 2004; 15: 3813 – 25. Yantiss RK, et al. Gastrointestinal stromal tumors: an ultrastructural study. Int J Surg Pathol 2002; 10(2): 101 – 13. Fletcher CD, et al. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Hum Pathol 2002; 33(5): 459 – 65. Miettinen M, Sobin LH, Sarlomo-Rikala M. Immunohistochemical spectrum of GISTs at different sites and their differential diagnosis with a reference to CD117 (KIT). Mod Pathol 2000; 13(10): 1134 – 42. West RB, et al. The novel marker, DOG1, is expressed ubiquitously in gastrointestinal stromal tumors irrespective of KIT or PDGFRA mutation status. Am J Pathol 2004; 165(1): 107 – 13. Blay P, et al. Protein kinase C theta is highly expressed in gastrointestinal stromal tumors but not in other mesenchymal neoplasias. Clin Cancer Res 2004; 10: 4089 – 95. Miettinen M, et al. Evaluation of malignancy and prognosis of gastrointestinal stromal tumors: a review. Hum Pathol 2002; 33(5): 478 – 83. Emory TS, et al. Prognosis of gastrointestinal smooth-muscle (stromal) tumors: dependence on anatomic site. Am J Surg Pathol 1999; 23(1): 82 – 7. Heinrich MC, et al. PDGFRA activating mutations in gastrointestinal stromal tumors. Science 2003; 299: 708 – 10. Antonescu CR, et al. Association of KIT exon 9 mutations with nongastric primary site and aggressive behavior: KIT mutation analysis and clinical correlates of 120 gastrointestinal stromal tumors. Clin Cancer Res 2003; 9(9): 3329 – 37. Prakash S, et al. Gastrointestinal stromal tumors in children and young adults: a clinicopathologic, molecular, and genomic study of 15 cases and review of the literature. J Pediatr Hematol Oncol 2005; 27(4): 179 – 87. Price VE, et al. Clinical and molecular characteristics of pediatric gastrointestinal stromal tumors (GISTs). Pediatr Blood Cancer 2005; 45(1): 20 – 4. Palazzo L, et al. Endosonographic features predictive of benign and malignant gastrointestinal stromal cell tumours. Gut 2000; 46(1): 88 – 92. Rothlin M, Schob O. Laparoscopic wedge resection for benign gastric tumors. Surg Endosc 2001; 15: 893 – 5. Otani Y, et al. Laparoscopic wedge resection of gastric submucosal tumors. Surg Laparosc Endosc Percutan Tech 2000; 10: 19 – 23. Demetri G, et al. Optimal management of patients with Gastrointestinal Stromal Tumor (GIST), expansion and update of NCCN clinical practice guidelines, 2004. J Natl Compr Cancer Netw 2004; 2(Suppl.): S1 – 26.
428
GASTROINTESTINAL TUMORS
38. Clary BM, et al. Gastrointestinal stromal tumors and leiomyosarcoma of the abdomen and retroperitoneum: a clinical comparison. Ann Surg Oncol 2001; 8(4): 290 – 9. 39. Pierie JP, et al. The effect of surgery and grade on outcome of gastrointestinal stromal tumors. Arch Surg 2001; 136(4): 383 – 9. 40. Rajan DK, et al. Sarcomas metastatic to the liver: response and survival after cisplatin, doxorubicin, mitomycin-C, Ethiodol, and polyvinyl alcohol chemoembolization. J Vasc Interv Radiol 2001; 12(2): 187 – 93. 41. Mavligit GM, et al. Gastrointestinal leiomyosarcoma metastatic to the liver. Durable tumor regression by hepatic chemoembolization infusion with cisplatin and vinblastine. Cancer 1995; 75(8): 2083 – 8. 42. Edmonson JH, et al. Contrast of response to dacarbazine, mitomycin, doxorubicin, and cisplatin (DMAP) plus GM-CSF between patients with advanced malignant gastrointestinal stromal tumors and patients with other advanced leiomyosarcomas. Cancer Invest 2002; 20: 605 – 12. 43. Dematteo RP, et al. Clinical management of gastrointestinal stromal tumors: before and after STI-571. Hum Pathol 2002; 33(5): 466 – 77. 44. Plaat BE, et al. Soft tissue leiomyosarcomas and malignant gastrointestinal stromal tumors: differences in clinical outcome and expression of multidrug resistance proteins. J Clin Oncol 2000; 18(18): 3211 – 20. 45. Buchdunger E, et al. Inhibition of the Abl protein-tyrosine kinase in vitro and in vivo by a 2-phenylaminopyrimidine derivative. Cancer Res 1996; 56(1): 100 – 4. 46. Buchdunger E, et al. Abl protein-tyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-kit and plateletderived growth factor receptors. J Pharmacol Exp Ther 2000; 295(1): 139 – 45. 47. Heinrich MC, et al. Inhibition of c-kit receptor tyrosine kinase activity by STI 571, a selective tyrosine kinase inhibitor. Blood 2000; 96(3): 925 – 32. 48. Corless CL, et al. PDGFRA mutations in gastrointestinal stromal tumors: frequency, spectrum and in vitro sensitivity to imatinib. J Clin Oncol 2005; 23: 5357 – 64. 49. Joensuu H, et al. Effect of the tyrosine kinase inhibitor STI571 in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med 2001; 344(14): 1052 – 6. 50. van Oosterom A, et al. Safety and efficacy of imatinib (STI571) in metastatic gastrointestinal stromal tumours: a phase I study. Lancet 2001; 358: 1421 – 3. 51. van Oosterom A, et al. 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 2002; 38(Suppl 5): S83 – 7. 52. Druker BJ, et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001; 344(14): 1031 – 7. 53. Druker BJ, et al. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N Engl J Med 2001; 344(14): 1038 – 42. 54. Van Glabbeke M, et al. Prognostic factors of toxicity and efficacy in patients with Gastro-Intestinal Stromal Tumors (GIST) treated with imatinib: a study of the EORTC-STBSG, ISG and AGITG. Chicago, Illinois: The American Society of Clinical Oncology, 2003. 55. Judson I et al., EORTC Soft Tissue and Bone Sarcoma Group. Imatinib pharmacokinetics in patients with gastrointestinal stromal tumour: a retrospective population pharmacokinetic study over time. Cancer Chemother Pharmacol 2005; 55(4): 379 – 86. 56. Verweij J, et al. Progression-free survival in gastrointestinal stromal tumours with high-dose imatinib: randomised trial. Lancet 2004; 364(9440): 1127 – 34. 57. Demetri G, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med 2002; 347: 472 – 80. 58. von Mehren M, et al. High incidence of durable responses induced by Imatinib mesylate (Gleevec in patients with unresectable and metastatic gastrointestinal stromal tumors (GISTs). Orlando, Florida: The American Society of Clinical Oncology, 2002: A1608.
59. Verweij J, et al. Imatinib Mesylate is an active agent for GIST but does not yield responses in other soft tissue sarcomas that are unselected for a molecular target. Eur J Cancer 2003; 39: 2006 – 11. 60. Blanke C, et al. Evaluation of the safety and efficacy of an oral molecularly-targeted therapy, STI571, in patients (pts) with unresectable or metastatic gastrointestinal Stromal Tumors (GISTS) expressing c-kit (CD117). Proceedings of the American Society of Clinical Oncology, 2001. San Francisco, California: American Society of Clinical Oncology, 2001: 1. 61. Van den Abbeele A, Badawi R. Use of positron emission tomography in oncology and its potential role to assess response to imatinib mesylate therapy in gastrointestinal stromal tumors (GISTs). Eur J Cancer 2002; 38(Suppl 5): S60 – 5. 62. Van den Abbeele A. F18-FDG-PET provides evidence of biological response to STI571 in patients with malignant gastrointestinal stromal tumors (GIST). In Proceedings of the American Society of Clinical Oncology, San Francisco, California, 2001: 362a. 63. Choi H, et al. Correlation of computerized tomography (CT) and proton emission tomography (PET) in patients with metastatic GIST treated at a single institution with imatinib mesylate. Chicago, Illinois: American Society of Clinical Oncology, 2003: A3290. 64. Blanke C, et al. Long-term follow up of advanced gastrointestinal stromal tumor (GIST) patients treated with imatinib mesylate. In 2004 Gastrointestinal Cancer Symposium, 2004, A2. 65. Demetri G, et al. Lack of progression is the most clinically significant measure of a patient’s clinical benefit: correlating the effects of imatinib mesylate therapy in gastrointestinal stromal tumor (GIST) with survival benefits. In Proceedings of the European Society of Medical Oncology, Vienna, Austria, 2004. 66. Benjamin R, et al. Phase III dose-randomized study of Imatinib mesylate (IM) for GIST: intergroup S0033 early results. Proc Am Soc Clin Oncol, 2003: 22: 814 (abstr 3271). 67. Rankin C, et al. Dose effect of imatinib (IM) in patients (pts) with metastatic GIST - Phase III Sarcoma Group Study S0033. In Annual Meeting of the American Society of Clinical Oncology, New Orleans, Louisiana, 2004: A9005. 68. Judson I et al., EORTC Soft Tissue and Bone Sarcoma Group. Imatinib pharmacokinetics in patients with gastrointestinal stromal tumour: a retrospective population pharmacokinetic study over time. Cancer Chemother Pharmacol 2005; 55(4): 376 – 86. 69. Blay J, et al. Continuous vs intermittent imatinib treatment in advanced GIST after one year: a prospective randomized phase III trial of the French Sarcoma Group. New Orleans, Louisiana: American Society of Clinical Oncology, 2004: A9006. 70. Le Cesne A, et al. Interruption of Imatinib (IM) in GIST patients with advanced disease: updated results of the prospective French Sarcoma Group randomized phase III trial on survival and quality of life. J Clin Oncol 2005; 23(16S): 823s. 71. Mol CD, et al. Structural basis for the autoinhibition and STI-571 inhibition of c-Kit tyrosine kinase. J Biol Chem 2004; 279: 31655 – 63. 72. Hirota S, et al. Effects of loss-of-function and gain-of-function mutations of c-kit on the gastrointestinal tract. J Gastroenterol 2000; 35(Suppl 12): 75 – 9. 73. Lasota J, et al. Mutations in exon 11 of c-Kit occur preferentially in malignant versus benign gastrointestinal stromal tumors and do not occur in leiomyomas or leiomyosarcomas. Am J Pathol 1999; 154(1): 53 – 60. 74. Lasota J, et al. Mutations in exons 9 and 13 of KIT gene are rare events in gastrointestinal stromal tumors. A study of 200 cases. Am J Pathol 2000; 157(4): 1091 – 5. 75. Rubin B, et al. Activation is a ubiquitous feature of gastrointestinal stromal tumors. Cancer Res 2001; 61: 8118 – 21. 76. Hirota S, et al. Gain-of-function mutations of platelet-derived growth factor receptor alpha gene in gastrointestinal stromal tumors. Gastroenterology 2003; 125: 660 – 7. 77. Heinrich M, et al. KIT mutational status predicts response to STI571 in patients with metastatic gastrointestinal tumors (GISTs). Orlando, Florida: American Society of Clinical Oncology, 2002: 6a. 78. Foster R, et al. Molecular basis of the constitutive activity and STI571 resistance of Asp816Val mutant KIT receptor tyrosine kinase. J Mol Graph Model 2004; 23(2): 139 – 52.
GASTROINTESTINAL STROMAL TUMORS 79. Ma Y, et al. The c-KIT mutation causing human mastocytosis is resistant to STI571 and other KIT kinase inhibitors; kinases with enzymatic site mutations show different inhibitor sensitivity profiles than wild-type kinases and those with regulatory-type mutations. Blood 2002; 99(5): 1741 – 4. 80. Heinrich M, et al. Correlation of clinical response to imatinib and target kinase genotype in patients with metastatic KIT+ GISTS (CALGB 150105/SWOG S0033). J Clin Oncol 2005; 23(16S): 3s. 81. Maki R, et al. SU11248 in patients with imatinib-resistant GIST: results from a continuation trial. J Clin Oncol 2005; 23(18S): 818s. 82. Demetri G, et al. Phase III dose-randomized study of imatinib mesylate (Gleevec, sti571) for GIST: intergroup S0033 early results. Orlando, Florida: American Society of Clinical Oncology, 2002: A1651. 83. Antonescu CR, et al. Acquired resistance to imatinib in gastrointestinal stromal tumor occurs through secondary gene mutation. Clin Cancer Res 2005; 11(11): 4182 – 90. 84. Wardelmann E, et al. Acquired resistance to imatinib in gastrointestinal stromal tumours caused by multiple KIT mutations. Lancet Oncol 2005; 6(4): 249 – 51. 85. Debiec-Rychter M, et al. Mechanisms of resistance to imatinib mesylate in gastrointestinal stromal tumors and activity of the PKC412 inhibitor against imatinib-resistant mutants. Gastroenterology 2005; 128(2): 270 – 9. 86. Chen LL, et al. A missense mutation in KIT kinase domain 1 correlates with imatinib resistance in gastrointestinal stromal tumors. Cancer Res 2004; 64(17): 5913 – 9. 87. Tamborini E, et al. A new mutation in the KIT ATP pocket causes acquired resistance to imatinib in a gastrointestinal stromal tumor patient. Gastroenterology 2004; 127(1): 294 – 9. 88. Hohenberger P, et al. Tumor resection following imatinib pretreatment in GI stromal tumors. Proc Am Soc Clin Oncol 2003; 22: 818. 89. Van Oosterom A, et al. Combination signal transduction inhibition: a phase I/II trial of the oral mTOR-inhibitor everolimus (E, RAD001) and imatinib mesylate (IM) in patients (pts) with gastrointestinal stromal tumor (GIST) refractory to IM. In ASCO Annual Meeting Proceedings, New Orleans, Louisiana, 2004: 3002. 90. Reichardt P, et al. A phase I/II trial of the oral PKC inhibitor PKC412 and imatinib mesylate in patients with gastrointestinal stromal tumors (GIST) refractory to imatinib (IM). Vienna, Austria: European Society of Medical Oncology, 2004. 91. Demetri GD, et al. SU11248, a multi-targeted tyrosine kinase inhibitor, can overcome imatinib (IM) resistance caused by diverse
92.
93.
94.
95. 96.
97.
98.
99. 100.
101.
102.
103.
429
genomic mechanisms in patients (pts) with metastatic gastrointestinal stromal tumor (GIST). New Orleans, Louisiana: American Society of Clinical Oncology, 2004: A3001. Demetri G, et al. Clinical activity and tolerability of the multi-targeted tyrosine kinase inhibitor SU11248 in patients (pts) with metastatic gastrointestinal stromal tumor (GIST) refractory to imatinib mesylate. Proc Am Soc Clin Oncol, 2003: 22: 814 (abstr 3273). Van den Abbeele AD, et al. Imaging kinase target inhibition with SU11248 by FDG-PET in patients (pts) with imatinib-resitant gastrointestinal stromal tumors (1-R GIST). J Clin Oncol 2005; 23(16S): 817s. Demetri G, et al. Phase 3, multicenter, randomized, double-blind, placebo-controlled trial of SU11248 in patients (pts) following failure of imatinib for metastatic GIST. J Clin Oncol 2005; 23(16S): 308s. Shah N, et al. Overriding imatinib resistance with a novel ABL kinase inhibitor. Science 2004; 305: 399 – 401. Sawyers C, et al. Hematologic and cytogenetic responses in imatinibresistant chronic phase chronic myelogenous leukemia patients treated with the Dual SRC/ABL kinase inhibitor BMS-354835: results from a phase I dose escalation study. San Diego, California: American Society of Hematology, 2004: A1. Takahashi R, et al. Expression of vascular endothelial growth factor and angiogenesis in gastrointestinal stromal tumor of the stomach. Oncology 2003; 64(3): 266 – 74. Bono P, et al. Serum KIT and KIT ligand levels in patients with gastrointestinal stromal tumors treated with imatinib. Blood 2004; 103(8): 2929 – 35. Trent J, et al. Apoptotic and anti-vascular activity of imatinib in GIST patients. J Clin Oncol 2005; 23(16S): 816s. van Oosterom A, et al. A phase I/II trial of the oral mTOR-inhibitor everolimus (E) and imatinib mesylate (IM) in patients (pts) with gastrointestinal stromal tumor (GIST) refractory to IM: study update. J Clin Oncol 2005; 23(16S): 824s. Growney JD, et al. Activation mutations of human c-KIT resistant to imatinib are sensitive to the tyrosine kinase inhibitor PKC412. Blood 2005; 106: 721 – 4. O’Hare T, et al. In vitro activity of Bcr-Abl inhibitors AMN107 and BMS-354825 against clinically relevant imatinib-resistant Abl kinase domain mutants. Cancer Res 2005; 65: 4500 – 5. Weisberg E, et al. Characterization of AMN107, a selective inhibitor of native and mutant Bcr-Abl. Cancer Cell 2005; 7: 129 – 41.
Section 6 : Gastrointestinal Tumors
38
Small Cell Carcinomas of the Gastrointestinal Tract Alexandria T. Phan and Paulo M. Hoff
INTRODUCTION Small cell carcinoma (SCC) is a highly malignant tumor composed of small round or egg-shaped cells with little cytoplasm. Clinically, these carcinomas are characterized by a high proliferative rate, early metastatic potential, high sensitivity to cytotoxic agents, and a poor long-term survival rate. The most common SCC site is the lung, which was initially described by Barnard in 1926. SCC of the lung is associated with smoking and comprises approximately 25% of all pulmonary cancers.1 In 1930, Duguid and Kennedy2 reported a case of extrapulmonary SCC. Since then, SCC has been reported to arise in almost every organ: skin, thymus, thyroid, nasal cavity, paranasal sinuses, larynx, hypopharynx, salivary glands, uterine cervix, endometrium, prostate, kidney, breast, bladder, and throughout the gastrointestinal (GI) tract.3 Extrapulmonary SCCs are rare, accounting for fewer than 1000 new cases yearly in the United States.4 In a large review by Fer and colleagues5 that included 400 patients diagnosed with SCC, only 20 (5%) of the lesions were believed to have originated in extrapulmonary organs, with most having originated in the lungs and later metastasizing to other organs. GI SCC belongs to a large family of neuroendocrine tumors that include a spectrum of tumors, from the indolent carcinoid and the islet cell tumors all the way to SCC. This chapter will focus on primary GI SCC, which are aggressive, large cell neuroendocrine tumors that behave like and are treated as SCC.
EPIDEMIOLOGY GI SCC is diagnosed infrequently, and McKeown6 is credited with the initial description of this entity. He reported two cases of esophageal SCC among 9000 autopsies in 1952. Since then, there have been reports of SCC originating throughout the GI tract, from the esophagus all the way to the rectum. A recent literature search by Brenner et al.7 identified 544 cases of GI SCC, representing 0.1% to 1% of
all GI malignancies diagnosed in the respective period. The most common GI SCC tumor site is the upper GI tract – esophagus and stomach (64%) – followed by the colon and rectum (20%) (see Table 1). Despite their diverse anatomic origins, these carcinomas look and behave similarly, usually presenting with metastasis at the time of diagnosis and progressing very rapidly. The patient’s median survival duration is usually less than 1 year, ranging from a few weeks to several months. Cure is usually only possible when the tumor is identified before the development of metastasis. Older reports of an occasional indolent course may have been associated with either inadequate diagnosis or confusion with other neuroendocrine tumors, such as classic carcinoid.
CLINICAL MANIFESTATIONS Patients with GI SCC have traditionally presented with locally advanced or metastatic disease, both of which are considered incurable. The recent increase in the use of screening endoscopy and colonoscopy have resulted in an increase in the diagnosis of SCC at earlier stages, but the impact of this change is still unclear. Constitutional symptoms are frequent, and other symptoms are site specific. GI bleeding and obstructive symptoms, such as dysphagia, nausea, vomiting, and jaundice, are common. Like SCC of the lung, paraneoplastic syndromes secondary to ectopic hormonal secretion, are not uncommon in GI SCC.4 Patients should be evaluated for paraneoplastic syndromes when this is clinically indicated. Ectopic production of vasoactive intestinal peptide,8 calcitonin,9 adrenocorticotropic hormone,10 and antidiuretic hormone11 have been described in the literature. Treatment should follow the guidelines used for paraneoplastic syndromes attributed to SCC of the lung.12
STAGING AND WORKUP Because these GI carcinomas are uncommon, the diagnosis should first exclude a primary lung SCC. After the histologic
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
SMALL CELL CARCINOMAS OF THE GASTROINTESTINAL TRACT Table 1 Demographic features and sites of organs of GI SCC (Adaptation with approval from Brenner et al. JCO 2004).
Clinical characteristics Age (years) Male : female Location in the GI tract Esophagus Stomach Colon Rectum Gall bladder Pancreas Ampulla of Vater Common bile duct Liver Small bowel Total
64 (mean) 6.5 : 1 Cases
(%)
290 60 70 40 46 21 6 3 7 1 544
53.3 11.0 13.0 7.3 8.4 3.9 1.1 0.5 1.3 0.2 –
GI, gastrointestinal.
diagnosis is established, a complete evaluation should include an x-ray and a computed tomography (CT) scan of the chest. A bronchoscopy, to exclude an occult primary lung tumor, particularly in heavy smokers, may be considered.4 The usual evaluation of the affected organ (e.g., colonoscopy or gastroesophagoscopy) must be performed in addition to the specific tests required for SCC – bone marrow aspiration and biopsy, CT scan of the brain, abdominal imaging with CT or magnetic resonance imaging (MRI), and others as clinically indicated. The role of recently available positron emission tomography in the workup of GI SCC is yet to be defined but it is likely to become important. No specific staging system has been proposed for GI SCC.4,13,14 Some authors advocate using the same staging system used for SCC of the lung, which is based on data from the Veterans Administration Lung Study Group. At the limited stage, the cancer is confined to a localized anatomic region with or without lymph node involvement. In the extensive stage, the cancer has disseminated beyond local–regional boundaries.4,14 However, most GI oncologists prefer the tumor-node-metastasis (TNM) system with staging specific to the affected organ.13 Although clinicians may use either system for staging GI SCC, there is no validation of either approach regarding its predictive or prognostic value. This lack of validation is unlikely to change soon, because the incidence of this disease is so rare.
431
could subsequently differentiate into either epithelial or neuroendocrine carcinoma or both.23 SCCs from different organs cannot be differentiated by their histologic features, which are indistinguishable from features of SCC of the lung.23 – 29 The World Health Organization (WHO) categorizes SCC of the lung into three groups: oat cell carcinomas, intermediate cell carcinoma, and combined carcinomas. Combined carcinomas are those in which another recognizable histologic feature is found along with the histologic characteristics of SCC.30 Typical oat cell–type tumors have small cells, two to three times the size of a mature lymphocyte; a high nucleus–cytoplasm ratio and minimal nuclear clearing; inconspicuous nucleoli; and dispersed chromatin. The intermediate-type tumors have relatively large cells with vesicular nuclei, prominent nucleoli, and comparatively abundant cytoplasm, and they are actually intermediate entities between the typical and the oat cell carcinoids.26,31 The typical small cell tumor has membrane-bound, electron-dense neurosecretory cytoplasmic granules that can be seen on electron microscopy. The number of these granules correlates with the tumor’s degree of differentiation.32 Consistent with an aggressive clinical course, SCC has a high proliferative rate, usually with necrosis and high mitotic index (>10 mitoses per high-power field).7 On Grimelius staining, the cells are argyrophilic and react with chromogranin. Immunohistochemistry is important in establishing the diagnosis; most tumors test positive for synaptophysin, chromogranin, Leu-7, and neuron-specific enolase (NSE).33 – 36 (see Figures 1 and 2). Some characteristics may not be present in all tumors, so an accurate diagnosis requires an experienced pathologist. Thyroid transcription factor-1, which is often negative in the case of GI SCC, may be a useful marker for differentiating between GI SCC and lung SCC.37
MOLECULAR CHARACTERISTICS The molecular events leading to the development of extrapulmonary SCC have not yet been elucidated. However,
HISTOLOGIC CHARACTERISTICS The histogenesis of extrapulmonary SCC is controversial. The original theories were that SCC arises from amine precursor uptake and decarboxylation (APUD) cells.3,15 – 17 Since 1968, when Pearse and colleagues18 proposed the acronym APUD to describe a group of cells with similar features and presumed embryological origin, considerable controversy, varying from neuroectodermal19 to neuroendodermal origins,20,21 has surrounded the embryogenesis of these cells. The observation that SCC is frequently found “combined” with squamous cells and adenocarcinomas22 led investigators to propose the origin as being a totipotent stem cell, which
Figure 1 H&E staining of poorly differentiated neuroendocrine carcinoma.
432
GASTROINTESTINAL TUMORS
durable responses is low. Specific recommendations for treatment of GI SCC for different organs are discussed in the following section.
CHARACTERISTICS OF AND TREATMENT RECOMMENDATIONS FOR SPECIFIC SITES OF GI SCC Esophageal SCC
Figure 2 Synaptophysin staining in poorly differentiated neuroendocrine carcinoma.
cytogenetic studies may be useful because they may help differentiate “true” extrapulmonary tumors from metastatic SCC that arise from an occult lung primary. Deletion of the short arm of chromosome 3 is commonly observed in lung primary tumors, but not in extrapulmonary SCC.38,39 Other molecular characteristics of GI SCC have been described, but whether or not they have a role in the pathogenesis/tumogenesis of this entity is unknown. Overexpression of p53,27,28,40 loss of Rb expression,27,41 Bcl-2 positivity,27 loss of p16,28,41 and mutation of k-ras28 have been reported as being associated with GI SCC.
GENERAL TREATMENT GUIDELINES Because of the relative infrequency of SCC of extrapulmonary origin, treatment recommendations are based on the therapies for SCC of the lung. Local treatment should be prescribed as part of the treatment regimen for patients who have localized disease or to control local symptoms. Because in most cases the disease has metastasized by the time of diagnosis, combination chemotherapy regimens are an essential part of the treatment plan.14,42 Published data for effective treatment of GI SCC are understandably limited. Regimens commonly used for SCC of the lung, such as etoposide plus cisplatin,43 – 50 have produced good responses.51 Other common combinations include cyclophosphamide, doxorubicin, and vincristine (CAV) and cyclophosphamide, doxorubicin, vincristine, and etoposide (CAVE).52 Recently, researchers from the Minnie Pearl Cancer Research Network presented preliminary data from a phase II study of paclitaxel, carboplatin, and etoposide (PCE) for extrapulmonary SCC, and indicated tolerable toxicity and good activity.53 The combination of irinotecan and cisplatin seems to be very active as well.54 Newer agents and regimens55,56 are being evaluated in patients with SCC of the lung, but their role in extrapulmonary SCC is currently undefined. Unfortunately, although responses to chemotherapy are high, the rate of
The esophagus is the most common site of extrapulmonary SCC of the GI tract. The reported incidence of SCC in relation to the total number of esophageal cancers varies widely, ranging from as low as 0.05% among 1918 patients with esophageal cancer57 to as high as 15% in some studies.58 Recent reports by investigators in the United States cite an incidence of approximately 1–2% of all cases of esophageal cancer.4,59 Because of the limited number of cases of esophageal SCC, no definite risk factors can be described. A possible association with tobacco use has been suggested, but esophageal SCC has also been described among nonsmokers.59 This tumor has an anatomic preference for the middle and lower thirds of the esophagus,3,59 with a slightly higher incidence in men than in women (3 : 2), and its onset usually occurs after the fifth decade of life.11,59 Common presenting symptoms include dysphagia, weight loss, chest pain, and abdominal pain.13,60,61 If untreated, the prognosis is poor, with median survival duration measured in weeks and no long-term survivals.62 Patients who are treated aggressively have a median survival duration of approximately 7 months, and patients occasionally survive for over 2 years.59,63 Treatment with surgery and radiation alone was used in the past, but the results were disappointing. The recent use of aggressive combination chemotherapy regimens developed for SCC of the lung has resulted in markedly improved but usually transient responses.42,64 A multimodality approach to treatment seems to be important. Craig and colleagues65 combined subtotal esophagectomy with chemotherapy or radiotherapy for seven patients with localized disease, and reported a mean survival duration of 20 months. One patient treated with surgery and chemotherapy survived for more than 96 months. Data from several single-institution reports suggest that the optimal treatment for patients with localized disease is a multimodality regimen consisting of chemotherapy designed for lung SCC, followed by local–regional consolidation, and either surgical resection or radiotherapy.62,66,67 It has been suggested that surgical resection improves long-term survival for limited-stage disease.68 Patients with advanced disease should be treated with chemotherapy alone; if they respond, local–regional therapy should be considered. Patients who have a poor performance status and cannot be treated aggressively or for whom treatment has failed may benefit from close follow-up and the palliative measures commonly used for patients with esophageal SCCs and adenocarcinomas, including supportive care, radiotherapy, laser therapy, and stenting.
Gastric SCC SCC of the stomach occurs infrequently. The first welldocumented case was reported in 1976,69 and since then,
SMALL CELL CARCINOMAS OF THE GASTROINTESTINAL TRACT
fewer than 50 cases have been cited in the literature.9 It is not surprising that the majority of cases, including the largest series,70 were reported by Japanese investigators, and this suggests that similar risk factors maybe involved in gastric SCC and adenocarcinoma. A SCC incidence of less than 1% has been reported in patients with gastric cancer,70 but the incidence may be as low as 0.01%.69 Most gastric SCCs are intermediatetype tumors and are combined with other histologies with no predilection for location in the stomach. Men are at a twofold increased risk for gastric SCC compared to women, and the most common presenting symptoms are epigastralgia, nausea, melena, anorexia, and weight loss. At diagnosis, the majority of patients have advanced disease. The liver is the most common site of distant metastasis but involvement of the brain, lungs, bones, lymph nodes, and bone marrow have been reported.70 The median survival duration is 9 months,9 and few studies have described long-term survivals.3,69 Combination chemotherapy may prolong survival, but too few cases are available for a definitive recommendation. Of interest, in 1950 Bernatz71 reported a 5-year-survival rate of close to 70% in 72 patients with SCC of the stomach who were treated with surgical resection. However, most of these cancers had a mixed histologic pattern, and, more importantly, patients with the diagnosis of lymphoma rather than SCC may have been included in the series.4,71,72
SCC of the Colon and Rectum SCC of the colon and rectum are aggressive carcinomas that appear and behave similar to their counterparts in the lung.24 Clery and associates73 first described SCC of the colon and rectum in 1961 and suggested that the incidence of this entity represented 0.2% of all colon cancers. Since then, fewer than 100 cases of SCC of the colon and rectum have been reported. Slightly over a third of the cases occurred in the rectum, followed by the cecum, sigmoid, transverse colon, and ascending colon.32 The clinical presentation of this cancer includes hematochezia, weight loss, fatigue, anorexia, abdominal pain, and changes in bowel habits,25,74 but the symptoms are usually similar to those of adenocarcinoma at a similar anatomic location. No preference among the sexes has been identified, and most of these carcinomas are diagnosed in patients 60 to 70 years of age. No risk factors have been firmly identified, but a case report suggests that patients who have ulcerative colitis might have a higher risk.32 In about 15% of patients, the disease is confined to the organ or to the regional nodes, but 85% of patients present with distant metastasis. The liver is the organ most commonly affected by metastasis; other organs include the lymph nodes, bones, bone marrow, and skin.75 – 77 Although SCCs frequently show coexistent squamous and glandular differentiation, the metastases observed almost always have small cell histologic characteristics.26 Despite earlier reports of long-term survivors among patients with resectable colorectal carcinomas,73 the median survival duration for patients with SCC of the colorectal area remains less than 1 year.25,76,77 Use of multiagent
433
chemotherapy, including drugs that are effective against bronchogenic SCC, has been recommended and documented in case reports, but because of the rarity of this tumor, no large clinical trials have been conducted.25,26,76 Patients with localized disease should be treated with a combined modality approach, including local treatment with surgery or radiotherapy and chemotherapy. For patients with advanced disease, the treatment should focus on chemotherapy.14,26,76
SCC of the Pancreas SCC of the pancreas was originally described in 1951,78 and a large retrospective series showed that 1–1.5% of all pancreatic cancers consistently have some small cell histologic features.3,79 Only 19 cases of pure pancreatic small cell cancer have been reported in the literature.80 The disease has no known risk factors, and its epidemiology is not well defined. Common presenting symptoms include jaundice and abdominal discomfort. Of the 15 patients who did not receive chemotherapy, the median survival duration was 2 months.76,81 Of the four patients treated with chemotherapy, one patient achieved a partial response and three obtained complete responses, although two of the three patients with complete responses relapsed and died within a year. The third patient, who had widely metastatic disease at diagnosis, was alive and free of disease 9 years after treatment with cisplatin and etoposide.5,51,80 On the basis of limited information, the optimal treatment for SCC of the pancreas includes chemotherapy with the same agents used in the treatment of lung SCC, with radiotherapy as consolidation.
SCC at Other GI Sites SCCs have also been documented in the ampulla of Vater and the small intestines. These tumors are so rare that only a few cases have been reported. Toker81 cited the only welldocumented case of SCC of the jejunum, which occurred in a 17-year-old girl who, when treated with surgery, had a prolonged disease-free survival. Six cases of SCC of the ampulla of Vater have been described in the literature since the first report by Swanson et al. in 1986.33,82 The patients, five men and one woman, tended to be of older age. The median survival duration was 10 months, with no long-term survivals.33,83 The usual presentation of this tumor includes signs of biliary obstruction and rapid development of metastases. Patients with good performance status should be treated with chemotherapy, similar to those with SCC of the lung, but no definitive data are available regarding the effectiveness of chemotherapy for this setting.
CONCLUSION GI SCC consists of a rare group of malignancies characterized by aggressive clinical history and chemotherapy sensitivity, with very few reports of durable remission or cure. SCC has been reported in almost every organ of the GI tract, with the esophagus being the most frequently reported site. Most of the information about the treatment is derived from what we know about SCC of the lung. Treatment of extensive
434
GASTROINTESTINAL TUMORS
or advanced stage disease is based on platinum-containing regimens. Optimal treatment of limited-stage disease includes the use of a multimodality approach combining chemotherapy, radiation therapy, and possibly surgery.
REFERENCES 1. Livingston RB. Small cell carcinoma of the lung. Blood 1980; 56: 575 – 84. 2. Duguid J, Kennedy A. Oat-cell tumors of mediastinal glands. J Pathol Bacteriol 1930; 33: 93 – 9. 3. Ibrahim NB, Briggs JC, Corbishley CM. Extrapulmonary oat cell carcinoma. Cancer 1984; 54: 1645 – 61. 4. Remick SC, Hafez GR, Carbone PP. Extrapulmonary small-cell carcinoma. A review of the literature with emphasis on therapy and outcome. Medicine (Baltimore) 1987; 66: 457 – 71. 5. Fer MF, Greco FA, Oldham RK. Poorly differentiated neoplasms and tumors of unknown origin: introduction. Semin Oncol 1982; 9: 393 – 5. 6. McKeown F. Oat-cell carcinoma of the oesophagus. J Pathol Bacteriol 1952; 64: 889 – 91. 7. Brenner B, et al. Small-cell carcinomas of the gastrointestinal tract: a review. J Clin Oncol 2004; 22: 2730 – 9. 8. Watson KJ, et al. Watery diarrhea-hypokalemia-achlorhydria syndrome and carcinoma of the esophagus. Gastroenterology 1985; 88: 798 – 803. 9. O’Byrne KJ, et al. Extrapulmonary small cell gastric carcinoma. A case report and review of the literature. Acta Oncol 1997; 36: 78 – 80. 10. Tateishi R, et al. Argyrophil cell carcinoma (apudoma) of the esophagus. A histopathologic entity. Virchows Arch A Pathol Anat Histol 1976; 371: 283 – 94. 11. Doherty MA, McIntyre M, Arnott SJ. Oat cell carcinoma of esophagus: a report of six British patients with a review of the literature. Int J Radiat Oncol Biol Phys 1984; 10: 147 – 52. 12. Thomas L, Kwok Y, Edelman MJ. Management of paraneoplastic syndromes in lung cancer. Curr Treat Options Oncol 2004; 5: 51 – 62. 13. Vrouvas J, Ash DV. Extrapulmonary small cell cancer. Clin Oncol (R Coll Radiol) 1995; 7: 377 – 81. 14. Van Der Gaast A, et al. Chemotherapy as treatment of choice in extrapulmonary undifferentiated small cell carcinomas. Cancer 1990; 65: 422 – 4. 15. Bensch KG, et al. Oat-cell carcinoma of the lung. Its origin and relationship to bronchial carcinoid. Cancer 1968; 22: 1163 – 72. 16. Gould VE, et al. The APUD cell system and its neoplasms: observations on the significance and limitations of the concept. Surg Clin North Am 1979; 59: 93 – 108. 17. Skrabanek P, Powell D. Unifying concept of non-pituitary ACTHsecreting tumors: evidence of common origin of neural-crest tumors, carcinoids, and oat-cell carcinomas. Cancer 1978; 42: 1263 – 9. 18. Pearse AG. Common cytochemical and ultrastructural characteristics of cells producing polypeptide hormones (the APUD series) and their relevance to thyroid and ultimobranchial C cells and calcitonin. Proc R Soc Lond B Biol Sci 1968; 170: 71 – 80. 19. Pearse AG, Takor TT. Neuroendocrine embryology and the APUD concept. Clin Endocrinol (Oxf) 1976; 5(Suppl): 229S – 244S. 20. Sidhu GS. The endodermal origin of digestive and respiratory tract APUD cells. Histopathologic evidence and a review of the literature. Am J Pathol 1979; 96: 5 – 20. 21. Bensch KG. The problem of classifying peripheral endocrine tumors. Hum Pathol 1983; 14: 383 – 5. 22. Arai K, Matsuda M. Gastric small-cell carcinoma in Japan: a case report and review of the literature. Am J Clin Oncol 1998; 21: 458 – 61. 23. Ho KJ, et al. Small cell carcinoma of the esophagus: evidence for a unified histogenesis. Hum Pathol 1984; 15: 460 – 8. 24. Jass J, Sobin LH. Histological typing of intestinal tumors. WHO International Histological Classification of Tumors. Berlin, Germany: Springer-Verlag, 1989. 25. Wick MR, Weatherby RP, Weiland LH. Small cell neuroendocrine carcinoma of the colon and rectum: clinical, histologic, and ultrastructural study and immunohistochemical comparison with cloacogenic carcinoma. Hum Pathol 1987; 18: 9 – 21.
26. Burke AB, Shekitka KM, Sobin LH. Small cell carcinomas of the large intestine. Am J Clin Pathol 1991; 95: 315 – 21. 27. Takubo K, et al. Primary undifferentiated small cell carcinoma of the esophagus. Hum Pathol 1999; 30: 216 – 21. 28. Maitra A, et al. Small cell carcinoma of the gallbladder: a clinicopathologic, immunohistochemical, and molecular pathology study of 12 cases. Am J Surg Pathol 2001; 25: 595 – 601. 29. Sarsfield P, Anthony PP. Small cell undifferentiated (‘neuroendocrine’) carcinoma of the colon. Histopathology 1990; 16: 357 – 63. 30. WHO. The World Health Organization histological typing of lung tumours. Second edition. Am J Clin Pathol 1982; 77: 123 – 36. 31. Gaffey MJ, Mills SE, Lack EE. Neuroendocrine carcinoma of the colon and rectum. A clinicopathologic, ultrastructural, and immunohistochemical study of 24 cases. Am J Surg Pathol 1990; 14: 1010 – 23. 32. Yaziji H, Broghamer WL Jr. Primary small cell undifferentiated carcinoma of the rectum associated with ulcerative colitis. South Med J 1996; 89: 921 – 4. 33. Sarker AB, et al. An immunohistochemical and ultrastructural study of case of small-cell neuroendocrine carcinoma in the ampullary region of the duodenum. Acta Pathol Jpn 1992; 42: 529 – 35. 34. Bunn PA Jr, et al. Small cell lung cancer, endocrine cells of the fetal bronchus, and other neuroendocrine cells express the Leu-7 antigenic determinant present on natural killer cells. Blood 1985; 65: 764 – 8. 35. Lloyd RV, et al. Distribution of chromogranin A and secretogranin I (chromogranin B) in neuroendocrine cells and tumors. Am J Pathol 1988; 130: 296 – 304. 36. Tapia FJ, et al. Neuron-specific enolase is produced by neuroendocrine tumours. Lancet 1981; 1: 808 – 11. 37. Ordonez NG. Value of thyroid transcription factor-1 immunostaining in distinguishing small cell lung carcinomas from other small cell carcinomas. Am J Surg Pathol 2000; 24: 1217 – 23. 38. Ledermann JA. Extrapulmonary small cell carcinoma. Postgrad Med J 1992; 68: 79 – 81. 39. Carney D, McCann A, Corbally N. Molecular Genetics of Lung Cancer. Chichester, England: John Wiley and Sons, 1990. 40. Lam KY, et al. Esophageal small cell carcinomas: clinicopathologic parameters, p53 overexpression, proliferation marker, and their impact on pathogenesis. Arch Pathol Lab Med 2000; 124: 228 – 33. 41. Parwani AV, et al. Immunohistochemical and genetic analysis of nonsmall cell and small cell gallbladder carcinoma and their precursor lesions. Mod Pathol 2003; 16: 299 – 308. 42. Kelsen DP, et al. Small-cell carcinoma of the esophagus: treatment by chemotherapy alone. Cancer 1980; 45: 1558 – 61. 43. Einhorn LH. Initial therapy with cisplatin plus VP-16 in small-cell lung cancer. Semin Oncol 1986; 13: 5 – 9. 44. Batist G, et al. Etoposide (VP-16) and cisplatin in previously treated small-cell lung cancer: clinical trial and in vitro correlates. J Clin Oncol 1986; 4: 982 – 6. 45. Feld R, et al. Canadian multicenter randomized trial comparing sequential and alternating administration of two non-cross-resistant chemotherapy combinations in patients with limited small-cell carcinoma of the lung. J Clin Oncol 1987; 5: 1401 – 9. 46. Wolf M, et al. Cisplatin/etoposide versus ifosfamide/etoposide combination chemotherapy in small-cell lung cancer: a multicenter German randomized trial. J Clin Oncol 1987; 5: 1880 – 9. 47. Bunn PA Jr, Carney DN. Overview of chemotherapy for small cell lung cancer. Semin Oncol 1997; 24: S7-69 – S7-74. 48. Figoli F, et al. Cisplatin and etoposide as second-line chemotherapy in patients with small cell lung cancer. Cancer Invest 1988; 6: 1 – 5. 49. Mascaux C, et al. A systematic review of the role of etoposide and cisplatin in the chemotherapy of small cell lung cancer with methodology assessment and meta-analysis. Lung Cancer 2000; 30: 23 – 36. 50. Maksymiuk AW, et al. Sequencing and schedule effects of cisplatin plus etoposide in small-cell lung cancer: results of a North Central Cancer Treatment Group randomized clinical trial. J Clin Oncol 1994; 12: 70 – 6. 51. Morant R, Bruckner HW. Complete remission of refractory small cell carcinoma of the pancreas with cisplatin and etoposide. Cancer 1989; 64: 2007 – 9.
SMALL CELL CARCINOMAS OF THE GASTROINTESTINAL TRACT 52. Jackson DV Jr, et al. Improvement of long-term survival in extensive small-cell lung cancer. J Clin Oncol 1988; 6: 1161 – 9. 53. Miranda FT, et al.: Paclitaxel/carboplatin/etoposide (PCE) therapy for advanced poorly differentiated neuroendocrine (PDNE) carcinoma: a Minnie Pearl Cancer Research Network phase II trial. In Proceedings of Clinical Oncology, Abstract 4058, 2005. 54. Hou Z, et al. A pilot study of irinotecan plus cisplatin in patients with metastatic high-grade neuroendocrine carcinoma. Proc Am Soc Clin Oncol 2003; 22: 375. 55. Ardizzoni A et al., The European Organization for Research and Treatment of Cancer Early Clinical Studies Group and New Drug Development Office, and the Lung Cancer Cooperative Group. Topotecan, a new active drug in the second-line treatment of small-cell lung cancer: a phase II study in patients with refractory and sensitive disease. J Clin Oncol 1997; 15: 2090 – 6. 56. Sculier JP, et al. Randomized trial comparing induction chemotherapy versus induction chemotherapy followed by maintenance chemotherapy in small-cell lung cancer. European Lung Cancer Working Party. J Clin Oncol 1996; 14: 2337 – 44. 57. Turnbull AD, et al. Primary malignant tumors of the esophagus other than typical epidermoid carcinoma. Ann Thorac Surg 1973; 15: 463 – 73. 58. Horai T, et al. A cytologic study on small cell carcinoma of the esophagus. Cancer 1978; 41: 1890 – 6. 59. Huncharek M, Muscat J. Small cell carcinoma of the esophagus. The Massachusetts General Hospital experience, 1978 to 1993. Chest 1995; 107: 179 – 81. 60. Sabanathan S, Graham GP, Salama FD. Primary oat cell carcinoma of the oesophagus. Thorax 1986; 41: 318 – 21. 61. McFadden DW, Rudnicki M, Talamini MA. Primary small cell carcinoma of the esophagus. Ann Thorac Surg 1989; 47: 477 – 80. 62. Ohmura Y, et al. Small cell carcinoma of the esophagus: a case report. Jpn J Clin Oncol 1997; 27: 95 – 100. 63. Nichols GL, Kelsen DP. Small cell carcinoma of the esophagus. The Memorial Hospital experience 1970 to 1987. Cancer 1989; 64: 1531 – 3. 64. Rosenthal SN, Lemkin JA. Multiple small cell carcinomas of the esophagus. Cancer 1983; 51: 1944 – 6. 65. Craig SR, et al. Primary small-cell cancer of the esophagus. J Thorac Cardiovasc Surg 1995; 109: 284 – 8. 66. Medgyesy CD, et al. Small cell carcinoma of the esophagus: the University of Texas M. D. Anderson Cancer Center experience and literature review. Cancer 2000; 88: 262 – 7. 67. Madroszyk A, et al. Small-cell carcinoma of the esophagus: report of three cases and review of the literature with emphasis on therapy. Ann Oncol 2001; 12: 1321 – 5.
435
68. Yachida S, et al. Long-term survival after resection for small cell carcinoma of the esophagus. Ann Thorac Surg 2001; 72: 596 – 7. 69. Matsusaka T, Watanabe H, Enjoji M. Oat-cell carcinoma of the stomach. Fukuoka Igaku Zasshi 1976; 67: 65 – 73. 70. Matsui K, et al. Small cell carcinoma of the stomach: a clinicopathologic study of 17 cases. Am J Gastroenterol 1991; 86: 1167 – 75. 71. Bernatz P. Small Cell Neoplasm of the stomach: A Clinicopathologic Study, Graduate School. Minnesota: University of Minnesota, 1950. 72. Richardson RL, Weiland LH. Undifferentiated small cell carcinomas in extrapulmonary sites. Semin Oncol 1982; 9: 484 – 96. 73. Clery AP, Dockerty MB, Waugh JM. Small-cell carcinoma of the colon and rectum. A clinicopathologic study. Arch Surg 1961; 83: 164 – 72. 74. Hussein AM, et al. Small cell carcinoma of the large intestine presenting as central nervous systems signs and symptoms. Two case reports with literature review. J Neurooncol 1990; 8: 269 – 74. 75. Pazdur R, et al. Neuroendocrine small cell carcinomas in miscellaneous primary sites: implications for staging and therapy. Anticancer Res 1981; 1: 335 – 40. 76. Redman BG, Pazdur R. Colonic small cell undifferentiated carcinoma: a distinct pathological diagnosis with therapeutic implications. Am J Gastroenterol 1987; 82: 382 – 5. 77. Mills SE, Allen MS, Cohen AR Jr. Small-cell undifferentiated carcinoma of the colon. A clinicopathological study of five cases and their association with colonic adenomas. Am J Surg Pathol 1983; 7: 643 – 51. 78. Miller JR, Baggenstoss AH, Comfort MW. Carcinoma of the pancreas; effect of histological types and grade of malignancy on its behavior. Cancer 1951; 4: 233 – 41. 79. Cubilla AL, Fitzgerald PJ. Cancer of the pancreas (nonendocrine): a suggested morphologic classification. Semin Oncol 1979; 6: 285 – 97. 80. Wahid NA, et al. Response of small cell carcinoma of pancreas to a small cell lung cancer regimen: a case report. Cancer Invest 1996; 14: 335 – 9. 81. Toker C. Oat cell tumor of the small bowel. Am J Gastroenterol 1974; 61: 481 – 3. 82. Swanson PE, et al. Primary duodenal small-cell neuroendocrine carcinoma with production of vasoactive intestinal polypeptide. Arch Pathol Lab Med 1986; 110: 317 – 20. 83. Lee CS, Machet D, Rode J. Small cell carcinoma of the ampulla of vater. Cancer 1992; 70: 1502 – 4.
Section 7 : Gynecological Cancers
39
Extra-ovarian Primary Peritoneal Carcinomas Alberto E. Selman and Larry J. Copeland
INTRODUCTION Extraovarian primary peritoneal carcinoma (EOPPC) is found in 7–21% of patients with suspected epithelial ovarian carcinoma (EOC).1 – 5 Basically, EOPPC is an extraovarian adenocarcinoma involving the peritoneum and spreads throughout the peritoneal cavity with minimal or no ovarian involvement, and no obvious primary site. Most of EOPPC cases reported have been of serous histology, and in many cases, the malignancy developed following bilateral oophorectomy for reasons other than cancer.3 Histopathological, immunohistochemical (IHC), and clinical similarities have been observed between EOPPC and EOC; specifically the serous variety, but EOPPC involves the ovarian surface only minimally (microscopic disease) or spares the ovaries entirely.6 On the other hand, molecular and epidemiological studies have reported some differences between the two diseases.7,8 Disagreement and controversy continue to exist in the literature among authorities who consider EOPPC a different clinical entity from EOC.1,9 – 11 Most investigators consider these entities to be similar in clinical presentation and course, except that EOPPC patients by disease definition present at a more advanced stage.
HISTORICAL BACKGROUND Numerous terms have been used to describe this entity, including serous surface carcinoma of the peritoneum, papillary tumor of the peritoneum, serous surface papillary carcinoma, serous papillary surface carcinoma of the ovary, serous borderline tumor of the peritoneum, extraovarian pelvic serous tumor, multifocal extraovarian serous carcinoma, extraovarian papillary serous carcinoma, mesothelioma, and papillary serous carcinoma of the peritoneum (PSCP). Swerdlow, in a 1959 case report, first described malignant mesothelioma (MM) with diffuse involvement of the peritoneum, no obvious primary site, and grossly normal ovaries in a 27-year-old woman.12 Since then, many authors have
reported similar cases.1,3,13 – 15 In 1977, Kannerstein pointed out the importance of distinguishing EOPPC from MM.3 Since its first description by Swerdlow, EOPPC has been recognized in females; its histologic similarities to MM and papillary serous ovarian carcinoma (PSOC) have caused difficulties in classifying it as an independent clinicopathologic entity.3 It is recognized as a distinct clinicopathologic tumor that arises from mesothelial cells under M¨ullerian influences.16,17 Some authors have regarded the term mesothelioma and EOPPC as synonymous.18,19 Parmley et al. have also proposed that all tumors arising from the pleural and peritoneal cavities be classified as mesotheliomas.20 Foyle et al. described 25 peritoneal tumors in women and divided them histologically into five groups: mesothelial hyperplasia, well-differentiated diffuse papillary mesothelioma, diffuse papillary mesothelioma, atypical diffuse mesothelioma, and papillary carcinoma.13 Only three tumors closely resemble papillary or tubopapillary diffuse MM, the type associated with asbestos, which occurs in the pleural cavities in both sexes. There were eight welldifferentiated mesotheliomas and these were associated with indolent behavior. However, in 10 cases, tumor resembled PSOC; these tumors were different and progressed rapidly. They concluded that EOPPC should not be merged with the general group of diffuse mesotheliomas. Other authors have also opposed the grouping of EOPPC and mesothelioma of the peritoneum because of important epidemiologic, histologic, ultrastructural, and biological differences.1,3 Researches have demonstrated the link between peritoneal mesotheliomas and asbestos exposure which is not present in EOPPC.21 Furthermore, peritoneal mesotheliomas have a male predilection and are rarely seen in women.22
BIOLOGY AND EPIDEMIOLOGY The histogenesis of this tumor is not known with certainty, but its origin may be a metaplastic process in the secondary M¨ullerian system.16 In 1972, Lauchlan first included the female peritoneum in the definition of the secondary
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
438
GYNECOLOGICAL CANCERS
M¨ullerian system.16 Since then, EOPPC is better understood as a neoplasm that arises from mesothelial cells under M¨ullerian influence. In 1974, Parmley and Woodruff demonstrated that pelvic peritoneum had the potential to differentiate into a M¨ullerian type of epithelium.20 Ovarian surface epithelium and the peritoneum share a common embryonic origin, and tumors with various histological types resembling EOC can develop primarily in the peritoneum.23 Embryologically, the coelomic epithelium invaginates to form the M¨ullerian ducts and also gives rise to the layer of mesothelial cells that line the peritoneal cavity and cover the cortical surfaces of the ovaries.17,24 EOPPC may thus arise from the multipotential mesothelial cells lining the peritoneal cavity. During fetal development, before the secretion of M¨ullerian inhibiting factor (MIF) by the fetal testes, the M¨ullerian ducts develop in male fetuses.17 MIF subsequently inhibits their development and they degenerate and completely disappear by 9–12 weeks. This M¨ullerian influence, however brief, may account for the rare occurrence of lesions resembling EOPPC in the male peritoneum, including rare lesions involving the tunica vaginalis testis, and epididymis.25 – 27 The true incidence of EOPPC remains unknown, although estimates place its frequency relative to ovarian cancer in the range of 1 : 10.1,28,29 Better recognition of this entity in recent years is probably responsible for a further increase in its relative frequency, which is approaching 18% of laparotomies performed for ovarian carcinoma and for revealing primary peritoneal carcinoma.7,30 A 14.9% increased frequency of peritoneal cancers was noted in an article titled “Annual report to the nation on the status of cancer (1973 through 1998)”, featuring cancers with recent increasing trends.31 The increasing incidence in females could reflect better classification practices due to a gradual realization and greater understanding of the characterization of primary peritoneal serous tumors and better identification of borderline ovarian and primary mucinous tumors compared with metastatic tumors. Initial studies focused on the epidemiologic differences between MM and EOPPC. MM are particularly rare in women in whom the most common malignant peritoneal neoplasm is the EOPPC. About two-thirds of patients with MM are middle-aged or elderly men, and in more than 80% of these cases there is evidence of asbestos exposure. In addition, MM responds poorly to chemotherapy and median survival time is short.32 There are numerous case reports of very small series of patients with EOPPC but larger series are rare. The paper concerning the largest series (74 patients) was published 15 years ago.1 This important study unveiled the characteristics of patients with EOPPC, and other studies compared the prognosis of EOPPC to that of EOC.2,30,33 Studies on the molecular pathogenesis suggest some differences in the cancers affecting these two regions. While ovarian cancer appears to have a uniclonal origin, at least some of the primary peritoneal cancers have a multifocal origin.29,34 The majority of studies have found few epidemiologic differences between peritoneal and epithelial ovarian malignancies. Several studies have shown, although others have not, that women with EOPPC are older than women with ovarian
cancer.8,35 Halperin et al. found that women with primary peritoneal cancer had a significantly earlier menarche, a higher number of births, and a lower incidence of family history for positive gynecologic malignancies, compared with women with ovarian cancer.36 It is not known whether pregnancy and oral contraceptive use decrease the risk of peritoneal cancer, as they do for ovarian cancer. Likewise, it is unclear whether factors that increase the risk of ovarian cancer, such as infertility, also increase the risk of primary peritoneal carcinomas. One of the important issues regarding ovarian cancer, recently discussed in the literature, is familial occurrence. Mutations of the tumor suppressor gene BRCA1 have been implicated in the development of familial ovarian and breast cancer.37,38 Women from families with multiple cases of breast and ovarian cancer, especially those who carry cancer-associated mutation of BRCA1 or BRCA2 are at increased lifetime risk for peritoneal carcinoma, even after previous surgery to remove the ovaries, fallopian tubes, and uterus. Over 90% of peritoneal cancer in patients with hereditary breast–ovarian cancer (HBOC) syndrome kindreds and associated with BRCA1 and BRCA2 mutations are serous carcinomas, which is equivalent to the proportion of ovarian cancers that are serous carcinomas in similar patients.39 Prophylactic oophorectomy has been found to reduce the risk for ovarian cancer in women from HBOC kindreds and those who carry cancer-associated BRCA1 and BRCA2 mutations, leaving a residual risk for peritoneal carcinomatosis of well less than 5%.40 The role of BRCA1 gene mutations in the development of EOPPC is uncertain.29 Bandera et al. screened 17 patients with peritoneal cancer and found BRCA1 germ-line mutations in 3 (17.6%) of 17 EOPPC patients.41 Karlan et al. reported that three EOPPC patients who underwent genetic testing carried BRCA1 mutations.34 They developed the disease at the age of 38, 39, and 55 years. In one of these cases the ovaries were histologically without evidence of malignancy, while in the other two cases the two ovaries and each of the metastatic sites studied differed in their patterns of oncogene overexpression. Bandera et al. identified in 2 of 17 EOPPC patients the 185 delAG germ-line BRCA1 mutation described in Ashkenazi Jewish population.42 The family history of one patient was notable, a mother and five aunts had breast or ovarian cancer. The other patient had a personal history of breast cancer. Both patients exhibited allelic loss of the normal BRCA1 allele in their tumor. A third patient was found to have a previously undescribed exon 11 single base pair substitution at nucleotide 1239 (CAG to CAC) resulting in a missense mutation (Gln to His). The patient had no family or personal history of breast or ovarian cancer, and her tumor did not exhibit loss of heterozygosity. Halperin et al. found two BRCA1 mutations, but no BRCA2 mutations in 28 women with EOPPC.36 Karlan et al. suggest that EOPPC may be a phenotypic variant of hereditary ovarian cancer.34 The fact that women with BRCA1 and possibly BRCA2 mutations are at increased risk for EOPPC has implications regarding recommendations for screening and prophylactic oophorectomy. In a study of 290 Jewish women who were at risk for ovarian cancer because of their
EXTRA-OVARIAN PRIMARY PERITONEAL CARCINOMAS
family history,43 intensive surveillance by use of CA-125 and ultrasound did not seem to be an effective means of diagnosing early stage ovarian cancer. During the study, six women developed stage IIIC peritoneal cancer, and all had had normal ovaries on transvaginal ultrasound imaging. All of the primary peritoneal cancers developed in women with BRCA1 mutations. Two studies evaluated the benefits of prophylactic oophorectomy and the potential risk of subsequent EOPPC. Kauff et al. studied 170 women with BRCA1 or BRCA2 mutations, 35 years of age or older, who had not undergone bilateral oophorectomy but chose to undergo either surveillance for ovarian cancer or risk-reducing salpingo-oophorectomy.44 During a mean follow-up of 24.2 months, breast cancer was diagnosed in 3 of the 98 women who chose risk-reducing salpingo-oophorectomy and peritoneal cancer was diagnosed in 1 woman in this group. Among the 72 women who chose surveillance, breast cancer was diagnosed in 8, ovarian cancer in 4, and peritoneal cancer in 1. The time to breast cancer or BRCA-related gynecologic cancer was longer in the salpingo-oophorectomy group, with a hazard ratio for subsequent breast cancer or BRCA-related gynecologic cancer of 0.25 (95 percent confidence interval, 0.08 to 0.74). In the study by Rebbeck et al., there were 551 women with disease-associated germ-line BRCA1 or BRCA2 mutations.45 Ovarian cancer developed in 259 women who had undergone bilateral prophylactic oophorectomy and in 292 matched controls who had not undergone the procedure. In a subgroup of 241 women with no history of breast cancer or prophylactic mastectomy, the incidence of breast cancer was determined in 99 women who had undergone bilateral prophylactic oophorectomy and in 142 matched controls. The length of postoperative follow-up for both groups was at least 8 years. Six women who underwent prophylactic oophorectomy (2.3%) received a diagnosis of stage I ovarian cancer at the time of the procedure; two women (0.8%) received a diagnosis of papillary serous peritoneal carcinoma 3.8 and 8.6 years after bilateral prophylactic oophorectomy. Among the controls, 58 women (19.9%) received a diagnosis of ovarian cancer, after a mean follow-up of 8.8 years. With the exclusion of the six women whose cancer was diagnosed at surgery, prophylactic oophorectomy significantly reduced the risk of coelomic epithelial cancer. Of the 99 women who underwent bilateral prophylactic oophorectomy and who were studied to determine the risk of breast cancer, breast cancer developed in 21 (21.2%), as compared to 60 (42.3%) in the control group. Each of these studies concluded that bilateral prophylactic oophorectomy significantly reduced the risk of breast cancer and EOC. However, the risk of peritoneal cancer remains.
PATHOLOGY Histopathologic similarities have been drawn between EOPPC and PSOC.6 For study purposes, the Gynecologic Oncology Group (GOG) developed a set of criteria for EOPPC diagnosis:15 1. The ovaries are either absent or normal in size (<4 cm largest diameter).
439
2. The involvement in the extraovarian sites is greater than that of the surface of either ovary. 3. Microscopically, the ovaries are either not involved with tumor or exhibit only serosal or cortical implants, less than 5 mm in depth. 4. The histopathological and cytological characteristics of the tumor are predominantly of the serous type. Other types of EOPPC have been described, including endometrioid, mucinous, clear cell, Brenner, and malignant mixed M¨ullerian types.4,46 – 48 Eltabbakh et al. reported that the median survival of patients who had nonserous histology was not significantly different from that of patients with serous histology.22 Altaras et al. reported one patient with papillary clear cell EOPPC alive without evidence of disease 76 months following diagnosis.4 Two EOPPC variations appear less virulent.49,50 The peritoneal serous borderline tumors (PSBTs) behave similarly to their ovarian counterparts. These PSBTs have an excellent prognosis, although in rare cases transformation to carcinoma has been observed on follow-up examination.49,51 Forty-one to ninety-nine percent PSBTs are accompanied by endosalpingiosis, suggesting an origin therein.49,51,52 Histologically, PSBTs are identical to the peritoneal implants found in association with ovarian serous borderline tumors. Patients with PSBTs are typically under the age of 35 years, and often infertile.52 Malignant tumors are rare in young women.30 The second, less virulent variety is the serous psammocarcinoma of the peritoneum. This latter group has a proportionately larger number of psammoma bodies, and less aggressive cytologic appearance with absent or moderate nuclear atypicality and rare mitotic figures. Conservative management, both fertility-preserving surgery and the withholding of adjuvant chemotherapy, should be considered in the management of this disease.53 The presently accepted criteria for making the distinction between EOC and EOPPC are based on minimal scientific evidence. Therefore, it is likely that some tumors designated as EOPPC are actually small ovarian cancers that find the peritoneum a more hospitable site for growth than the ovary. Possible reasons why it may provide a more favorable environment is the greater density of the ovarian tissue, which may inhibit invasion of tumor cells originating within superficial layers, and the production of a tumorinhibitory substance by the ovarian stroma,54 which has been demonstrated in vitro.55 Besides separating EOPPC and EOC, there are different entities that need to be distinguished (see Table 1). Histologically, EOPPC should be included in the differential diagnosis of all papillary serous lesions of the peritoneum, including endosalpingiosis, mesothelial hyperplasia, MM, and metastatic adenocarcinoma.49,52 1. Endosalpingiosis: Refers to the presence of histologically benign glands lined by tubal-type epithelium outside the confines of the fallopian tube. This disorder is often associated with chronic salpingitis and ovarian serous borderline tumors. The association with chronic salpingitis suggests that shedding of tubal epithelial cells onto the peritoneum is a route of development of endosalpingiosis.
440
GYNECOLOGICAL CANCERS
Table 1 Differential diagnosis of EOPPC.
Variations of EOPPC Peritoneal serous borderline tumors Psammocarcinomas of the peritoneum Endosalpingiosis Florid mesothelial hyperplasia Malignant mesotheliomas Metastatic adenocarcinoma EOPPC, extraovarian primary peritoneal carcinoma.
Endosalpingiosis is also a legitimate candidate, a precursor to EOPPC. In one series, 2 out of 14 carcinomas of this type were associated with this disorder.51 2. Florid mesothelial hyperplasia: Hyperplasia of mesothelial cells is a common response to inflammation and chronic effusions. Psammoma bodies may be present but rarely are as numerous as in EOPPC.24 3. MMs: Under light microscopy, differentiation between the two types of tumor may be difficult. Kannerstein et al. describe some morphologic and histochemical differences between the mesothelioma and EOPPC.3,22 Although not of absolute differential diagnostic value, the presence of psammoma bodies, epithelial mucin, columnar cells, and the absence of hyaluronic acid favors the diagnosis of EOPPC over mesothelioma. Warhol et al. compared the ultrastructural features of MM and EOC, which are clinically similar to EOPPC.56 They found that the mesotheliomas were characterized ultrastructurally by a greater content of tonofilaments, lack of mucin, fewer cilia, and dense-core granules of the neurosecretory type. Mesotheliomas may exactly mimic EOPPC, and definitive distinction requires IHC and/or electron microscopy.57 Although the signs and symptoms of peritoneal mesothelioma and EOPPC are similar, the response to treatment and survival is generally poorer in patients with peritoneal mesothelioma. MM of the pleura are approximately 10 times as frequent as those of the peritoneum, and MM of the peritoneum is extremely rare in the absence of pleural MM and a history of asbestos exposure.58,59 Epidemiologically, MM is more common in older men, especially men with a history of asbestos exposure. EOPPC, an uncommon tumor in women, is likely to be overlooked in men, unless it is considered in the differential diagnosis. In males, another malignant tumor that can mimic an EOPPC is MM of the tunica vaginalis testis.26,59 4. Metastatic adenocarcinoma: This is ruled out by the absence of a primary tumor elsewhere. Moll et al. demonstrated p53 overexpression in 83% of 29 EOPPC patients.60 The authors did not discuss the significance of p53 overexpression on survival. Kowalski et al. described p53 overexpression in 48% of 44 EOPPC patients, similar to the 59% incidence in patients with EOC.7 The authors did not find that p53 overexpression was predictive of prognosis within the EOPPC or the EOC groups. Ben-Baruch et al., using IHC of archival material, demonstrated p53 overexpression in 42.4% of 75 EOPPC patients.33 EOPPC patients, whose tumors demonstrated p53 overexpression, had a shorter median survival than those
whose tumors did not (11 vs 23.5 months, respectively). The difference did not achieve statistical significance.
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS Although several studies have described the clinicopathologic features of this tumor entity, the clinical behavior of EOPPC remains obscure.1 – 4,6,11,14,15,22,30,50,61,62 It is not clear whether the EOPPC patients differ from PSOC patients with regard to their epidemiological characteristics. It has been suggested that biologically these tumors behave similarly to EOC of similar stage.63 The natural history is similar to the ovarian counterpart, with the disease occurring mostly in women between 57 and 67 years and with a clinical presentation that does not differ from that of advanced stages of EOC.4 Ben-Baruch et al. did not find significant differences in patient’s characteristics between EOPPC and EOC with regard to mean age at diagnosis, parity, and menopausal status.33 The mean age of EOPPC diagnosis was 61.1 years, similar to that reported by Ransom et al., Fowler et al., Altaras et al., and Fromm et al. who found a mean age of 60, 61.4, 61.2, and 57.4 years, respectively.1,4,5,62 Ben-Baruch et al. did not demonstrate significant differences between EOC and EOPPC in the operative findings.33 The risk of ascites fluid volume exceeding 1000 mL and the proportion of patients with stage IV disease was the same among these two entities. The rate of stage IV disease was found to be 28%, which is similar to the rate of 29% and 32% reported by Altaras et al. and Killackey and Davis.4,50 To our knowledge, there have been two reports of this tumor occurring in children. In 1983, Ulbright et al. described a 11-year-old girl with an extraovarian serous carcinoma of the retroperitoneum, and Wall et al. reported an adolescent girl with EOPPC whose tumor responded to paclitaxel after showing limited response to two other chemotherapeutic regimens, one of which included carboplatin.64 The most common symptoms are abdominal pain and distension.1,2,4,6,11,13 – 15,30,50,61,62 Other EOPPC patients also presented constipation, nausea, vomiting, loss of weight, loss of appetite, malaise, dyspareunia, and urinary symptoms.1,22 Common presenting signs include ascites, pelvic-abdominal mass and peripheral edema.1,22 The presence of psammoma bodies in the cervicovaginal smear of a EOPPC patient has also been documented.65 The common sites of disease at the time of laparotomy are the omentum, abdominal and pelvic peritoneum, ovaries, and serosa of the bowel.1,61,66 Eltabbakh et al. suggested that the incidence of central nervous system metastases in EOPPC patients (1.4%) is similar to that of EOC patients.67 A patient with inappropriate antidiuretic hormone secretion after suboptimal cytoreduction of a stage III EOPPC has also been described.68 Altaras et al. and Rose et al. reported the usefulness of CA125 in diagnosis and follow-up of EOPPC patients.4,66 In patients whose preoperative CA125 values are known, this tumor marker was elevated in 94.4% of the cases.61 Mills et al. reported elevated CA125 values in eight EOPPC patients.2 Altaras et al. described elevated CA125 values in three EOPPC patients and found that CA125 measurements correlated with the clinically determined status of the
EXTRA-OVARIAN PRIMARY PERITONEAL CARCINOMAS
disease.4 Similar to the situation in EOC patients, CA125 values may be useful in the diagnosis of EOPPC patients and follow-up of their response to therapy. The prognostic significance of estrogen and progesterone receptor analysis in patients with EOPPC is controversial.69 – 71 A confounding problem when comparing studies on estrogen and progesterone receptors from different institutes is the different ways in which laboratories determine estrogen and progesterone receptor status. Eltabbakh et al. employed IHC, a technique that reduces the number of false elevated results.61,70 In their study of EOPPC patients, estrogen and progesterone receptors were positive in 50% and 6.3% of the cases, respectively. Estrogen and progesterone positivity did not correlate significantly with survival. However, the median survival of EOPPC patients whose tumors were progesterone receptor positive was almost twice that of patients whose tumors were progesterone receptor negative (40 vs 21.2 months). In practice, the difficulty is to identify EOPPC patients without subjecting patients with other less treatable malignancies to invasive procedures. In this group of patients, exhaustive investigations are usually carried out, including computed tomography (CT) of the abdomen and pelvis, barium studies, and endoscopy of the gastrointestinal tract. In patients who present ascites as the sole clinical feature (absence of a pelvic or abdominal mass), a diagnostic paracentesis is performed, and malignant cells may be seen on cytology. Some authors have suggested that the detection of signet ring cells in ascites can be taken to exclude primary ovarian or peritoneal carcinoma.72 In the absence of these findings, all women with ascites containing adenocarcinoma cells should undergo laparoscopy
and/or laparotomy as a diagnostic and potentially therapeutic debulking procedure. If during surgery no primary site is found but there are peritoneal tumor deposits, then it is recommended the patient undergo a total hysterectomy, bilateral salpingo-oophorectomy, omentectomy, and cytoreductive surgery followed by adjuvant chemotherapy (see Figure 1). In short, the surgical management is as for ovarian cancer.
TREATMENT There is no separate staging system for EOPPC. Most investigators have used the International Federation of Gynecology and Obstetrics (FIGO) staging system for ovarian cancer in EOPPC patients.73 Most cases reported in the literature have been stage III or IV. Raju et al. suggested that EOPPC should be treated as ovarian tumors of similar grade and stage.6 Their clinical behavior, including response to treatment, is equivalent to that of EOC with a comparable extent of disease.1,4,6 Accordingly, the standard regimen used in the treatment of EOC is generally administered to EOPPC patients.1,62 The goal of surgical treatment is cytoreduction to no gross residual disease (see Table 2).62 Chemotherapy is commonly administered after cytoreductive surgery. It is generally agreed that these patients should be managed following the aggressive chemotherapeutic regimens established for patients with advanced ovarian carcinoma. However, because of the rarity of this tumor, reported experience with chemotherapeutic agents is limited (see Table 3). Since 1979, cisplatin-based multiagent chemotherapy has been the standard treatment for patients with EOC and, consequently, for EOPPC patients. Several authors have reported that the response of EOPPC
Ascites (absence of a pelvic or abdominal mass)
Diagnostic paracentesis
Signet ring cells in ascites
Primary ovarian or peritoneal carcinoma is excluded
No more invasive procedures
Adenocarcinoma cells
Laparoscopy and/or laparotomy (diagnostic and potentially therapeutic)
Peritoneal tumor deposits without primary site
Surgical management is as for ovarian cancer
Chemotherapy established for patients with advanced ovarian carcinoma
Figure 1 Treatment algorithm.
441
442
GYNECOLOGICAL CANCERS
Table 2 Survival of EOPPC based on cytoreduction status.
Study
Patients
Strand et al.9
18
Fromm et al.1
74
Fowler et al.5
36
Ben-Baruch et al.33
25
Eltabbakh et al.61
75
Kennedy et al.75
36
Residual disease
Median survival time (months)
<3 >3 <2 >2 <1.5 >1.5 <2 >2 <1 >1 <1 >1
31 11 25 26 18 17 46 20 40 18 N/A 32.8
N/A, nonavailable.
Table 3 Response of EOPPC to first-line chemotherapy.
Study Chen and Flam14
Number of patients 3
Mills et al.2
10
Dalrymple et al.30
31
Fromm et al.1 Altaras et al.4
44 7
Bloss et al.15
33
Menzin et al.76 Piver et al.77
4 46
Bloss et al.78 Kennedy et al.75
N/A 38
Regimen PAC (N = 2), DC (N = 1) PAC (N = 3), PC (N = 4), alkylators (N = 2) Platinum-based (N = 26), chlorambucil (N = 5) PC, P, M (N = 44) PAC (N = 5), PC (N = 2) PAC (N = 29), PC (N = 4) TP (N = 4) PAC (N = 25) TP (N = 21) PC Paclitaxel/Ca (N = 26) Paclitaxel/P (N = 12)
Overall response (%) 100 80
32.3 63.6 100 63.6 100 62.5 70 65 87
P, cisplatin; PAC, cisplatin, doxorubicin, cyclophosphamide; PC, cisplatin, cyclophosphamide; TP, paclitaxel, cisplatin; Ca, carboplatin; M, melphalan; N/A, nonavailable; DC, doxorubicin, cisplatin.
patients to platinum-based chemotherapy is similar to that of PSOC patients and have subsequently recommended treating patients with EOPPC in a fashion similar to that used in patients with EOC.1,4,9,14,15,30,62,74 Other investigators, however, have failed to confirm these findings.2,11 Long-term survival has been reported predominantly in patients with optimal cytoreduction and platinum-based chemotherapy.14,62 Numerous regimens have been utilized with varying degrees of success. Recommended chemotherapy has evolved into combination treatment with taxanes and platinum. Fromm et al. described 74 patients with primary peritoneal cancer treated with surgery and postoperative chemotherapy, of whom 29.2% received single-agent therapy while 70.8% were given a multiagent regimen.1 Half of all patients received cisplatin, either singly or in combination with other cytotoxic drugs. An overall response rate to first-line chemotherapy (multiple-regimens) of 63.6% was achieved
with a median survival of 24 months. They achieved a complete response in 22.7% and a partial response in 40.9% of patients. In 31.8% cases, the disease progressed despite chemotherapy. They demonstrated that patients who received cisplatin-based regimens had a median survival of 31.5 months, whereas those who did not had a median survival of 19.5 months. This difference was not statistically significant. Survival rate was not influenced by residual tumor at primary surgery. The improved response seen with platinum-based regimens has been confirmed by other investigators.4,15,30,50,62 A number of authors have reported response rates of approximately 60% to platinum-based chemotherapy after cytoreduction.5,15 These response rates were comparable to those achieved by the authors at their own institutions with similar combinations in patients with ovarian carcinomas, and who have subsequently recommended treating patients with EOPPC in a fashion similar to that used in patients with EOC. Several case –control studies have compared the response rates of ovarian and peritoneal carcinoma to platinumbased chemotherapy. Mulhollan et al. evaluated 33 cases of EOPPC and compared them with 54 cases of EOC.74 With at least a 4-year follow-up period, no differences in the median survival time (17 vs 18 months) were noted. Bloss et al. reported a retrospective, case –controlled study comparing the response and survival to cytoreductive surgery followed by cisplatin-based chemotherapy of 33 women, with PSCP versus 33 cases with EOC.15 The authors concluded that there was no significant difference in median survival between the cases (20.8 months) and controls (27.8 months). In two other case –control studies, patients with EOPPC were reported to have reduced response rates to platinumbased chemotherapy, compared with control patients with EOC.50,79 The GOG conducted a prospective phase II trial of cisplatin and cyclophosphamide in 36 women with EOPPC.78 Additionally, the study compares these patients with 130 women with PSOC undergoing identical therapy to determine if it may be reasonable to include EOPPC patients in future ovarian trials. After primary surgery, patients were treated with cisplatin and cyclophosphamide every 21 days for six cycles. Those women without clinical evidence of disease after study treatment then underwent a reassessment laparotomy. The clinical response rate for EOPPC to the treatment regimen was 65% compared to 59% for women with PSOC. Surgical complete responses were identical in the two groups (20%). While overall survival was not significantly different between the two groups, women with EOPPC demonstrated a greater risk of progression-free survival failure than women with PSOC. Based on this study, women with EOPPC were included in all GOG treatment studies of EOC. In 1996, a randomized GOG trial demonstrated a significant survival advantage in patients with advanced EOC, whose residual disease was >1 cm, treated with paclitaxel plus cisplatin compared to similar patients who were treated with cisplatin plus cyclophosphamide.80 As a result of this
EXTRA-OVARIAN PRIMARY PERITONEAL CARCINOMAS
study, the combination of paclitaxel and cisplatin is considered first-line chemotherapy for patients with EOC. The impact on response and survival of this combination, and the relative contribution of adding a taxane to the first-line therapy in EOPPC patients, have not been thoroughly evaluated. Four EOPPC patients treated with paclitaxel (135 mg m−2 ) and cisplatin (50–75 mg m−2 ) for six cycles have been reported.76 Reassessment surgery demonstrated complete surgical response in one and partial surgical response in three patients. Kennedy et al. treated 38 EOPPC patients (36 stage IIIC and 2 stage IV).75 All patients received paclitaxel (135 mg m−2 or 175 mg m−2 ), and cisplatin or carboplatin. Median progression-free survival was 15 months and median overall survival 40 months. Survival for optimally debulked patients was significantly better than for suboptimally debulked patients (median 32.8 months). Other authors have found that the addition of paclitaxel to cisplatin does not improve response rates for patients with EOPPC.61 Eltabbakh evaluated extreme drug resistance (EDR) assays among 20 consecutive women with EOPPC.81 The results of the EDR assay and response to chemotherapy were compared with those among women with EOC. There was no significant difference in the incidence of EDR to cisplatin, carboplatin, paclitaxel, doxorubicin, cyclophosphamide, ifosfamide, etoposide, hexamethylmelamine, and topotecan among patients with EOPPC and those with EOC. The response rate of EOPPC patients to chemotherapy was 80% and unrelated to EDR to the individual drugs used in combination chemotherapy. The EDR profile and response to cisplatin-based chemotherapy among women with EOPPC were similar to those among women with EOC. These findings support treating both conditions similarly. EDR to individual drugs does not preclude response to combination chemotherapy. The combination of paclitaxel and a platinum compound is active in the re-treatment of patients with ovarian or peritoneal carcinoma who had disease recurrence >or = 6 months following this combination.82 Since, in preclinical and clinical trials, paclitaxel has been effective in treating advanced and platinum-refractory EOC, it should be considered in the treatment of patients with platinum-resistant EOPPC.83 – 85 Eltabbakh et al. reported that patients who received paclitaxel alone or in combination as second-line chemotherapy had significantly longer survival than patients who received chemotherapy without paclitaxel (median survival 23 vs 8.2 months, respectively, p = 0.026).61 In one report, a paclitaxel dose of 175 mg m−2 given as 24-hour continuous intravenous infusion resulted in a rapid partial response with good palliation of symptoms.85 Another case reported a complete response after therapy with a paclitaxel dose of 420 mg m−2 given over 24 hours as a continuous intravenous infusion in phase I trial.64 The fact that paclitaxel may exhibit a dose-response in the treatment of EOC can explain the presence of a complete response in the case that used 420 mg m−2 and only a partial response when 210–250 mg m−2 dose was used.86,87 Further studies are required to determine the best dose schedule of paclitaxel. Carboplatin may be active in paclitaxel-resistant EOPPC. This clinical phenomenon has been reported recently in
443
patients treated with EOC, previously treated with paclitaxel and whose most recent platinum-based therapy was 12 months prior to carboplatin reinduction.88 It has also been reported in patients with ovarian carcinoma whose primary treatment was single-agent paclitaxel.89 Carboplatin therapy should be considered in patients with paclitaxel refractory EOPPC without platinum chemotherapy for 12 months.90 Other forms of treatment have also been described, including radiotherapy and hormonal treatments with estrogen and progesterone preparations.1 However, their roles in treatment are undefined.
PROGNOSIS The prognosis of EOPPC compared to EOC continues to be a subject of debate in the literature. Several comparative or case-control studies seem to suggest that these two tumors have a similar prognosis.1,9,15,30,62,91 While other series suggest poorer survival in EOPPC,2,11,13,50 some investigators have reported better prognosis in EOPPC.14,92 The prognostic factors in patients with EOC are better defined than those in patients with EOPPC. Multivariate analysis of 21 240 cases with primary EOC showed that stage, histology, grade, age, presence of ascites, lymph node status, and race were predictors of survival.93 Eltabbakh et al. demonstrated that age, surgical stage, performance status, and degree of cytoreductive surgery are significant prognostic factors in EOPPC patients, and that performance status and primary debulking surgery are independent factors.61 Mulhollan et al. investigated the significance of ovarian involvement on survival of EOPPC patients.74 These authors demonstrated that the size of the ovarian tumor and the amount of ovarian stromal invasion had no significant effect on survival. The amount of residual disease may be an important prognostic factor for both EOPPC and EOC patients.9,33,94 However, controversy exits concerning the ability to perform optimal surgical debulking and its clinical significance in EOPPC. Optimal cytoreductive surgery defined as <2 cm of residual tumor, has been accomplished in 33 to 69% of patients with EOPPC.1,2,62 Strand et al. found that successful surgical cytoreduction resulted in better response to chemotherapy and prolonged survival.9 Fromm et al. were able to accomplish optimal debulking surgery only in 41.2% of 74 patients with EOPPC.1 They demonstrated that patients receiving multiagent chemotherapy (i.e. cisplatin and cyclophosphamide) had a statistically greater median survival rate compared to those receiving a single-agent regimen (melphalan). The median survival of EOPPC patients treated with combination chemotherapy was very similar to that of EOC patients who received the same treatment. Fromm et al. observed that neither age nor the presence of residual disease ≥2 cm after cytoreductive surgery was predictive of survival. Among the pathological factors these authors examined, only the absence of mitosis was significantly predictive in survival. The presence of vascular invasion and the proportion of papillary areas in the tumor failed to predict survival. Mills et al. as well as Fromm et al. did not find a significant prognostic value in optimal cytoreductive surgery
444
GYNECOLOGICAL CANCERS
in EOPPC patients.1,2 In the study by Ransom et al., only three patients who had long-term survival had optimal tumor cytoreduction.62 Fowler et al. found that patients who underwent optimal cytoreduction had an increased survival rate, although the difference did not reach statistical significance.5 These results are similar to those of BenBaruch et al.33 In their study comparing 25 EOPPC patients with stages III-IV PSOC, the survival was significantly better in patients with optimal debulking only in the EOC group. The survival curve for patients with a residual disease ≥ 2 cm was almost identical for the EOPPC and EOC patients. Their rate of successful debulking and the result with platinum-based combination chemotherapy were the same in both groups. This is in contrast to a previous report of a lower rate of optimal cytoreduction and decreased response to platinum-based chemotherapy in the EOPPC group.50 Eltabbakh et al. demonstrated that optimal cytoreductive surgery resulting in less than 1 cm residual tumor was achieved in 65.3% of their patients.61 Optimal cytoreductive surgery was a favorable prognostic factor in both univariate and multivariate analysis. The difference in the results of these investigators could be explained by the number of patients and the definition of optimal cytoreductive surgery. Eltabbakh et al. also suggested a possible value of secondary cytoreductive surgery in EOPPC patients that was not discussed in previous reports. They found that patients who underwent cytoreductive surgery following recurrence or progression of disease had longer survival rate than patients who did not undergo secondary cytoreductive surgery (median survival 12.2 vs 3.1 months, respectively). The overall 5-year survival is about 20%, which is similar to the survival rate of patients with advanced ovarian cancer.30,62 These results were achieved with surgery without lymphadenectomy.95 The question remains whether retroperitoneal lymphadenectomy improves the survival rate of EOPPC patients. In addition, the inclusion of paclitaxel into adjuvant treatment protocols and/or the use of high-dose chemotherapy may improve the survival in EOPPC patients and has to be proven in clinical trials. Median survivals of 21, 17, 17.8, 19, 23, 24, and 23.5 months of EOPPC patients have been documented.1,5,9,30,33,61,62 However, long-term survival after chemotherapy of more than 5 years has been documented in the literature.14 Petru et al. reported a median survival of only 10 months for 14 EOPPC patients.94 However, since seven of these patients did not undergo surgical debulking, he concluded that the amount of residual disease might present an important prognostic factor.62 Foyle et al. reported that out of nine patients with documented follow-up, eight were dead 1.5 years after diagnosis.13 White et al. in a 1985 publication of 11 cases, found that eight patients were dead within 3 years of diagnosis despite chemotherapy.28 They reported median survival of 15 and 16 months for single (n = 3) and multiple (n = 8) agent chemotherapy regimens given as first-line.
AUTHORS’ RECOMMENDATIONS It is important to consider this diagnosis in any woman with ascites but no pelvic mass. The differential diagnosis of
diffuse peritoneal involvement includes both neoplastic and non-neoplastic lesions (peritoneal tuberculosis). Neoplastic lesions include malignant peritoneal mesothelioma, pseudomyxoma peritonei, and diffuse peritoneal carcinomatosis. The most common primary sites for peritoneal carcinomatosis, apart from the ovary, are the uterus and the cervix, the kidneys, the pancreas, and the breast. MM of the peritoneum is extremely rare in the absence of malignant pleural mesothelioma and without a history of asbestos exposure. It is more common in middle-aged men. The clinical management is the same as for ovarian cancer and it is important that these patients be directed to a gynecological cancer center. Failure to consider the diagnosis can result in treatment delay or inappropriate referral. We may never be able to scientifically distinguish between primary ovarian and primary peritoneal serous carcinomas in every case. EOPPC patients should be reported separately from those with ovarian carcinoma but should be treated in a similar fashion.
REFERENCES 1. Fromm GL, Gershenson DM, Silva EG. Papillary serous carcinoma of the peritoneum. Obstet Gynecol 1990; 75: 89 – 95. 2. Mills SE, et al. Serous surface papillary carcinoma: a clinicopathologic study of 10 cases and comparison with stage III-IV ovarian serous carcinoma. Am J Surg Pathol 1988; 12: 827 – 34. 3. Kannerstein M, et al. Papillary tumors of the peritoneum in women: mesothelioma or papillary carcinoma? Am J Obstet Gynecol 1977; 127: 306 – 14. 4. Altaras MM, et al. Primary peritoneal papillary serous adenocarcinoma : clinical and management aspects. Gynecol Oncol 1991; 40: 230 – 6. 5. Fowler JM, et al. Peritoneal adenocarcinoma (serous) of M¨ullerian type: a subgroup of women presenting with peritoneal carcinomatosis. Int J Gynecol Cancer 1994; 4: 43 – 51. 6. Raju U, et al. Primary papillary serous neoplasia of the peritoneum: a clinico-pathological and ultrastructural study of eight cases. Hum Pathol 1989; 20: 426 – 36. 7. Kowalski LD, et al. A Matched-case comparison of extraovarian versus primary ovarian adenocarcinoma. Cancer 1997; 79: 1587 – 94. 8. Eltabbakh GH, et al. Epidemiologic differences between woman with extra-ovarian primary peritoneal carcinoma and women with epithelial ovarian cancer. Obstet Gynecol 1998; 91: 254 – 9. 9. Strand CM, et al. Peritoneal carcinomatosis of unknown primary site in women. A distinct subset of adenocarcinoma. Ann Int Med 1989; 111: 213 – 7. 10. Fox H. Primary neoplasia of the female peritoneum. Histopathology 1993; 23: 103 – 10. 11. Gooneratne S, et al. Serous papillary carcinoma of the ovary. A clinicopathologic study of 16 cases. Int J Gynecol Pathol 1982; 1: 258 – 69. 12. Swerdlow M. Mesothelioma of the pelvic peritoneum resembling papillary cystadenocarcinoma of the ovary: case report. Am J Obstet Gynecol 1959; 77: 197 – 200. 13. Foyle A, Al-Jabi M, Mc Caughey WTE. Papillary peritoneal tumors in woman. Am J Surg Pathol 1981; 5: 241 – 9. 14. Chen KTK, Flam MS. Peritoneal papillary serous carcinoma with longterm survival. Cancer 1986; 58: 1371 – 3. 15. Bloss JD, et al. Extraovarian peritoneal serous papillary carcinoma: a case-control retrospective comparison to papillary adenocarcinoma of the ovary. Gynecol Oncol 1993; 50: 347 – 51. 16. Lauchlan SC. The secondary M¨ullerian system. Am J Obstet Gynecol 1972; 27: 133 – 46. 17. Parmley T. Embryology of the female genital tract. In Kurman RJ (ed) Blaustein’s Pathology of the Female Genital Tract, 3rd ed. New York: Springer-Verlag New York Inc., 1987: 1 – 14.
EXTRA-OVARIAN PRIMARY PERITONEAL CARCINOMAS 18. Hertig AT. Proceedings of Eighteenth Seminar. Chicago, Illinois: American Society of Clinical Pathology, 1952: 49. 19. Rosenbloom MA, Foster RB. Probable pelvic mesothelioma. Report of a case and review of literature. Obstet Gynecol 1961; 18: 213 – 22. 20. Parmley TH, Woodruff JD. The ovarian mesothelioma. Am J Obstet Gynecol 1974; 120: 234 – 41. 21. Selikoff IJ, Churg J, Hammond EC. Relation between exposure to asbestos and mesothelioma. N Engl J Med 1965; 272: 560 – 5. 22. Kannerstein M, Churg J. Peritoneal mesothelioma. Hum Pathol 1977; 8: 83 – 94. 23. Gompel C, Silverberg SG. The female peritoneum. In Gompel C, Silverberg SG (eds) Pathology in Gynecology and Obstetrics. Philadelphia, Pennsylvania: Lippincott, 1994: 414 – 447. 24. Clement PB. Endometriosis, lesions of the secondary M¨ullerian system, and pelvic mesothelial proliferation. In Kurman RJ (ed) Blaustein’s Pathology of the Female Genital Tract, 3rd ed. New York: SpringerVerlag New York Inc., 1987: 516 – 559. 25. Shah IA, et al. Papillary serous carcinoma of the peritoneum in a man. Cancer 1998; 82: 860 – 6. 26. Jones MA, Young RH, Scully RE. Malignant mesothelioma of the tunica vaginalis:a clinicopathologic analysis of 11 cases with review of the literature. Am J Surg Pathol 1995; 19: 815 – 25. 27. Remmle W, et al. Serous papillary cystic tumor of borderline malignancy with focal carcinoma arising in testis: a case report with immunohistochemical and ultrastructural observations. Hum Pathol 1992; 23: 75 – 9. 28. White PF, Merino MJ, Barwick KW. Serous surface papillary carcinoma of the ovary: a clinical, pathological, ultrastructural, and immunohistochemical study of 11 cases. Pathol Annu 1985; 20: 403 – 18. 29. Eltabbakh GH, Piver MS, Werness BA. Primary peritoneal adenocarcinoma metastatic to the brain. Gynecol Oncol 1997; 66: 160 – 3. 30. Dalrymple JC, et al. Extraovarian peritoneal serous papillary carcinoma. A clinicopathologic study of 31 cases. Cancer 1989; 64: 110 – 5. 31. Howe HL, et al. Annual report to the nation on the status of cancer (1973 through 1998), featuring cancers with recent increasing trends. J Natl Cancer Inst 2001; 93: 824 – 42. 32. Markaki S, et al. Primary malignant mesothelioma of the peritoneum: a clinical and immunohistochemical study. Gynecol Oncol 2005; 96: 860 – 4. 33. Ben-Baruch G, et al. Primary peritoneal serous papillary carcinoma: a study of 25 cases and comparison with stage III-IV ovarian papillary serous carcinoma. Gynecol Oncol 1996; 60: 393 – 6. 34. Karlan BY, et al. Secreted ovarian stromal substance inhibits ovarian epithelial cell proliferation. Gynecol Oncol 1995; 59: 67 – 74. 35. Chu CS, et al. Primary peritoneal carcinoma: a review of the literature. Obstet Gynecol Surv 1999; 54: 323 – 35. 36. Halperin R, et al. Primary peritoneal serous papillary carcinoma: a new epidemiologic trend? A matched-case comparison with ovarian serous papillary cancer. Int J Gynecol Cancer 2001; 11: 403 – 8. 37. Miki Y, et al. A strong candidate for the breast and ovarian susceptibility gene BRCA1. Science 1994; 266: 66 – 71. 38. Shattuck-Eidens D, et al. A collaborative survey of 80 mutations in the BRCA1 breast and ovarian cancer susceptibility gene. JAMA 1995; 273: 535 – 41. 39. Casey MJ, Bewtra C. Peritoneal carcinoma in women with genetic susceptibility: implications for Jewish populations. Fam Cancer 2004; 3: 265 – 81. 40. Piver MS. Hereditary ovarian cancer: lessons from the first twenty years of the Gilda Radner familial ovarian cancer registry. Gynecol Oncol 2002; 85: 9 – 17. 41. Bandera CA, et al. Germline BRCA1 mutations in women with papillary serous carcinoma of the peritoneum(EOPPC). Proc Am Assoc Cancer Res 1997; 38: 82. 42. Bandera CA, et al. BRCA1 gene mutations in women with papillary serous carcinoma of the peritoneum. Obstet Gynecol 1998; 92: 596 – 600. 43. Liede A, et al. Cancer incidence in a population of Jewish women at risk of ovarian cancer. J Clin Oncol 2002; 20: 1570 – 7.
445
44. Kauff ND, et al. Risk-reducing salpingo-oophorectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 2002; 346: 1609 – 15. 45. Rebbeck TR, et al. Prophylactic oophorectomy in carriers of BRCA1 or BRCA2 mutations. N Engl J Med 2002; 346: 1616 – 22. 46. Clark JE, et al. Endometrioid-type cystadenocarcinoma arising in the mesosalpinx. Obstet Gynecol 1979; 54: 656 – 8. 47. Lee KR, Verma U, Belinson JL. Primary clear cell carcinoma of the peritoneum. Gynecol Oncol 1991; 41: 259 – 62. 48. Mirc JL, Fenoglio-Preiser CM, Husseinzadeh N. Malignant mixed M¨ullerian tumor of extraovarian secondary M¨ullerian system: report of two cases and review of the literature. Arch Pathol Lab Med 1995; 119: 1044 – 9. 49. Bell DA, Scully RE. Serous borderline tumors of the peritoneum. Am J Surg Pathol 1990; 14: 230 – 9. 50. Killackey MA, Davis AR. Papillary serous carcinoma of the peritoneal surface: matched-case comparison with papillary serous ovarian carcinoma. Gynecol Oncol 1993; 51: 171 – 4. 51. Weir MM, Bell DA, Young RH. Grade 1 peritoneal serous carcinoma: a report of 14 cases and comparison with 7 peritoneal serous psammocarcinomas and 19 peritoneal serous borderline tumors. Am J Surg Pathol 1998; 22: 849 – 62. 52. Biscotti CV, Hart WR. Peritoneal serous papillomatosis of low malignant potential (serous borderline tumors of the peritoneum): a clinicopathologic study of 17 cases. Am J Surg Pathol 1992; 16: 467 – 75. 53. Whitcomb BP, et al. Primary peritoneal psammocarcinoma: a case presenting with an upper abdominal mass and elevated CA-125. Gynecol Oncol 1999; 73: 331. 54. Scully RE. The Eltabbakh/Piver article reviewed. Oncology 1998; 12: 820. 55. Karlan BY, et al. Peritoneal serous papillary carcinoma, a phenotypic variant of familial ovarian cancer: implications for ovarian cancer screening. Am J Obstet Gynecol 1999; 180: 917 – 28. 56. Warhol MJ, Hunter NJ, Corson JM. An ultrastructural comparison of mesotheliomas and adenocarcinomas of the ovary and endometrium. Int J Gynecol Pathol 1982; 1: 125 – 34. 57. Eyden BP, Banik S, Harris M. Malignant epithelial mesothelioma of the peritoneum: observation on a problem case. Ultrastruct Pathol 1996; 20: 337 – 44. 58. Kannerstein M, et al. A critique of the criteria for the diagnosis of diffuse malignant mesothelioma. Mt Sinai J Med 1977; 44: 485 – 94. 59. Ascoli V, et al. Malignant mesothelioma of the tunica vaginalis testis in a young adult. J Urol Pathol 1996; 5: 75 – 83. 60. Moll UM, Valea F, Chumas J. Role of p53 alteration in primary peritoneal carcinoma. Int J Gynecol Pathol 1997; 16: 156 – 62. 61. Eltabbakh GH, et al. Prognostic factors in extra-ovarian primary peritoneal carcinoma. Gynecol Oncol 1998; 71: 230 – 9. 62. Ransom DT, et al. Papillary serous adenocarcinoma of the peritoneum: a review of 33 cases treated with platin-based chemotherapy. Cancer 1990; 66: 1091 – 4. 63. Khoury N, et al. A comparative immunohistochemical study of peritoneal and ovarian serous tumors, and mesotheliomas. Hum Pathol 1990; 21: 811 – 9. 64. Wall JE, et al. Effectiveness of paclitaxel in treating papillary serous carcinoma of the peritoneum in an adolescent. Am J Obstet Gynecol 1995; 172: 1049 – 52. 65. Shapiro SP, Nunez C. Psammoma bodies in the cervicovaginal smear in association with papillary tumor of the peritoneum. Obstet Gynecol 1983; 61: 130 – 4. 66. Rose PG, Reale FR. Papillary serous carcinoma of the peritoneum following endometrial cancer. Obstet Gynecol 1991; 78: 80. 67. Eltabbakh GH, Piver MS. Extraovarian primary peritoneal carcinoma. Oncology 1998; 12: 813 – 9. 68. Resnik E, Bender D. Syndrome of inappropriate antidiuretic hormone secretion in papillary serous surface carcinoma of the peritoneum. J Surg Oncol 1996; 61: 63 – 5. 69. Geisler JP, et al. Estrogen and progesterone receptor status as prognostic indicators in patients with optimally cytoreduced stage IIIC serous cystadenocarcinoma of the ovary. Gynecol Oncol 1996; 60: 424 – 7.
446
GYNECOLOGICAL CANCERS
70. Kommos F, et al. Steroid receptors in ovarian carcinoma: immunohistochemical determination may lead to new aspects. Gynecol Oncol 1992; 47: 317 – 22. 71. Sevelda P, et al. Estrogen and progesterone receptor content as a prognostic factor in advanced epithelial ovarian carcinoma. Br J Obstet Gynecol 1990; 97: 706 – 12. 72. Della-Fiorentina SA, et al. Primary peritoneal carcinoma: a treatable subset of patients with adenocarcinoma of unknown primary. Aust N Z J Surg 1996; 66: 124 – 5. 73. Staging Announcement FIGO Cancer Committee. Cancer committee of the international federation of gynecology and obstetrics. Gynecol Oncol 1986; 25: 383 – 5. 74. Mulhollan TJ, et al. Ovarian involvement by serous surface papillary carcinoma. Int J Gynecol Pathol 1994; 13: 120 – 6. 75. Kennedy AW, et al. Experience with platinum-paclitaxel chemotherapy in the initial management of papillary serous carcinoma of the peritoneum. Gynecol Oncol 1998; 71: 288 – 90. 76. Menzin AW, et al. Surgically documented responses to paclitaxel and cisplatin in patients with primary peritoneal carcinoma. Gynecol Oncol 1996; 62: 55 – 8. 77. Piver MS, et al. Two sequential studies for primary peritoneal carcinoma: induction with weekly cisplatin followed by either cisplatindoxorubicin-cyclophosphamide or paclitaxel-cisplatin. Gynecol Oncol 1997; 67: 141 – 6. 78. Bloss JD, et al. A phase II trial cisplatin and cyclophosphamide in the treatment of extraovarian peritoneal serous papillary carcinoma with comparison to papillary serous ovarian carcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 1998; 68: 109. 79. Halperin R, et al. Immunohistochemical comparison of primary peritoneal and primary ovarian serous papillary carcinoma. Int J Gynecol Pathol 2001; 20: 341 – 5. 80. McGuire WP, et al. Cyclophosphamide and cisplatin compared with paclitaxel plus cisplatin in patients with stage III and IV ovarian cancer. N Engl J Med 1996; 334: 1 – 6. 81. Eltabbakh GH. Extreme drug resistance assay and response to chemotherapy in patients with primary peritoneal carcinoma. J Surg Oncol 2000; 73: 148 – 52. 82. Rose PG, et al. Second-line therapy with paclitaxel and carboplatin for recurrent disease following first-line therapy with paclitaxel and platinum in ovarian or peritoneal carcinoma. J Clin Oncol 1998; 16: 1494 – 7.
83. Nicoletti MI, et al. Antitumor activity of taxol (NSC-125973) in human ovarian carcinomas growing in the peritoneal cavity of nude mice. Ann Oncol 1993; 4: 151. 84. Einzig AI, et al. Phase II study and long-term follow up of patients treated with Taxol for advanced ovarian adenocarcinoma. J Clin Oncol 1992; 10: 1748. 85. Wilailak S, et al. Peritoneal papillary serous carcinoma: response to taxol in a platinum resistant disease. Eur J Gynaecol Oncol 1995; 16: 187 – 9. 86. Eisenhauer EA, et al. European-Canadian randomized trial of paclitaxel in relapsed ovarian cancer: high-dose versus low-dose and long versus short infusion. J Clin Oncol 1994; 12: 2654. 87. Holmes FA, et al.. Current status of clinical trials with paclitaxel and docetaxel. In: George GI, et al. (eds) ACS Symposium. Taxane Anticancer Agent: Basic Science and Current Status. Washington, DC: American Cancer Society, 1995: 3 – 31. 88. Kavanagh JJ, et al. Carboplatin reinduction after taxane in patients with platinum-refractory epithelial ovarian cancer. J Clin Oncol 1995; 13: 1584. 89. Thigpen T, et al. Cisplatin as salvage therapy in ovarian carcinoma treated initially with single agent paclitaxel: a Gynecologic Oncology Group Study. Proc Am Soc Clin Oncol 1996; 15: 778. 90. Herrada J, et al. Remission with carboplatin of paclitaxel resistant primary peritoneal papillary serous carcinoma: case report. Eur J Gynaecol Oncol 1997; 18: 39 – 41. 91. Dubernard G, et al. Prognosis of stage II or IV primary peritoneal serous papillary carcinoma. Eur J Surg Oncol 2004; 30: 976 – 81. 92. Piura B, et al. Peritoneal papillary serous carcinoma: study of 15 cases and comparison with stage III-IV ovarian papillary serous carcinoma. J Surg Oncol 1998; 68(3): 173 – 8. 93. Kosary CL. FIGO stage, histologic grade, age, and race as prognostic factors in determining survival for cancers of the female gynecological system: an analysis of 197387 SEER cases of cancers of the endometrium, cervix, ovary, vulva, and vagina. Semin Surg Oncol 1994; 10: 31 – 46. 94. Petru E, et al. Primary papillary serous carcinoma of the peritoneum: a report of experiences, Geburtsh Frauenheilk 1992; 5: 533 – 5. 95. Winter R. Lymphadenectomy. In Burghardt E, et al. (eds) Surgical Gynecologic Oncology. Stuttgart, Germany: Thieme Verlag, 1993: 281.
Section 7 : Gynecological Cancers
40
Borderline Tumors and Other Rare Epithelial Tumors of the Ovary
Teresa P. D´ıaz-Montes, Russell Vang, Deborah K. Armstrong and Robert E. Bristow
In 1929, Taylor first described a subset of ovarian tumors that he termed semimalignant.1 These lesions had a more favorable outcome than other ovarian cancers, but they were not separately classified by the International Federation of Gynecology and Obstetrics (FIGO) and the World Health Organization (WHO) until the early 1970s.2,3 In 1961, the Cancer Committee of the FIGO suggested a system that subdivided the ovarian tumors into three types: benign cystadenomas, cystadenocarcinomas of low malignant potential, and cystadenocarcinomas.2 This classification became effective in 1971. The WHO applied the designation of tumor of borderline malignancy and added the synonym of carcinoma of low malignant potential in their 1973 classification.3 During the past few decades, multiple terms have been applied to these neoplasms such as tumors of borderline malignancy, carcinomas of low malignant potential, borderline tumors, tumors of low malignant potential, and atypical proliferative tumors. Today no consensus has been reached on the preferred terminology of these tumors, although the most favored term is borderline ovarian tumor. The terms borderline, atypical proliferative, and low malignant potential tumors should be considered as synonymous. During the Borderline Ovarian Tumor Workshop held at Bethesda, Maryland, in 2003, a consensus was reached as to not designate these tumors as carcinomas of low malignant potential or any other type of carcinoma.4
women.5 Borderline tumors account for approximately 15% of all epithelial ovarian tumors. On an average, the age at diagnosis is approximately 10 years younger than that of women with malignant ovarian cancer and ranges from 39 to 45 years.6 – 9 Borderline ovarian tumors share an epidemiologic risk factor profile similar to that of frankly malignant ovarian tumors, but they are associated with a significantly better prognosis and clinical outcome. In a case –control study, Harlow et al. reported that factors linked to a lower risk of (i.e. protective against) borderline ovarian tumors included the use of oral contraceptives, prior pregnancy, and lactation.10 Use of oral contraceptives has been associated with a 60% reduction in risk. However, the magnitude of the association was independent of duration of use, age at first use, or years since last usage. Compared to nulliparous women, the relative risk of developing a borderline ovarian tumor was 0.7 for women who had given birth to one or two children and 0.4 for women with three or more children. There was no consistent influence of increasing age at first live birth. Among nulliparous women, a further increase in risk was present in those who reported a history of infertility. Adjusting for parity, a history of lactation was associated with a 50% reduction in risk. Neither age at menarche, menopause, or first live birth was a significant risk factor for borderline ovarian tumor development. This study is representative of other reports and provides clear evidence that borderline ovarian tumors have a similar epidemiological and reproductive risk factor profile compared to their more malignant counterparts.
EPIDEMIOLOGY
PATHOLOGY
Borderline ovarian tumors are rare. The Surveillance Epidemiology and End Results (SEER) program indicates a United States incidence of approximately 2.5 per 105 women years.5 The incidence rates of borderline ovarian tumors were higher among white women compared with black
Serous Borderline Tumors
HISTORICAL BACKGROUND
Gross
Serous borderline tumors (SBTs) are bilateral in 23–82% of cases.11 The tumors are usually uni- or multicystic, and in
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
448
GYNECOLOGICAL CANCERS
Figure 1 Serous borderline ovarian tumor. Figure 3 Micropapillary serous borderline tumor.
most series the reported mean/median size ranges from 7 to 12 cm.12 – 17 The external surface may be smooth or demonstrate an exophytic tumor growth pattern (see Figure 1). The internal surfaces are usually lined by white to tan papillary excrescences. Histology
SBTs are subclassified into two forms – the typical type (atypical proliferative serous tumor) and the micropapillary type (noninvasive micropapillary serous carcinoma). Terms that have been used to classify these tumors but which are considered synonymous are SBT, serous tumor of low malignant potential, and atypical proliferative serous tumor.4,18 Both types of SBT lack destructive stromal invasion and contain papillary proliferations that are either intracystic and/or exophytic on the surface of the ovary. The typical SBT is composed of broad papillary structures that progressively branch in a hierarchical fashion into smaller papillae, terminating in epithelial tufts (see Figure 2). The micropapillary SBTs often arise on a background of typical SBT and contain
large papillary structures in which small and elongated papillae immediately project from the underlying large papillae without the hierarchical branching of typical SBT (the socalled “medusa appearance”). These resulting micropapillary structures lining the large papillae are generally five times longer than they are wide. Numerous detached micropapillary buds may fill the spaces between the larger papillae (see Figure 3). Since typical SBTs may have minor areas of micropapillary architecture, a 5-mm area of pure micropapillary/cribriform growth needs to be present to qualify for a diagnosis of micropapillary SBT. The cysts and papillae of SBT are lined by cuboidal or columnar serous epithelium without frankly malignant nuclei. Microinvasion consists of individual cells or nests within stroma, but the size of any individual focus cannot measure greater than 3–5 mm in greatest linear extent or 10 mm2 in overall area.11,18 Extraovarian disease principally exists in the form of implants, which are classified as noninvasive or invasive. Noninvasive implants are histologically classified as epithelial or desmoplastic types. Implants are considered as invasive when infiltration into underlying tissue is present,19 although modified criteria of (i) solid nests within clear spaces or (ii) an appearance resembling the micropapillary SBT have been used by others.20 Extraovarian disease may also be in the form of lymph node involvement by SBT.
Mucinous Borderline Tumors Gastrointestinal Type Gross The tumors generally are unilateral and have smooth external surfaces although capsular rupture may be present. The reported mean size ranges from 17 to 20 cm.21 – 23 The cut surfaces are usually multicystic and are lined by smooth surfaces that occasionally contain papillary excrescences.
Figure 2 Typical serous borderline tumor.
Histology Terms that have been used to classify these tumors but which are considered synonymous are mucinous borderline tumor of the intestinal type, mucinous tumor of low malignant potential, and atypical proliferative
BORDERLINE TUMORS AND OTHER RARE EPITHELIAL TUMORS OF THE OVARY
449
Endometrioid, Clear cell, and Brenner (Transitional Cell) Borderline Tumors Gross
Endometrioid, clear cell, and Brenner borderline tumors of the ovary are almost always unilateral. The cut surfaces are usually solid, but cystic components may be present. Histology
These tumors contain crowded glands or nests. The endometrioid and transitional cell types may show papillary architecture. The background tumor may contain adenofibroma (or benign Brenner tumor for the transitional cell type) or endometriosis (for the endometrioid or clear cell types). Intraepithelial carcinoma or microinvasion may be seen. Extraovarian implants have not been well characterized for these types or borderline tumors. Figure 4 Mucinous borderline tumor, gastrointestinal type.
mucinous tumor of gastrointestinal type.4,18,24 The tumors are composed of crowded glands and cysts that are lined by stratified mucinous epithelium with goblet cells of gastrointestinal type (see Figure 4). Destructive stromal invasion is absent. A background of mucinous cystadenoma may be evident. Intraepithelial carcinoma is diagnosed when the nuclei show marked nuclear atypia. Microinvasion consists of individual cells or nests within stroma, but the largest dimension of any individual focus is smaller than 3–5 mm in greatest linear extent or 10 mm2 in area.24 The distinction of the upper limit of glandular crowding in a mucinous borderline tumor from the confluent/expansile pattern of invasive mucinous carcinoma is based on the same size criteria for microinvasion.24 Implants are generally not seen and if encountered should cause concern for the possibility of a misclassified mucinous tumor secondarily involving the ovary. Endocervical-like Type Gross Endocervical type mucinous borderline tumors are bilateral in 13–40% of cases.25 – 28 The reported mean size ranges from 8 to 13 cm.25 – 28 The external surfaces may be smooth or contain exophytic tumor or capsular rupture. The cut surfaces may be uni- or multicystic, and the cysts are lined by papillary excrescences. Histology Terms that have been used to classify these tumors include mucinous borderline tumor of the endocervical-like or M¨ullerian type and atypical proliferative seromucinous tumor.26 – 28 The tumors lack destructive stromal invasion and have a papillary architecture as seen in SBTs. The epithelium is of mucinous columnar type without goblet cells. In addition, mixtures of serous, endometrioid, squamous, and indifferent cell types may be seen. The stroma of occasional large papillae may be edematous and contain abundant neutrophils. Associated endometriosis may be present. Microinvasion, micropapillary/cribriform types of ovarian tumors, intraepithelial carcinoma, and extraovarian implants may be encountered.
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS Borderline tumors, as with other ovarian tumors, are difficult to detect clinically until they are advanced in size or stage. Like their malignant counterparts, the most common presenting symptoms are abdominal/pelvic pain or pressure, increasing abdominal girth or abdominal distension, and perception of an abdominal mass.29 Approximately 25% of patients remain asymptomatic,30 presenting with a mass on physical or pelvic examination, at the time of surgery, or as an incidental finding on sonography. Other symptoms reported are abnormal premenstrual bleeding,29,30 torsion,31 intraperitoneal hemorrhage,30 weight loss,32 and dyspareunia.33 In most cases, the diagnosis of borderline ovarian tumor is rendered during intraoperative or postoperative pathologic evaluation. No preoperative tumor markers or radiological features can accurately identify a pelvic mass as a borderline ovarian tumor. Several studies have attempted to determine sonographic findings distinguishing borderline ovarian tumors from both benign and invasive malignant tumors. Exacoustos et al. reported that the presence of papillae into the cyst cavity from the cyst wall, was significantly more frequent in borderline tumors (48%) than it was in benign (4%) and invasive tumors (4%).34 Intracystic solid tissue was observed in 48% of invasive tumors, but in only 18% of borderline and 7% of benign tumors.34 Pascual et al. also reported that 63% of the cases evaluated had intracystic papillae.35 However, neither papillae, nor other sonographic features constituted highly sensitive sonographic markers of borderline ovarian tumors. Computed tomography (CT) and magnetic resonance imaging (MRI) have also been evaluated to determine findings that differentiate borderline ovarian tumors from their malignant counterparts. DeSouza et al.36 evaluated CT and MRI features and tumor marker levels that could differentiate borderline ovarian tumors from stage I ovarian tumors. They reported that borderline ovarian tumors were complex masses with imaging features similar to stage I invasive ovarian carcinoma, but the thickness of septations and the size of solid components were significantly larger in stage I ovarian cancer.36 However, neither feature allowed for
450
GYNECOLOGICAL CANCERS
confident differentiation of borderline ovarian tumors from stage I invasive ovarian carcinoma. CA125 is the most useful tumor marker currently available for monitoring response to therapy among patients with epithelial ovarian cancer; however, it has not proven to be a reliable screening test because of its low specificity. The use of CA125 has also been studied among patients diagnosed with borderline ovarian tumors. Chambers et al.37 obtained CA125 levels from 18 patients and found elevations above 35 U mL−1 in only four cases (22.2%). No value was noted to exceed 100 U mL−1 . Rice et al.38 found that CA125 levels were elevated in 92% of women with advanced-stage borderline ovarian tumors, but in only 40% of women with stage I. All of the patients with advanced-stage disease had serous histology, compared to only 48% of those with stage I disease, leading these authors to conclude that elevated CA125 levels correlate with advanced-stage disease in patients with serous borderline ovarian tumors.38
STAGE DISTRIBUTION AND PROGNOSIS The surgical staging of borderline ovarian tumors is done in accordance with the same FIGO guidelines used for invasive ovarian cancer. A recent review of 370 patients with borderline ovarian tumors from the Norwegian Radium Institute revealed that 84% of the patients were diagnosed with stage I disease.39 This is consistent with the review performed by Sutton of 12 published studies, which revealed that 80% of the 946 patients diagnosed with borderline ovarian tumors presented with stage I disease.40 The importance of comprehensive surgical staging in ovarian cancer also applies to borderline ovarian tumors. In a study performed by Hopkins et al., 15 patients with apparent early stage disease underwent a restaging operation and residual disease was found in seven patients.41 One patient remained as stage IA, one patient was upstaged to stage IB, and five patients were upstaged to either FIGO stage II or III disease. The authors concluded that reexploration for staging in borderline ovarian tumors would yield a significant number of positive results.41 Snider et al. also noted a 19% incidence of upstaging in 27 patients diagnosed with stage I disease.42 Since the prognosis for patients with nodal involvement is similar to those without nodal involvement, it suggests that this is a metaplastic and not a metastatic process.43,44 The prognosis depends upon the stage and histologic features of the tumor, but in general is good. A review of 2818 women with borderline ovarian tumors from the SEER program reported 5- and 10-year relative survival rates of 99 and 97% respectively for stage I, 98 and 90% respectively for stage II, 96 and 88% respectively for stage III, and 77 and 69% respectively for stage IV.45 Sherman et al. demonstrated long-term survivals for patients with borderline tumors staged as distant was similar to women with localized carcinoma.46 Given the favorable long-term survival outcome of borderline ovarian tumors, the clinical significance of surgical upstaging by reexploration for apparent early stage disease has been questioned by a number of investigators.
SURGICAL MANAGEMENT Early Stage Disease Surgery is the cornerstone of treatment for patients diagnosed with early stage borderline tumors of the ovary. Apparent stage I borderline ovarian tumors diagnosed intraoperatively by frozen section should be managed, depending on their childbearing status, with bilateral salpingo-oophorectomy, hysterectomy, and staging when childbearing is completed versus conservative management, meaning ovarian cystectomy or unilateral oophorectomy, with staging, when fertility preservation is desired. In one large series of patients with borderline ovarian tumors, Lin et al.47 reported that intraoperative frozen-section analysis was obtained in 196 (77%) cases. Of the 193 cases for which the frozen-section diagnosis was known, 117 (61%) were diagnosed correctly as borderline ovarian tumors, 52 (27%) cases were interpreted as invasive ovarian cancer, and 24 (12%) cases were thought to be benign ovarian tumors intraoperatively. Overall, 66% of the patients had at least one staging biopsy performed, and 34% had no staging biopsy. Only 12% of patients were completely staged. Among all patients, staging biopsies were positive for extraovarian disease in 37% of cases. Approximately 47% (80 of 169) of patients who underwent biopsies were upstaged as a result of positive biopsies, with 41% (70 of 169) having extrapelvic spread. For patients with apparent stage I borderline ovarian tumors who have previously undergone a surgical procedure that did not include staging, we recommend staging/reexploration by means of a laparotomy or laparoscopy only if micropapillary SBT histology is present or if there is evidence of bulky residual disease on a postoperative CT scan. If neither of the above criteria is present, reexploration for the expressed purpose of surgical staging is unlikely to have a meaningful impact on long-term clinical outcome and is not recommended. Rather, close clinical surveillance should be sufficient. For patients in whom an intraoperative diagnosis of mucinous borderline ovarian tumors is obtained, appendectomy and a thorough gastrointestinal evaluation should be performed to rule out a primary gastrointestinal tumor metastatic to the ovary. Since many women affected with borderline ovarian tumors have not completed childbearing, the efficacy and safety of conservative surgery is an important concern. There is no evidence that a conservative approach has an adverse effect on survival in patients with stage I borderline ovarian tumors.48 For this group of patients, unilateral oophorectomy or ovarian cystectomy with staging is a reasonable option. Conservative management should include visualization of the contralateral ovary, because the risk of bilateral involvement, especially in SBTs, is 40 percent.49 If there is a suspicious mass or lesion, ovarian cystectomy or wedge biopsy may be performed. Random biopsies of the contralateral ovary are not recommended if no gross abnormalities are seen because of the low yield. Several authors have reported on the safety of conservative surgery for patients with borderline ovarian tumors who desire future fertility. Although performing a unilateral oophorectomy or an ovarian cystectomy, or both, does not appear to significantly
BORDERLINE TUMORS AND OTHER RARE EPITHELIAL TUMORS OF THE OVARY
affect long-term overall survival rates, recurrence rates are higher for women who undergo conservative management. In the study of Tazelaar et al.,50 41 of 61 patients with stage IA borderline tumors were treated with total abdominal hysterectomy and bilateral salpingo-oophorectomy. The rest (20 patients) were treated with a variety of conservative procedures including cystectomy with (1 patient) and without (3 patients) a contralateral ovarian wedge biopsy, and unilateral salpingo-oophorectomy with (6 patients) and without (10 patients) a contralateral wedge biopsy. After a mean followup time of 89 months, subsequent borderline tumors had developed in three patients (15%) initially treated conservatively and in two patients (5%) initially treated with abdominal hysterectomy and bilateral salpingo-oophorectomy. In the report by Lim-Tan et al.,51 35 patients with serous borderline ovarian tumors underwent unilateral ovarian cystectomy, bilateral cystectomy, or unilateral cystectomy with contralateral oophorectomy or salpingo-oophorectomy. All but two of the patients had stage I disease. Tumor persisted or recurred only in the ovary that had been subjected to cystectomy in 2 (6%) of 33 patients with stage I tumors, in both the ipsilateral and contralateral ovary in one patient (3%), and in the contralateral ovary in the other (3%). They concluded that involvement of the resection margin of the cystectomy specimen and the removal of more than one cyst from an ovary was almost always associated with persistence or recurrence of tumor. Because of the increased risk of recurrence, ovarian cystectomy in the management of borderline ovarian tumors is not widely embraced and must be limited to exceptional cases (e.g. bilateral lesions). Morris et al. reported that 42% of women who underwent conservative surgery for serous borderline ovarian tumors developed a recurrence.52 In view of these findings, women who undergo conservative management should be closely monitored for disease recurrence, especially if they are left with only one remaining ovary.
Advanced-stage and Recurrent Disease Approximately 20% of patients with borderline ovarian tumors will present with advanced-stage disease at the time of diagnosis. The recommended primary surgical management of these patients is identical to that applied to patients with invasive ovarian cancer. Rates of recurrence and death for women with stages II and III disease vary from 5 to 30%.12 – 16,19,20,53,54 However, even with advancedstage borderline ovarian tumors, the long-term survival rate approaches 70%.45 As with invasive ovarian cancer, maximal surgical cytoreduction to no gross residual disease is therefore recommended. The role of conservative surgical management, meaning unilateral oophorectomy or ovarian cystectomy, for advanced-stage disease is less well accepted than for early stage borderline tumors of the ovary. Zanetta et al. demonstrated that 40% of the patients treated with conservative management for advanced-stage disease had a recurrence compared with 12.9% of recurrences after nonconservative management.54 Given these observations, conservative management (e.g. unilateral oophorectomy or ovarian cystectomy) for advanced-stage disease should be undertaken with caution.
451
In contrast to invasive ovarian cancer, the median time to recurrence for borderline tumors is 5 to 7 years. For patients who develop recurrent disease, secondary cytoreductive surgery is the treatment of choice. Crispens et al.,55 evaluated the outcome of 53 patients treated for progressive or recurrent serous borderline ovarian tumors and emphasized the importance of surgery in the management of these tumors. These authors found that patients with progressive or recurrent disease who could be optimally cytoreduced to residual disease less than or equal to 2 cm had a significantly better survival compared with patients who could not be optimally cytoreduced. The response to chemotherapy, hormonal therapy, and radiotherapy was universally poor. Among 45 patients who received nonsurgical therapy, only 6 patients (13%) had partial responses and 6 patients (13%) had complete responses. Twenty-one out of 45 patients (47%) had stabilization of disease with nonsurgical therapy. It should be noted that the clinical significance of an objective response to adjuvant therapy is unclear, given the indolent clinical course of borderline ovarian tumors.
ADJUVANT THERAPY For patients with disease apparently confined to the ovaries, adjuvant chemotherapy is not recommended. Barnhill et al.48 reported a Gynecologic Oncology Group (GOG) prospective study in which 146 patients with stage I serous borderline ovarian tumors were observed without adjuvant therapy. With a median follow-up of 42.2 months, no patient developed recurrent disease. For most patients with stage I tumors, long-term disease-free survival can be expected. To underscore this point, a large meta-analysis demonstrated a disease-free survival rate of 98.2% and a disease-specific survival rate of 99.5% for women with stage I disease.53 Clinical significance notwithstanding, objective responses to platinum-based chemotherapy among patients with advanced-stage borderline ovarian tumors have been reported at the time of second-look surgery. Gershenson et al. reported complete responses to chemotherapy at secondlook laparotomy in 8 of 20 patients with macroscopic residual disease after initial cytoreductive surgery and in 5 of 12 patients with microscopic residual disease after initial surgery.56 Barakat et al. reported that two of seven patients with macroscopic residual borderline ovarian tumors and seven of eight patients with microscopic disease had pathologic complete remissions at second-look laparotomy after platinum-based chemotherapy.57 With a mean followup of 64 months, only one patient had died of progressive disease. Importantly, there was no difference in survival rate between patients who received chemotherapy and those who did not. Sutton et al.32 reported the GOG data using a subset of 32 women with advanced-stage borderline ovarian tumors that were optimally cytoreduced. The patients were randomized to treatment with cisplatin and cyclophosphamide with or without adriamycin. Fifteen of 32 patients underwent second-look surgery, and nine showed evidence of persistent disease. However, at a median followup of 31.7 months, 31 of 32 patients were alive. Only one patient died and it was unrelated to the ovarian disease
452
GYNECOLOGICAL CANCERS
process. Because of the low percentage of actively dividing cells that are present in borderline ovarian tumors, these are thought to be relatively resistant to standard cytotoxic agents. As noted earlier, even patients with advanced-stage disease can be expected to have excellent overall survival rates. Therefore, patients must be counseled that the role of adjuvant chemotherapy in advanced-stage disease is still unclear. Patients should be managed with surgical debulking with the objective of removing all visible disease. Adjuvant chemotherapy is usually reserved for patients with invasive extraovarian implants, although a meaningful survival benefit has been difficult to demonstrate even for this select group of patients. In patients with noninvasive implants, surgical debulking should be sufficient and chemotherapy is not recommended. The management of recurrent disease must be individualized. The decision to treat must be carefully considered, balancing the patient’s symptoms, tumor growth rate, extent of disease, and overall life goals. There are no randomized or prospective studies evaluating the benefit of chemotherapy in patients with recurrent borderline ovarian tumors. As demonstrated by Crispens et al.,55 patients with progressive or recurrent disease who could be optimally cytoreduced to less than or equal to 2 cm maximal residual disease had a significantly better survival compared with patients who could not be optimally cytoreduced. The responses to chemotherapy, hormonal therapy, and radiotherapy were poor. These findings did not demonstrate a significant response to nonsurgical therapy among patients with persistent or recurrent disease. In this setting, surgery should be the preferred management option for patients diagnosed with recurrent borderline ovarian tumors. Chemotherapy may be indicated if unresectable disease is present, or for tumors demonstrating a more rapid growth rate with progressive symptomatology. In summary, the recommended management of clinically apparent early stage borderline ovarian tumors includes bilateral salpingo-oophorectomy with hysterectomy and surgical staging for women who have completed childbearing. For young patients with apparent early stage disease who desire fertility preservation, unilateral oophorectomy or ovarian cystectomy with staging procedures is an acceptable alternative, although this approach may predispose to a higher risk of recurrence. For advanced-stage and recurrent disease, cytoreductive surgery is the cornerstone treatment, while adjuvant chemotherapy is reserved for select cases only (e.g. unresectable disease, invasive metastatic implants, rapid growth rate with progressive symptomatology).
OTHER RARE EPITHELIAL TUMORS OF THE OVARY Clear cell Ovarian Carcinoma Clear cell ovarian carcinomas comprise approximately 3% of ovarian epithelial neoplasm.58 The mean age at diagnosis is 55 years.59 Patients usually present with symptoms related to a pelvic or abdominal mass. Clear cell carcinomas are the most common epithelial ovarian neoplasm to be associated with paraneoplastic hypercalcemia. About half of the cases are associated with endometriosis.60 Approximately
half of the patients present with stage I and 15% with stage II disease.61 Tumors usually range up to 30 cm in diameter; with a mean of about 15 cm.58 There is conflicting data on the behavior of these tumors. In some studies, the prognosis appears similar to that of other ovarian carcinomas,62,63 while in others, the prognosis is said to be worse.59,61,64,65 However, when controlled by stage, the survival of patients with clear cell carcinoma may be slightly lower than that of patients with serous carcinoma. The treatment of clear cell carcinoma is similar to that of other epithelial cell types of ovarian cancer. Recio et al.66 reported that women with clear cell carcinomas treated with platinumbased chemotherapy were at significantly increased risk of thromboembolic complications compared to those with non–clear cell carcinomas, with a corresponding negative impact on survival.
Mucinous Ovarian Carcinoma After exclusion of metastatic tumors of the ovaries, primary ovarian mucinous carcinomas are uncommon. They comprise approximately 4% of ovarian epithelial neoplasm.58 The mean age at diagnosis is 53 years.58 As these tumors can be quite large, the clinical presentation is generally that of a large pelvic or abdominal mass and abdominal distension. Approximately 63% of cases are stage I disease.61 The overall prognosis is better than that for serous carcinoma of the ovary mostly because of the early stage of presentation, however when stratified by stage the behavior is similar to that of serous ovarian carcinoma.61,67 – 69 Advanced-stage mucinous carcinoma is uniformly fatal. The treatment of mucinous carcinoma is similar to that of other types of epithelial cell ovarian cancer.
Endometrioid Ovarian Carcinoma Endometrioid ovarian carcinomas comprise 6% of ovarian surface epithelial tumors.58 The mean patients’ age at diagnosis is 56 years.58 The most common presenting symptoms are abdominal distention and abdominal or pelvic pain. Abnormal vaginal bleeding could be a frequent symptom and could be related to the association of endometrioid ovarian carcinoma with endometrial hyperplasia and carcinoma.70 The tumor size usually ranges from 12 to 20 cm with a mean of about 15 cm.58 A high proportion of endometrioid carcinomas of the ovary are diagnosed at an early stage with 52% of cases presenting with stage I or II disease.61 The reported association of endometrioid carcinomas with ovarian endometriosis is around 10%, although in a well-documented study of stage I cases, 40% were associated with endometriosis, one-third of which arose in the endometriosis.71 It has been stated that these tumors have a better prognosis than serous ovarian carcinomas but this is mostly related to the fact that a great majority of the patients present with early stage disease. Treatment of endometrioid carcinomas is generally the same as that of other ovarian epithelial carcinomas.
Malignant Mixed Mesodermal Tumor (Carcinosarcoma) Malignant mixed mesodermal tumors (MMMTs) of the ovary comprise less than 1% of ovarian neoplasm.58 The mean age
BORDERLINE TUMORS AND OTHER RARE EPITHELIAL TUMORS OF THE OVARY
of patients with this tumor is about 60 years.58 Approximately 74% of patients present with advanced-stage disease.70 The tumors are typically large, ranging from 15 to 20 cm in diameter.58 MMMTs are aggressive, and rapidly fatal tumors with a median survival of approximately one year.72 – 74 Brown et al. compared 65 patients with carcinosarcoma to 746 patients with serous adenocarcinoma.75 The objective response to platinum-based chemotherapy (25 vs 60%), progression-free survival (6.4 vs 12.1 months), and median survival (8.2 vs 20.7 months) were all statistically worse with carcinosarcoma. Sit et al. reported platinum combinations with either paclitaxel or ifosfamide with median survivals of 19 and 23 months, respectively.76
18.
19.
20.
21.
22.
REFERENCES 1. Taylor HC. Malignant and semimalignant tumors of the ovary. Surg Gynecol Obstet 1929; 48: 204 – 30. 2. International Federation of Gynecology and Obstetrics. Classification and staging of malignant tumors in the female pelvis. Acta Obstet Gynecol Scand 1971; 50: 1 – 7. 3. Serov SF, Scully RE, Sobin LH. International Histological Classification and Staging of Tumors. No. 9 Histologic Typing of Ovarian Tumors. Geneva: World Health Organization, 1973: 37 – 41. 4. Silverberg SG, et al. Borderline ovarian tumor workshop. Borderline ovarian tumors: key points and workshop summary. Hum Pathol 2004; 35: 910 – 7. 5. Mink P, Sherman ME, Devesa S. Incidence patterns of invasive and borderline ovarian tumors among white women and black women in the United States: results from the SEER program, 1978 – 1997. Cancer 2002; 95: 2380 – 9. 6. Nakashima N, et al. Ovarian epithelial tumors of borderline malignancy in Japan. Gynecol Oncol 1990; 38: 90 – 8. 7. Barnhill D, et al. Epithelial ovarian carcinoma of low malignant potential. Obstet Gynecol 1985; 65: 53 – 9. 8. Bostwick DG, et al. Ovarian epithelial tumors of borderline malignancy. A clinical and pathologic study of 109 cases. Cancer 1986; 58: 2052 – 65. 9. Katsube Y, Berg JW, Silverberg SG. Epidemiologic pathology of ovarian tumors: a histopathologic review of primary ovarian neoplasms diagnosed in the Denver Standard Metropolitan Statistical Area, 1 July31 December 1969 and 1 July-31 December 1979. Int J Gynecol Pathol 1982; 1(1): 3 – 16. 10. Harlow BL, et al. Case-control study of borderline ovarian tumors: reproductive history and exposure to exogenous female hormones. Cancer Res 1988; 48: 5849 – 52. 11. Bell DA, et al. Serous borderline (low malignant potential, atypical proliferative) ovarian tumors: workshop perspectives. Hum Pathol 2004; 35: 934 – 48. 12. Deavers MT, et al. Micropapillary and cribriform patterns in ovarian serous tumors of low malignant potential: a study of 99 advanced stage cases. Am J Surg Pathol 2002; 26: 1129 – 41. 13. Eichhorn JH, et al. Ovarian serous borderline tumors with micropapillary and cribriform patterns: a study of 40 cases and comparison with 44 cases without these patterns. Am J Surg Pathol 1999; 23: 397 – 409. 14. Goldstein NS, Ceniza N. Ovarian micropapillary serous borderline tumors. Clinicopathologic features and outcome of seven surgically staged patients. Am J Clin Pathol 2000; 114: 380 – 6. 15. Prat J, De Nictolis M. Serous borderline tumors of the ovary: a longterm follow-up study of 137 cases, including 18 with a micropapillary pattern and 20 with microinvasion. Am J Surg Pathol 2002; 26: 1111 – 28. 16. Seidman JD, Kurman RJ. Subclassification of serous borderline tumors of the ovary into benign and malignant types. A clinicopathologic study of 65 advanced stage cases. Am J Surg Pathol 1996; 20: 1331 – 45. 17. Smith Sehdev AE, Sehdev PS, Kurman RJ. Noninvasive and invasive micropapillary (low-grade) serous carcinoma of the ovary: a
23.
24.
25.
26.
27.
28.
29.
30.
31. 32.
33. 34. 35.
36.
37. 38. 39.
453
clinicopathologic analysis of 135 cases. Am J Surg Pathol 2003; 27: 725 – 36. Seidman JD, et al. Borderline ovarian tumors: diverse contemporary viewpoints on terminology and diagnostic criteria with illustrative images. Hum Pathol 2004; 35: 918 – 33. Bell DA, Weinstock MA, Scully RE. Peritoneal implants of ovarian serous borderline tumors. Histologic features and prognosis. Cancer 1988; 62: 2212 – 22. Bell KA, Smith Sehdev AE, Kurman RJ. Refined diagnostic criteria for implants associated with ovarian atypical proliferative serous tumors (borderline) and micropapillary serous carcinomas. Am J Surg Pathol 2001; 25: 419 – 32. Lee KR, Scully RE. Mucinous tumors of the ovary: a clinicopathologic study of 196 borderline tumors (of intestinal type) and carcinomas, including an evaluation of 11 cases with ’pseudomyxoma peritonei’. Am J Surg Pathol 2000; 24: 1447 – 64. Riopel MA, Ronnett BM, Kurman RJ. Evaluation of diagnostic criteria and behavior of ovarian intestinal-type mucinous tumors: atypical proliferative (borderline) tumors and intraepithelial, microinvasive, invasive, and metastatic carcinomas. Am J Surg Pathol 1999; 23: 617 – 35. Rodriguez IM, Prat J. Mucinous tumors of the ovary: a clinicopathologic analysis of 75 borderline tumors (of intestinal type) and carcinomas. Am J Surg Pathol 2002; 26: 139 – 52. Ronnett BM, et al. Mucinous borderline ovarian tumors: points of general agreement and persistent controversies regarding nomenclature, diagnostic criteria, and behavior. Hum Pathol 2004; 35: 949 – 60. Dube V, et al. Mucinous ovarian tumors of mullerian-type: an analysis of 17 cases including borderline tumors and intraepithelial, microinvasive, and invasive carcinomas. Int J Gynecol Pathol 2005; 24: 138 – 46. Rodriguez IM, Irving JA, Prat J. Endocervical-like mucinous borderline tumors of the ovary: a clinicopathologic analysis of 31 cases. Am J Surg Pathol 2004; 28: 1311 – 8. Rutgers JL, Scully RE. Ovarian mullerian mucinous papillary cystadenomas of borderline malignancy. A clinicopathologic analysis. Cancer 1988; 61: 340 – 8. Shappell HW, et al. Diagnostic criteria and behavior of ovarian seromucinous (endocervical-type mucinous and mixed cell-type) tumors: atypical proliferative (borderline) tumors, intraepithelial, microinvasive, and invasive carcinomas. Am J Surg Pathol 2002; 26: 1529 – 41. Hopkins MP, Kumar NB, Morley GW. An assessment of pathologic features and treatment modalities in ovarian tumors of low malignant potential. Obstet Gynecol 1987; 70: 923 – 9. Hart WR, Norris HJ. Borderline and malignant mucinous tumors of the ovary. Histologic criteria and clinical behavior. Cancer 1973; 31: 1031 – 45. Julian CG, Woodruff JD. The biologic behavior of low-grade papillary serous carcinoma of the ovary. Obstet Gynecol 1972; 40: 860 – 7. Sutton GP, et al. Stage III ovarian tumors of low malignant potential treated with cisplatin combination therapy (a Gynecologic Oncology Group study). Gynecol Oncol 1991; 41: 230 – 3. Yazigi R, Sandstad J, Munoz AK. Primary staging in ovarian tumors of low malignant potential. Gynecol Oncol 1988; 31: 402 – 8. Exacoustos C, et al. Preoperative sonographic features of borderline ovarian tumors. Ultrasound Obstet Gynecol 2005; 25(1): 50 – 9. Pascual MA, et al. Borderline cystic tumors of the ovary: gray-scale and color Doppler sonographic findings. J Clin Ultrasound 2002; 30(2): 76 – 82. DeSouza NM, et al. Borderline tumors of the ovary: CT and MRI features and tumor markers in differentiation from stage I disease. AJR Am J Roentgenol 2005; 184(3): 999 – 1003. Chambers JT, et al. Borderline ovarian tumors. Am J Obstet Gynecol 1988; 159(5): 1088 – 94. Rice LW, et al. Preoperative serum CA-125 levels in borderline tumors of the ovary. Gynecol Oncol 1992; 46(2): 226 – 9. Kaern J, Trope CG, Abeler VM. A retrospective study of 370 borderline tumors of the ovary treated at the Norwegian Radium Hospital from 1970 to 1982. A review of clinicopathologic features and treatment modalities. Cancer 1993; 71(5): 1810 – 20.
454
GYNECOLOGICAL CANCERS
40. Sutton GP. Ovarian tumors of low malignant potential. In Rubin SC, Sutton GP (eds) Ovarian Cancer. New York: McGraw-Hill, 1993: 425 – 449. 41. Hopkins MP, Morley GW. The second-look operation and surgical reexploration in ovarian tumor of low malignant potential. Obstet Gynecol 1989; 74: 375 – 8. 42. Snider DD, et al. Evaluation of surgical staging in stage I low malignant potential ovarian tumors. Gynecol Oncol 1991; 40: 129 – 32. 43. Shiraki M, et al. Ovarian serous borderline epithelial tumors with multiple retroperitoneal nodal involvement: metastasis or malignant transformation of epithelial glandular inclusions? Gynecol Oncol 1992; 46: 255 – 8. 44. Camatte S, et al. Lymph node disorders and prognostic value of nodal involvement in patients treated for a borderline ovarian tumor: an analysis of a series of 42 lymphadenectomies. J Am Coll Surg 2002; 195: 332 – 8. 45. Trimble CL, Kosary C, Trimble EL. Long-term survival and patterns of care in women with ovarian tumors of low malignant potential. Gynecol Oncol 2002; 86: 34 – 7. 46. Sherman ME, et al. Survival among women with borderline ovarian tumors and ovarian carcinoma: a population-based analysis. Cancer 2004; 100: 1045 – 52. 47. Lin PS, et al. The current status of surgical staging of ovarian serous borderline tumors. Cancer 1999; 85: 905 – 11. 48. Barnhill DR, et al. Preliminary analysis of the behavior of stage I ovarian serous tumors of low malignant potential: A Gynecologic Oncology Group study. J Clin Oncol 1995; 13: 2752 – 6. 49. Segal GH, Hart WR. Ovarian serous tumors of low malignant potential (serous borderline tumors): the relationship of exophytic surface tumor to peritoneal implants. Am J Surg Pathol 1992; 16: 577 – 83. 50. Tazelaar HD, et al. Conservative treatment of borderline ovarian tumors. Obstet Gynecol 1985; 66: 417 – 22. 51. Lim-Tan SK, Cajigas HE, Scully RE. Ovarian cystectomy for serous borderline tumors: A follow-up study of 35 cases. Obstet Gynecol 1988; 72: 775 – 80. 52. Morris RT, et al. Outcome and reproductive function after conservative surgery for borderline ovarian tumors. Obstet Gynecol 2000; 95: 541 – 7. 53. Seidman JD, Kurman RJ. Ovarian serous borderline tumors: A critical review of the literature with emphasis on prognostic indicators. Hum Pathol 2000; 31: 539 – 57. 54. Zanetta G, et al. Behavior of borderline tumors with particular interest to persistence, recurrence, and progression to invasive carcinoma: A prospective study. J Clin Oncol 2001; 19: 2658 – 64. 55. Crispens MA, et al. Response and survival in patients with progressive or recurrent serous ovarian tumors of low malignant potential. Obstet Gynecol 2002; 99: 3 – 10. 56. Gershenson DM, Silva EG. Serous ovarian tumors of low malignant potential with peritoneal implants. Cancer 1990; 65: 578 – 84. 57. Barakat RR, et al. Platinum-based chemotherapy for advanced-stage serous ovarian carcinoma of low malignant potential. Gynecol Oncol 1995; 59: 390 – 3. 58. Seidman JD, Russell P, Kurman RJ. Surface epithelial tumors of the ovary. In Kurman RJ (ed) Blaunstein’s Pathology of the Female Genital Tract. New York: Springer-Verlag; 2002: 810 – 904.
59. O’Brien ME, et al. Clear cell epithelial ovarian cancer (mesonephroid): bad prognosis only in early stages. Gynecol Oncol 1993; 49: 250 – 4. 60. Behbakht K, et al. Clinical characteristics of clear cell carcinoma of the ovary. Gynecol Oncol 1998; 70: 255 – 8. 61. Pecorelli S. FIGO annual report on the results of treatment in gynaecological cancer. J Epidemiol Biostat 1998; 23(3): 1 – 168. 62. Crozier MA, et al. Clear cell carcinoma of the ovary: a study of 59 cases. Gynecol Oncol 1989; 35: 199 – 203. 63. Jenison EL, et al. Clear cell adenocarcinoma of the ovary: a clinical analysis and comparison with serous carcinoma. Gynecol Oncol 1989; 32: 65 – 71. 64. Kennedy AW, et al. Survival probability in ovarian clear cell adenocarcinoma. Gynecol Oncol 1999; 74: 108 – 14. 65. Tammela J, et al. Clear cell carcinoma of the ovary: poor prognosis compared to serous carcinoma. Eur J Gynaecol Oncol 1998; 19: 438 – 40. 66. Recio FO, et al. Lack of improved survival plus increase in thromboembolic complications in patients with clear cell carcinoma of the ovary treated with platinum versus nonplatinum-based chemotherapy. Cancer 1996; 78: 2157 – 63. 67. Chaitin BA, Gershenson DM, Evans HL. Mucinous tumors of the ovary: a clinicopathologic study of 70 cases. Cancer 1985; 55: 1958 – 62. 68. Hoerl HD, Hart WR. Primary ovarian mucinous cystadenocarcinomas: a clinicopathologic study of 49 cases with long-term follow-up. Am J Surg Pathol 1998; 22: 1449 – 62. 69. Kikkawa F, et al. Mucinous carcinoma of the ovary: clinicopathologic analysis. Oncology 1996; 53: 303 – 7. 70. Le T, et al. Malignant mixed mesodermal ovarian tumor treatment and prognosis: a 20-year experience. Gynecol Oncol 1997; 65: 237 – 40. 71. Sainz de la Cuesta R, et al. Histologic transformation of benign endometriosis to early epithelial ovarian cancer. Gynecol Oncol 1996; 60: 238 – 44. 72. Andersen WA, et al. Platinum-based combination chemotherapy for malignant mixed mesodermal tumors of the ovary. Gynecol Oncol 1989; 32: 319 – 22. 73. Boucher D, Tetu B. Morphologic prognostic factors of malignant mixed mullerian tumors of the ovary: a clinicopathologic study of 15 cases. Int J Gynecol Pathol 1994; 13: 22 – 8. 74. Dehner LP, Norris HJ, Taylor HB. Carcinosarcomas and mixed mesodermal tumors of the ovary. Cancer 1971; 27: 207 – 16. 75. Brown E, et al. Carcinosarcoma of the ovary: 19 years of prospective data from a single center. Cancer 2004; 100(10): 2148 – 53. 76. Sit AS, et al. Chemotherapy for malignant mixed mullerian tumors of the ovary. Gynecol Oncol 2000; 79(2): 196 – 200.
FURTHER READING Czernobilsky B, Silverman BB, Mikuta JJ. Endometrioid carcinoma of the ovary: a clinico-pathologic study of 75 cases. Cancer 1970; 26: 1141 – 52.
Section 7 : Gynecological Cancers
41
Stromal Tumors of the Ovary
Jubilee Brown, Anuja Jhingran, Michael Deavers and Maurie Markman
INTRODUCTION Stromal tumors of the ovary represent 7% of all ovarian malignancies. This encompasses a diverse group of malignancies that vary in clinical presentation, natural history, prognosis, and recommended treatment. This chapter discusses the details of these rare tumors, provides a general treatment schema for such tumors, reviews details of pathologic features, and provides a detailed discussion of surgical and nonsurgical treatment recommendations for patients with each specific tumor type. In addition, the issue of conservative management for preservation of fertility is discussed.
OVERVIEW Historical Background Sertoli-stromal cell tumors, or androblastomas, were originally described in 1931 by Meyer, who theorized that they arose from the male blastema and therefore utilized the term “arrheno-blastoma” (from Greek arrhenos, male).1 Terminology was changed to “Sertoli-Leydig cell tumor” in 1958 based on the recommendation of Morris and Scully, who suggested that the prior term implied masculinization, which is often absent, and was more consistent with the general classification of sex cord-stromal tumors.2 Sex cord tumor with annular tubules (SCTAT) was first described in 1970 by Scully as a tumor associated with PeutzJeghers syndrome.3 Since that time, the behavior of this tumor has become better characterized, and it is now known that approximately 15% of these tumors are associated with adenoma malignum of the cervix.4 Steroid cell tumors, one of the more rare subtypes of stromal ovarian tumor, were historically designated as lipid cell tumors. However, it was determined that a substantial percentage of these tumors had no fatty component, so the name was changed to steroid cell tumors.5 In total, stromal tumors of the ovary represent a small portion of ovarian cancers, and an even smaller portion of the overall world cancer burden. However, for those women with this anomaly, many of whom are in their reproductive years, successful fertility-sparing treatment is crucial. Therefore,
research has continued into the science and best management of these tumors, with continuing change and progress in the field. This chapter will present the current data with historical perspective to understand the optimal treatment for patients with stromal tumors of the ovary.
Biology and Epidemiology The predicted incidence for all new cases of ovarian cancer in the United States is estimated to be 22 000 for 2005.6 Ninety percent of these malignancies are epithelial in origin, with the remaining 10% comprised of sex cord-stromal tumors, germ cell tumors, soft tissue tumors not specific to the ovary, unclassified tumors, and metastatic tumors.7 However, these data do not specify the exact numbers for stromal tumors of the ovary. Likewise, estimates from the Surveillance, Epidemiology and End Results (SEER) database between 1975 and 1998 suggest that for each 5-year interval between ages 15 and 40, the incidence of nongerm cell ovarian malignancy increases from 8 per million to 79 per million women per year.8 However, these data are also nonspecific for stromal ovarian tumors. In general, it has been estimated that stromal tumors of the ovary account for 7% of all ovarian malignancies,9 and many of these tumors occur in adolescent and young women.
Anatomy and Pathology Specialized gonadal stromal cells and their precursors can give rise to sex cord-stromal tumors of the ovary. They arise as masses in the pelvis, originating within one or both ovaries. These tumors can occur as an isolated histologic subtype or in combination. The classification is presented in Table 1.7 Specifically, granulosa cells and Sertoli cells arise from sex cord cells, while theca cells, Leydig cells, lipid cells, and fibroblasts arise from stromal cells and their pluripotential mesenchymal precursors. These cells are involved in the production of steroid hormones, and therefore physical manifestations of excess estrogen or androgen production are not infrequent at the time of diagnosis.10 Specific considerations with regard to pathology of certain tumor types are noted in the following subsections of this chapter.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
456
GYNECOLOGICAL CANCERS
Table 1 Classification of Stromal Tumors of the Ovary.7
1. Granulosa stromal cell tumors a. Granulosa cell tumors i. Juvenile ii. Adult b. Thecomas/fibromas i. Thecoma 1. Typical 2. Luteinized ii. Fibroma c. Cellular fibroma d. Fibrosarcoma e. Stromal tumor with minor sex cord elements f. Sclerosing stromal tumor g. Stromal luteoma h. Unclassified (fibrothecoma) 2. Sertoli-stromal cell tumors; androblastomas a. Well differentiated i. Sertoli cell tumor; tubular androblastoma ii. Sertoli-Leydig cell tumor iii. Leydig cell tumor b. Intermediate differentiation i. Variant – with heterologous elements c. Poorly differentiated (sarcomatoid) i. Variant – with heterologous elements d. Retiform e. Mixed 3. SCTAT 4. Gynandroblastoma 5. Steroid (lipid) cell tumor a. Stromal luteoma b. Leydig cell tumor c. Unclassified SCTAT, sex cord tumor with annular tubules.
Clinical Presentation and Diagnostic Considerations The diagnosis of a stromal tumor of the ovary is based on a thorough historical and physical examination and appropriate imaging techniques. The first hint may be the age of the patient in her adolescent or young adult years. Other presenting signs and symptoms are typical for patients with a pelvic mass, with bloating, pelvic pressure or pain, increase in abdominal girth, and gastrointestinal or urinary symptoms. The physical examination, including a pelvic and rectovaginal examination, usually suggests a pelvic mass. In some patients, especially those with granulosa cell tumors, evidence of hemoperitoneum can be present, with abdominal pain and tenderness, peritoneal signs, a fluid wave, and even hemodynamic instability.11 As noted, since stromal tumors of the ovary arise from steroid-producing cells, these tumors are often hormonally active, producing estrogen, progesterone, and androgens. Therefore, physical manifestations of excess estrogens or androgen production can be the presenting symptoms or signs of a stromal tumor.10 If this is the case, patients may report hirsutism or virilism, or if adolescents, they may describe isosexual precocious puberty. For patients during the reproductive years, presenting signs and symptoms related to hormonal changes include menorrhagia, irregular menstrual bleeding, and amenorrhea. Postmenopausal patients may note vaginal bleeding, breast enlargement or tenderness, and vaginal cornification.11
During the diagnostic and/or preoperative evaluation, abnormal uterine bleeding should prompt consideration for an endometrial biopsy. Of course, in women of reproductive age, pregnancy must first be excluded. Since endometrial hyperplasia can be a secondary effect of excess estrogen production by the ovarian stromal tumor, the endometrium must be evaluated. If this is not done in the office preoperatively, it must be done in the operating room upon the diagnosis of the ovarian stromal tumor.12 Imaging tests which may prove useful in the diagnosis of the adnexal mass include transvaginal ultrasound, computerized tomography, and magnetic resonance imaging. Of these, the ultrasound is often the best for distinguishing the details of pelvic anatomy. The findings may also identify hemoperitoneum or ascites. If hemoperitoneum is present, surgery should proceed immediately through a laparotomy, as is discussed below. Preoperative laboratory tests which may be helpful include inhibin A and B and CA125, in addition to the routine preoperative laboratory testing.13,14
Treatment The treatment of stromal ovarian tumors is determined by many factors, including patient age, parity, desire for future fertility, extent of disease, and comorbid conditions. The surgeon may be faced with a patient with an adnexal mass, the precise histologic classification of which is difficult to determine, even with the pathologic evaluation of frozen tissue sections. The surgeon must then follow general guidelines for non-epithelial ovarian tumors during the initial operative management and reevaluate the need for adjuvant or additional therapy on the basis of the final pathologic results. With close attention to all details, including histologic type, patient characteristics, and extent of disease, the need for reexploration and more extensive surgery can be minimized. General Treatment Guidelines – Surgical Therapy
When a pelvic mass is first diagnosed, the specific histologic diagnosis is unknown. However, using patient characteristics including age, physical diagnosis, and imaging characteristics as noted above, a stromal tumor of the ovary can be suspected. A frank discussion should always be held preoperatively with any woman of childbearing age who has an adnexal mass regarding her wishes for future fertility and her desires for maintaining ovarian and/or uterine function in light of the potential operative findings. Although this is often a difficult conversation for the physician to initiate, it is better discussed preoperatively with the patient than intraoperatively with the next of kin when a malignancy is encountered.15 Laparoscopy is appropriate in the occasional patient with a small solid adnexal mass or complex ovarian cyst.16,17 However, any patient with a large, solid adnexal mass or evidence of hemodynamic instability should undergo laparotomy through a vertical skin incision to allow for appropriate surgical staging, if necessary.18 Upon initial inspection, gross characteristics can suggest the diagnosis. A large, unilateral, solid adnexal mass, often yellow and multilobulated in
STROMAL TUMORS OF THE OVARY
appearance, or hemorrhagic with hemoperitoneum evident, can suggest a granulosa cell tumor or other sex cord-stromal tumor. Upon entering the peritoneal cavity, the surgeon should obtain pelvic washings and evacuate the hemoperitoneum, if present. The site of hemorrhage is most commonly the mass itself, and therefore surgical removal may stop the bleeding. A unilateral mass in a patient of any age should be removed by unilateral salpingo-oophorectomy and sent for immediate histologic evaluation.15,18 Cystectomy is not appropriate in this case. Also, the mass should not be ruptured or morcellated, as this results in the disease being classified as a more advanced stage and may adversely affect survival.19 Therefore, the tumor should not be morcellated to effect laparoscopic removal. In cases in which laparoscopy is used initially, a Cook’s bag with an extended incision should be utilized or the procedure should be converted to a laparotomy to avoid morcellating the tumor mass. Occasionally, to remove what is thought to be a benign dermoid cyst, an ovarian cystectomy is performed in an attempt to preserve ovarian tissue. In these cases, the tumor should be sent for immediate histologic evaluation, and in the event of a sex cord-stromal tumor, the entire ovary should be removed.15,18 No support exists in the literature for ovarian cystectomy in premenopausal patients with sex cord-stromal tumors. Articles that summarize “conservative management” of these tumors invariably describe unilateral salpingooophorectomy with conservation of the normal contralateral ovary in patients with limited disease. Therefore, unilateral salpingo-oophorectomy is the initial step in the treatment of patients with apparent limited disease.15,18 Once the diagnosis of a sex cord-stromal tumor is made, the entire abdominopelvic cavity should be explored, with attention paid to all peritoneal surfaces and abdominopelvic organs. A complete staging procedure should be performed, including cytologic evaluation of each hemidiaphragm, infracolic omentectomy, and peritoneal biopsies from each paracolic gutter, the vesicouterine fold, and the pouch of Douglas. Additionally, biopsies of any suspicious areas should be performed. Pelvic and para-aortic lymph node sampling are recommended for full staging, as a small percentage of patients with apparent early-stage disease will have positive lymph nodes on final pathologic review, thereby changing the stage, recommended treatment, and prognosis. The bowel should be inspected from the ileocecal valve to the ligament of Treitz, with specific evaluation for tumor implants and sites of obstruction. Tumor-reductive surgery should be performed in patients with advanced disease to reduce the tumor burden as much as possible, preferably leaving the patient with no macroscopic disease.15,18 Patients who have completed childbearing should undergo total abdominal hysterectomy and bilateral salpingooophorectomy regardless of the stage of disease. However, preservation of fertility is an essential consideration in young patients.15 If the contralateral ovary and/or uterine serosa are grossly involved by tumor, the surgeon may have no choice but to remove the uterus and both adnexae. If the contralateral ovary and uterine serosa appear normal, conservative management with preservation of the uterus
457
and contralateral adnexa is appropriate, as 95% of sex cordstromal tumors are unilateral. The treatment of patients who have had inadequate staging is a difficult issue. Limited information exists as to the best course of action for these patients. If the patient has documented large amounts of residual disease after a limited initial attempt at tumor reduction, repeat exploration with staging and tumor-reductive surgery would be reasonable. If the patient has had an inadequate exploration, such as through a small Pfannenstiel incision or through a limited laparoscopy, more information needs to be collected prior to making a decision about post-surgical treatment. We have recommended several options, including repeat laparoscopic or open exploration with full surgical staging or, in some circumstances, a physical examination, computed tomography (CT), and measurement of serum inhibin and serum CA125 levels. If the results of all of these are negative, the decision may be made to observe the patient clinically, with or without hormonal suppression therapy using leuprolide acetate.19 General Treatment Guidelines – Radiotherapy, Chemotherapy, Biologic Agents, Multimodality Agents
Since stromal tumors of the ovary are relatively rare, controlled clinical trials designed to determine which treatment regimens are best for certain histologic subtypes are not feasible. Most published studies combine most or all subtypes of stromal ovarian tumors, and therefore treatment recommendations are based on limited data. Most data have been gathered from patients with adult granulosa cell tumors, but occasionally other tumor types are encountered, and treatment is generalized to these types as well.15,20,21 Most patients with surgical stage I disease do not require adjuvant treatment.22 Patients with stage IC disease may benefit from some adjuvant therapy. Either platinum-based chemotherapy or hormonal therapy with leuprolide acetate has been recommended for this group of patients.21 Patients with more advanced disease are typically treated with combination chemotherapy. The data regarding platinum-based therapy originated in the 1970s and 1980s, with multiple investigators publishing anecdotal reports of several complete and partial responses to platinum-containing regimens, including vincristine, actinomycin D, and cyclophosphamide (VAC), doxorubicin/cisplatin, cyclophosphamide, doxorubicin, cisplatin (CAP), and altretamine/cisplatin.23 – 29 In 1986, Colombo investigated the combination of bleomycin, vinblastine, and cisplatin in patients treated up front for advanced disease, and found 9 of 11 patients responded, although severe toxicity also occurred.30 Subsequent trials used etoposide in place of vinblastine, and in 1996 Gershenson reported an 83% response rate in nine patients with advanced disease.31 Subsequently, in 1999 Homesley reported Gynecologic Oncology Group (GOG) 115, with 57 evaluable patients with stage II –IV disease. Sixty-one percent of patients experienced grade 4 myelotoxicity, and 37% of patients had a negative second-look surgery. Thus, 69% of patients with advanced stage primary and 51% of patients with recurrent disease
458
GYNECOLOGICAL CANCERS
remained progression-free. The progression-free interval was 24 months. As a result, many patients have been treated with three to four courses of bleomycin, etoposide, and cisplatin (BEP)32 (see Table 2). However, a recent report has shown paclitaxel and carboplatin to have good results and fewer toxic effects.33 Confirmation of equivalent outcomes between these two regimens awaits performance of a larger randomized trial. Patients with recurrent disease after a long diseasefree interval may undergo secondary cytoreductive surgery. In our series,34 we have many patients, including some with multifocal disease, who have enjoyed long-term survival after multiple tumor-reductive surgeries. In cases of widespread disease or disease refractory to surgery, chemotherapy and hormonal therapy are options for treatment. Although the response rate is higher earlier in the disease course and declines as the number of prior treatment regimens increases, paclitaxel in combination with carboplatin results in a 60% overall response rate with acceptable toxicity.33 Other chemotherapeutic agents with demonstrated response include carboplatin; BEP; cisplatin, doxorubicin, and cyclophosphamide; etoposide and cisplatin; VAC; oral etoposide; topotecan; liposomal doxorubicin; paclitaxel; ifosfamide and etoposide.34 – 36 Paclitaxel and carboplatin remain the most commonly used single agents at first and second relapse. Early in the treatment of recurrent disease,
Table 2 Protocol for BEP (bleomycin, etoposide, and cisplatin).
Maintenance fluids of D5NS with 10 mEq L−1 KCl and 8 mEq L−1 magnesium sulfate at 42 mL hour−1 are initiated on admission and continued during and 24 hour postchemotherapy. Thirty minutes prior to cisplatin each day, prehydrate with 1 L normal saline with 20 mEq KCl and 16 mEq magnesium sulfate at 250 mL hour−1 for 4 hours, and give Ondansetron 8 mg in 50 mL NS IVPB, and Dexamethasone 20 mg in 50 mL NS IVPB, and Diphenhydramine 50 mg in 50 mL NS. Cisplatin 20 mg m−2 day−1 in 1 L NS with 50 g mannitol IVPB over 4 hours on days 1 – 5 Etoposide 100 mg m−2 day−1 in 500 mL NS IVPB over 2 hours on days 1 – 5 Bleomycin 10 International units in 1L NS IVPB over 24 hours on days 1 – 3 Follow with: Ondansetron 8 mg in 50 mL NS IVPB q8h, and Albuterol nebulizers 2.5 mg q6h for 24 hours, and Prochlorperazine 10 mg in 50 mL NS IVPB q6h prn nausea. Regimen repeated every 28 days. NS, normal saline; IVPB, intravenous piggyback; D5NS, 5% dextrose in normal saline.
leuprolide acetate frequently results in the regression or stabilization of disease.37 Commonly used dosing schedules are listed in Table 3. Radiation therapy is also occasionally employed in the treatment of localized or symptomatic disease.18,38 – 41
Table 3 Common dosing schedules for chemotherapy and hormonal therapy.
Agent Paclitaxel/Carboplatin Paclitaxel Paclitaxel Carboplatin BEP
PAC
EP VAC
Oral etoposide Topotecan Doxil Ifosfamide/etoposide
Leuprolide acetate
Dose −2
175 mg m , AUC = 5 135 – 200 mg m−2 80 – 100 mg m−2 AUC = 5 B 15 units day 1, E 75 mg m−2 days 1 – 5, Cisplatin 20 mg m−2 days 1–5 Cisplatin 40 – 50 mg m−2 , Doxorubicin 40 – 50 mg m−2 , Cyclophosphamide 400 mg m−2 Etoposide 100 mg m−2 , Cisplatin 75 mg m−2 Vincristine 1.5 mg m−2 day 1 Actinomycin D 0.5 mg days 1–5 Cyclophosphamide 150 mg m−2 days 1 – 5 Etoposide 50 mg m−2 day−1 × 21 days Topotecan 1.5 mg m−2 day−1 × 5 days Doxil 40 mg m−2 Ifosfamide 1.2 g m−2 day−1 × 5 days Etoposide 100 mg m−2 day−1 × 5 days 7.5 mg or 22.5 mg
IV, intravenous; IM, intramuscular; AUC, area under the curve.
Route
Interval
IV IV IV IV IV
Every 3 Every 3 Weekly Every 4 Every 3
IV
Every 4 weeks
IV
Every 4 weeks
IV
Every 2 weeks
IV
Every 4 weeks
IV
Every 4 weeks
PO
Every 21 days
IV
Every 3 weeks
IV IV
Every 4 weeks Every 3 weeks
IV
Every 3 weeks
IM IM
Every 4 weeks Every 3 months
weeks weeks weeks weeks
STROMAL TUMORS OF THE OVARY
GRANULOSA STROMAL CELL TUMORS Granulosa Cell Tumors Granulosa cell tumors constitute between 2 and 5% of all ovarian cancers and represent 90% of the stromal ovarian tumors.18 The incidence of granulosa cell tumors varies from 0.58 to 1.6 cases per 100 000 women.42,43 Most occur during the reproductive years, but they have been reported from infancy to the 10th decade of life. Granulosa cell tumors occur in two distinct histologic varieties, adult and juvenile. The patient profile, histologic appearance, natural history, and recommended treatment differ between these subtypes. Adult Granulosa Cell Tumors
Adult granulosa cell tumors represent 95% of granulosa cell tumors. They can occur at any age but are most common in perimenopausal women. Patients often present with abnormal vaginal bleeding, abdominal distension and/or pain, and occasionally signs of virilism.21,44,45 Evaluation usually yields a solid adnexal mass. These tumors may rupture, resulting in hemoperitoneum and pain. If the patient has abnormal uterine bleeding, a preoperative endometrial biopsy or intraoperative endometrial curettage should be performed, as the excess estrogen produced by many granulosa cell tumors can lead to endometrial hyperplasia or malignancy. Up to 55% of patients have associated endometrial hyperplasia or polyps, independent of age; 4–20% of patients have a synchronous endometrial cancer, a risk identified in patients over 45 years of age.12,46 If malignancy is encountered, the uterus should be removed regardless of patient age, as is noted above. Occasionally, adult granulosa cell tumors are diagnosed during pregnancy.47,48 In such patients, hormonal manifestations are less common, and the tumors are large and often complicated by rupture.48 The clinico-pathological features of these tumors have been reported in several large series.19,21,43,49 – 51 The overall 20-year survival approximates 40%. The stage at presentation is the strongest prognostic factor, with a 5–10 year survival over 90% for stage I, 55% for stage II, and 25% for stage III tumors. Adult type granulosa cell tumors are characterized by late recurrence, and recurrence over a decade after the initial diagnosis is not unusual. The average time to recurrence is 5–10 years, but recurrence up to 30 years later has been reported.52 Other prognostic factors include tumor size, rupture, and bilaterality. In patients with stage I disease, recurrences are rare for tumors less than 5 cm, but recur at a rate of 20% for tumors 5–15 cm in size and over 30% for tumors greater than 15 cm.19 Molecular and pathologic markers have recently been evaluated as prognostic indicators for adult granulosa cell tumors. A high mitotic count appears to confer a worse prognosis, but the impact of atypia is less clear.53 – 55 Aneuploidy and Ki67 expression, markers of cellular proliferation, appear to confer a worse prognosis, but these results are somewhat controversial.55 – 58 Other molecular markers associated with poor prognosis in other tumors do not appear to play a role
459
in granulosa cell tumors. These include p53, c-myc, p21, ras, and c-erbB2.,55,56,59 Chromosomal abnormalities have also been recently evaluated in granulosa cell tumors. Detected abnormalities include trisomy 12, monosomy 22, and deletion of chromosome 6, but these studies await confirmation.60 – 65 Most recently, monosomy 22, often in conjunction with trisomy 14, has been detected in these tumors, as have deletions in 22q64 and frequent microsatellite instability.65 Surgical recommendations are presented above. Since only 5% of adult granulosa cell tumors are bilateral, it is appropriate to conserve a normal-appearing uterus and contralateral ovary in a reproductive-age woman with apparent earlystage disease that is confirmed by frozen section analysis to be an adult granulosa cell tumor.11,15,18,66 Surgical staging should be performed. Additionally, serum inhibin and CA125 levels, if not obtained preoperatively, should be obtained after surgery, as they may be helpful in postoperative follow-up to confirm the resolution of disease and identify recurrence. Even though there are no prospective randomized studies showing the value of radiotherapy in the treatment of granulosa cell tumors of the ovary, there are several retrospective studies that have demonstrated that the use of radiation therapy may prolong disease-free survival in select patients with advanced or recurrent disease.18,38 – 41 In one study by Wolf et al.38 14 patients with advanced or recurrent granulosa cell tumors were retrospectively evaluated. With a median follow-up time of 13 years, 10 patients were treated with whole abdominal/pelvic radiation therapy and four patients received only pelvic radiation therapy. Of these 14 patients, 6 achieved a complete clinical response and these responders were followed up for 5 to 12 years, and three of six patients remained in remission at the end of the study. The other three patients experienced relapse between 4 and 5 years after radiation. In another study,39 62 patients were evaluated, of which eight received radiation therapy for advanced disease that was incompletely resected. Of the eight patients who received radiation therapy, 4 achieved a complete response, including three with disease-free intervals of at least 4 years. Even though these studies show an improvement with adjuvant radiation therapy, there are other studies that have reported no benefit from adjuvant radiation therapy.21,44,53 Although adult granulosa cell tumors are indolent lesions, they can recur many years, even decades, following the initial diagnosis and treatment. Patients should be followed up at gradually increasing intervals with physical examinations and with serum inhibin A and CA125 measurements.20,67 – 69 Some have suggested that patients with granulosa cell tumors may be at increased risk for the development of breast cancer.21,44 However, the risk has not been adequately characterized to amend screening criteria, so patients should follow routine breast screening recommendations. Additionally, one report exists of adult granulosa cell tumors of the ovary in two first-degree relatives, but this is an isolated instance and may not speak of the biology of the tumor.70 Recommendations for the treatment of recurrent disease are outlined above. If recurrence is diagnosed, eventual
460
GYNECOLOGICAL CANCERS
disease-related prognosis is poor, with previous statistics showing that over 70% of patients died despite treatment with chemotherapy, radiation, or both.21 As noted above, secondary cytoreductive surgery, chemotherapy, radiation, and hormonal therapy all represent appropriate treatment approaches. Secondary cytoreductive surgery is most appropriate when there has been a relatively long disease-free interval and disease appears to be resectable.71 If the patient is not a surgical candidate, or if disseminated or unresectable disease is present, or if postoperative chemotherapy is deemed appropriate, chemotherapy may be an option. Too little information exists to determine the best approach with certainty, but given the largest reports, the best chemotherapeutic regimens seem to be platinum based. Either BEP if not previously used, or a taxane-platinum combination may be the most appropriate chemotherapeutic regimen for recurrent disease, yielding similar response rates of 54 and 72%, respectively.33 Hormonal therapy has also been reported to be effective. Many adult granulosa cell tumors express steroid receptors, so treatment with gonadotropin releasing hormone antagonists35,72 and progestins36,37 has been performed. Several responses to both categories of hormone have been reported. The use of radiation for the treatment of recurrent disease has also been reported, with several responses noted. However, based on the small numbers of patients, the data are anecdotal, the response rates are short, and the impact on survival remains unknown.18,23,38 – 41,71 Juvenile Granulosa Cell Tumors
Juvenile granulosa cell tumors represent only 5% of granulosa cell tumors, but are distinct from their adult counterpart in natural history and pathologic characteristics. Juvenile granulosa cell tumors tend to occur in adolescents and teenagers, although most adolescents and young adults with ovarian malignancies do have germ cell tumors of the ovary. One study identified 38 cases of pediatric ovarian tumors, and 15% were stromal ovarian tumors, all of which were juvenile granulosa cell tumors.73 From a clinical perspective, these tumors are quite distinct from adult granulosa cell tumors. At the time of diagnosis, isosexual precocity is not unusual. They usually present with a palpable mass on pelvic or rectal examination, are unilateral in over 95% of cases, and most are diagnosed as stage IA tumors.74,75 Even though the survival rate in patients with early-stage tumors is above 95%, patients should still be surgically staged, because advanced stage tumors tend to be more aggressive and less responsive to treatment, with a shorter disease-free interval than adult type tumors. Platinumbased chemotherapy is recommended for any patient with disease over stage IA. Therefore, it is essential to stage each patient, in order not to miss any occult disease which would require treatment. Also, late relapses are unusual. A clinical association has been described between juvenile granulosa cell tumors and Ollier’s disease (enchondromatosis) and Maffucci’s syndrome (enchondromatosis and hemangiomatosis). An increased risk for the development of breast cancer has also been reported.76
The gross appearance of these tumors resembles that of the adult variant. The most common presentation is a tumor with cystic and solid components. Uniformly solid tumors, uniformly cystic tumors, and hemorrhagic cysts can also occur. On microscopic examination, two cytologic characteristics distinguish juvenile from adult granulosa cell tumors: the nuclei of juvenile granulosa cell tumors are rounded and hyperchromatic with moderate to abundant eosinophilic or vacuolated cytoplasm, and the cells are frequently luteinized. Cytogenetic studies have identified trisomy 1277 and a deletion in chromosome 6q.78 Additionally, a high mitotic index may be a negative prognostic factor.79 Treatment of the patient with juvenile granulosa cell tumor follows the general treatment guidelines above. Over 95% of patients have unilateral disease, so conservative surgery (unilateral salpingo-oophorectomy) with preservation of fertility is almost always an option. Also, since the majority of these tumors occur in adolescent girls and young women, maintaining reproductive capacity without adversely affecting survival is of paramount importance. A complete staging procedure is a mandatory part of the procedure. In the rare patient with advanced disease, a total abdominal hysterectomy with bilateral salpingooophorectomy is necessary. Again, most trials have combined adult and juvenile types, so specific treatment guidelines for the juvenile type cannot be made. The role of adjuvant treatment is not entirely clear, but it is recommended that all patients with over stage IA disease receive adjuvant chemotherapy, historically with BEP,32 although a taxane/platinum combination may also be utilized as for adult granulosa cell tumors. Additionally, a recent study has suggested that postoperative cisplatinbased chemotherapy may improve survival specifically for the juvenile granulosa cell tumor type.79 Some have used methotrexate, dactinomycin, and cyclophosphamide with a suggested treatment benefit.80 The role of radiation is largely unknown. Unfortunately, when juvenile granulosa cell tumors recur, they usually do so after a shorter progression-free interval than is seen with adult granulosa cell tumors. Although many approaches to treatment have been used, including surgical cytoreduction, radiation therapy, and multiple chemotherapy regimens including high-dose chemotherapy, few sustained responses are seen in patients with recurrent juvenile granulosa cell disease. In our experience, responses have been achieved with BEP; paclitaxel and/or carboplatin; topotecan; bleomycin, vincristine, and cisplatin; etoposide and cisplatin; cisplatin, doxorubicin, and cyclophosphamide; highdose chemotherapy; and gemcitabine.34 According to a single report, patients with recurrent disease may benefit more from multiagent chemotherapy than single-agent chemotherapy.79 Hormonal therapy with leuprolide acetate has resulted in several cases of stable disease.35 Overall, the prognosis for patients with juvenile granulosa cell tumor remains good but is related to stage. The 5-year survival for patients with stage IA disease is 99%, but this declines to 60% for patients with advanced disease.75,79
STROMAL TUMORS OF THE OVARY
Thecomas/Fibromas This category of ovarian stromal tumor represents a spectrum of neoplasms with significant overlap which predominantly have clinically benign behavior and are derived from ovarian stromal cells. These tumors share many similar clinical characteristics and often cannot be assigned to either the distinct thecoma or fibroma category based on the clinical or microscopic examination. Thecomas
Thecomas represent approximately 1% of ovarian neoplasms. Most cases occur in postmenopausal women,21,81 and only 10% of patients are under the age of 30 years. Thecomas are often hormonally active and may cause abnormal vaginal bleeding, which is the most common presenting symptom. Additionally, 37–50% of patients with thecomas have endometrial hyperplasia and up to 27% have an associated endometrial carcinoma.21,81,82 Therefore, just as in granulosa cell tumors, patients presenting with abnormal bleeding should always have the endometrium sampled. Thecomas may also contain a population of luteinized cells (see Table 1), and these luteinized thecomas may be androgenic and cause virilization.83 Radiographically, most tumors (79%) are solid on CT and show delayed accumulation of contrast material.84 Some have suggested that the diagnosis can be suggested by poor penetration of the mass with acoustic shadowing on ultrasound.85 The specific radiological diagnosis, however, is nonspecific. Histologically, thecomas are composed of lipid-laden stromal cells with abundant pale cytoplasm. Their clinical behavior is usually benign and the prognosis is excellent. Occasionally, tumors may exhibit nuclear atypia and mitoses; however fibrosarcomas and luteinized granulosa cell tumors should be ruled out in these cases.86 Additionally, abnormalities in chromosome 12 have been described, but these have not been specifically related to prognosis.87 Because these tumors have a benign course, surgical resection alone without staging or adjuvant treatment is the appropriate therapy. Given the age distribution for patients with this tumor, preservation of fertility is not usually an issue. Conservative management may include only removal of the ovary or ovaries, but then the endometrium should be sampled with a dilation and curettage. If the patient has associated endometrial hyperplasia or carcinoma, or if she is postmenopausal, a total hysterectomy and bilateral salpingooophorectomy is appropriate.11 Fibromas
The most common stromal tumor of the ovary is the fibroma, representing 4% of all ovarian neoplasms. These are usually benign, unilateral and hormonally inactive tumors. The mean age at diagnosis is 48 years. Clinically, patients present with pelvic heaviness or pain and a mass. Gross examination reveals the tumors to be solid and white, although degenerative cystic cavities are not uncommon. The average size is 6 cm but size increases with the age at diagnosis.88 The presence of ascites is not uncommon, and it tends to occur with increasing tumor size; for example,
461
ascites is present in 30% of patients with tumors greater than 6 cm. One percent of patients will develop a hydrothorax, called Meigs’ syndrome.89
Cellular Fibroma/Fibrosarcoma Approximately 10% of fibromas will show light microscope evidence of hypercellularity, as well as mitoses. Tumors of low malignant potential are designated as those with an increased cellular density, only mild nuclear atypia, and less than three mitotic figures per 10 high-power fields. Fully malignant fibrosarcomas have increased cellularity, marked pleomorphism and frequent mitotic figures. The presence of trisomy 8 may be useful in distinguishing between the diagnosis of fibroma and fibrosarcoma.90 In contrast to the benign fibroma, fibrosarcomas are highly aggressive tumors.91 They are usually large, unilateral, and highly vascular. At the time of surgery, rupture, adhesions, hemorrhage, and necrosis are often seen.
Fibrothecoma As noted above, fibromas and thecomas represent a spectrum of neoplasms. It is not unusual for patients to have evidence of a mixed tumor with elements of fibroma and thecoma within one ovarian tumor. These are also benign, and are treated with definitive surgery alone. They do not require adjuvant treatment and do not recur or metastasize.11
Sertoli-Stromal Cell Tumors Sertoli-stromal cell tumors are also known as androblastomas, and they represent a group of tumors which differentiate toward testicular structures. They are more commonly referred to as Sertoli-Leydig cell tumors, and they may contain only Sertoli cells, or both Sertoli and Leydig cells. These rare tumors represent less than 1% of all ovarian tumors. As noted in Table 1, they are classified in five groups: well differentiated, intermediately differentiated, poorly differentiated, retiform, and mixed. Well-differentiated tumors include Sertoli cell tumors, and Sertoli-Leydig cell tumors. Sertoli-Leydig cell tumors tend to occur in young women with a mean age of 25 years. Well-differentiated tumors tend to occur approximately 10 years later than intermediate or poorly differentiated tumors. Conversely, the retiform type is usually diagnosed at a younger age than intermediate or poorly differentiated types.92,93 Pure Sertoli tumors are diagnosed in young women, and Sertoli-Leydig tumors tend to occur in women in their teens and twenties. Thus, fertility preservation is very important for many of these patients, and this is usually appropriate as over 95% of all tumors are unilateral with a normal uterus.15,66 Upon presentation, approximately 50% of patients demonstrate virilization. The differential diagnosis includes adrenal and gonadal hyperplasia and a neoplasm. The presence of a unilateral adnexal mass, however, palpated on examination or visualized on imaging, suggests an ovarian neoplasm as the source of the virilization. The size of the tumor, however, does not predict the ability to cause virilization, so a very small neoplasm can be responsible for a significant testosterone elevation and cause virilization.93
462
GYNECOLOGICAL CANCERS
The evaluation of a patient with virilization, or androgen excess, includes a transvaginal pelvic ultrasound to visualize the ovaries. Also, serum dehydroepiandrostenedione sulfate (DHEAS) and testosterone levels should be measured. An elevated DHEAS suggests that the adrenal gland is the source of the androgen excess. Conversely, an elevated testosterone suggests an ovarian source. A CT may visualize an adrenal mass responsible for secretion of DHEAS. Additional studies helpful in the workup of virilization include 17-OH progesterone (elevated in congenital adrenal hyperplasia) and cortisol (elevated in Cushing’s disease). Prior to surgical intervention, the offending mass should be detected by imaging or physical examination. It has been suggested that ovarian vein catheter studies can be helpful in detecting the source if the imaging and examination are unrevealing. Gross examination shows Sertoli-Leydig cell tumors to be solid or mixed cystic and solid, with no features pathognomonic for Sertoli-Leydig cell tumors on visual inspection. The size is variable, ranging from microscopic to 25 cm.93,94 Well-differentiated tumors tend to be smaller, and poorly differentiated tumors tend to be larger.92 Microscopic examination shows that well-differentiated tumors, which account for 11% of cases, have a predominantly tubular pattern. The Sertoli cells are cuboidal or columnar with round nuclei, but with no prominent nucleoli. Atypical nuclei are absent or rare and few mitotic figures are seen. The stroma contains nests of Leydig cells. As seen in Table 1, the most common variants are intermediate differentiation (54%) and poor differentiation (13%). These subgroups are characterized by a continuum of different patterns and combinations of cell types, with both Sertoli and Leydig components exhibiting various degrees of maturity. A retiform component is present in 15% of tumors, demonstrating tubules and cysts arranged in a pattern that resembles the rete testis. Twenty-two percent of cases contain heterologous elements, most commonly mucinous glands. Management of patients with Sertoli-stromal cell tumors follows the guidelines above. In young patients, fertility is of course a significant issue. In young patients, a unilateral salpingo-oophorectomy and staging is usually appropriate, as 95% of lesions are unilateral.15,66 However, in patients who have completed their childbearing, a total hysterectomy and bilateral salpingo-oophorectomy with staging procedure is the procedure of choice.11 Patients with stage IA well-differentiated tumors do not require adjuvant therapy. Patients with stage IC disease or greater, with poorly differentiated tumors of any stage, or with heterologous elements present have a 50 to 60% risk of recurrence.20 Owing to the rare nature of the disease and the lack of controlled trials, there is little scientific support for any treatment approach; however, adjuvant therapy seems reasonable given the substantial risk of recurrence. Radiation and hormone therapy have been described,92,95 but there is very limited information upon which to base treatment. Information regarding chemotherapy on these patients is available only as part of the larger, previously cited studies, and therefore, it is the practice of the authors to administer adjuvant therapy in the form of three to four courses of BEP or six courses of paclitaxel and carboplatin to these patients.
Stage is clearly the most important prognostic factor. At the time of diagnosis, over 90% of patients have stage IA disease. This is largely dependent on grade; in one series, every patient with a well-differentiated tumor was uniformly stage IA, but only 52% of patients with poorly differentiated tumors were stage IA.96 There have been no reported patients with advanced stages or recurrence, and only one death from disease has been reported in a patient with a well-differentiated tumor. However, 10% of intermediate, 60% of poorly differentiated, and 20% of retiform and heterologous subtypes show malignant behavior, leading to the recommendation for adjuvant treatment in these groups. Other poor prognostic factors include the presence of thyroid nodules, tumor size, mitotic activity, tumor rupture, features of rhabdomyosarcoma, and other heterologous elements, especially when containing mesenchymal elements. One report of familial occurrence has been reported.92,97,98 Patients with Sertoli-Leydig cell tumors can be followed up with physical examination and with serum α-fetoprotein, inhibin, and testosterone level measurement. Eighteen percent of patients with Sertoli-Leydig cell tumors recur, and of those that recur, two-thirds do so within the first year. Platinum-based chemotherapy is the mainstay of treatment upon recurrence, but anecdotal reports show that outcomes are poor for patients with recurrent disease.20
Sex Cord Tumor with Annular Tubules (SCTAT) The group of tumors known as ovarian SCTAT represents a separate category of sex cord-stromal tumors which was first described by Scully in 1970.3 It is controversial whether these tumors are more closely related to granulosa cell tumors or Sertoli-Leydig cell tumors, as the cellular elements appear to be somewhat intermediate in nature, but they do seem to represent a distinct entity. Microscopically, these tumors are characterized by either simple or complex ring-shaped tubules surrounding hyalinized basement membrane material. The clinical presentation is usually related to abnormal vaginal bleeding. Abdominal pain and intussusception have been reported but are less common. These tumors are uncommon in the adolescent population but can present with isosexual precocity. Clinically, there appear to be two subgroups of ovarian SCTATs. The first subgroup is associated with Peutz-Jeghers syndrome and is typically multifocal, bilateral, and almost always benign. These tumors are small and rarely palpable on examination. Patients with this tumor should be carefully screened for adenoma malignum of the cervix, as up to 15% of patients may have an occult lesion.4 Therefore, hysterectomy should be strongly considered in these patients. The second subgroup of ovarian SCTATs occurs incidentally, independent from Peutz-Jeghers syndrome. These tumors are usually larger and have a significant potential for malignant behavior. The cornerstone of treatment remains surgical resection, following the general guidelines above. These tumors tend to remain lateralized and have a tendency toward lymphatic spread. A percentage of these tumors recur and metastasize. M¨ullerian-inhibiting substance and inhibin
STROMAL TUMORS OF THE OVARY
are useful tumor markers. If the tumor recurs, repeat cytoreductive surgery is usually warranted. Chemotherapy remains unproved outside of anecdotal cases.99
Gynandroblastoma Gynandroblastomas are a separate, rare entity comprised of granulosa cell elements, and Sertoli cells. The specific cell of origin remains debated, but it may arise from undifferentiated mesenchyme.100 Gynandroblastomas are responsible for less than 1% of all ovarian stromal tumors. Microscopically, these combine elements of both “male and female directed cells”.101 These tumors must show unequivocal granulosa/Sertoli cell elements, must be well differentiated, and must demonstrate intimate mixing of the constituent cell types. Clinically, signs and symptoms are related to estrogen and androgen overproduction. Most patients are in their third to fifth decades of life.100,102,103 Androgen excess is present in 60% of patients with gynandroblastoma. Therefore, virilization is visible, but estrogenic stimulation of specific end organs still occurs, so endometrial hyperplasia is commonly noted.104 Most of these tumors are solid and large, measuring between 7 and 10 cm in size, with yellow-white cystic areas present. Also, most gynandroblastomas are appreciated on the pelvic examination because of their size. These tumors should also be staged and aggressively cytoreduced. Since these tumors are so uncommon, a distinct preoperative diagnosis of gynandroblastoma would be uncommon. However, the workup is similar to that of other patients with androgen excess and a unilateral adnexal mass – this is thought to be an androgen-producing ovarian neoplasm. Preoperative testing usually reveals elevated levels of testosterone100,104,105 or urinary 17-ketosteroids.101 Elevated androstenedione, dehydroepiandrosterone, dihydrotestosterone, and estradiol have also been variably reported.106 – 108 As with the other tumor types in this group, surgery is the cornerstone of therapy, and general guidelines can be followed as outlined above. In young women where reproductive function is important, a unilateral salpingo-oophorectomy and staging procedure are advocated. In postreproductive patients, hysterectomy and bilateral salpingo-oophorectomy are indicated, and in patients with advanced disease, cytoreductive surgery is appropriate. Patients with advanced disease have a poor prognosis.104 Therefore, these patients are treated with adjuvant treatment, although the data are very limited. Anecdotal reports of P32 and cyclophosphamide, actinomycin D, and vincristine combination chemotherapy are published. Serum estrogens or androgens can be followed as tumor markers.
Steroid (lipid) Cell Tumor Steroid, or lipid, cell tumors consist of three lesions based on their cell of origin and microscopic features: stromal luteomas, Leydig cell tumors, and steroid cell tumors not otherwise specified (NOS). Together these three entities represent less than 0.1% of all ovarian tumors.
463
Stromal Luteoma
Stromal luteomas are often small, with half of them being less than 5 cm.109 Microscopically, they consist of large, rounded or polyhedral cells resembling Leydig cells, luteinized ovarian stromal cells, and adrenocortical cells. They represent approximately 25% of steroid cell tumors, and have been identified during pregnancy. Most patients, however, are postmenopausal at the time of diagnosis. These tumors are benign steroid cell tumors that do not require staging or postoperative therapy. Childbearing potential should be maintained in the occasional young patient with this diagnosis.20 Leydig Cell Tumor
Leydig cell tumors represent 15–20% of all steroid cell tumors of the ovary. They are subdivided into tumors of hilar and nonhilar type, and both are benign. Clinically, they are unilateral with a median size of less than 3 cm.110 – 112 This tumor often presents in postmenopausal patients. Histologically, they consist only of Leydig cells, and crystals of Reinke are seen. These tumors are benign steroid cell tumors that do not require staging or postoperative therapy. Childbearing potential should be maintained in the occasional young patient with this diagnosis. Steroid Cell Tumors not Otherwise Specified
Steroid cell tumors NOS are a category of steroid cell tumor that can be malignant and aggressive. These are the most common type of steroid cell tumor, and therefore, when a steroid cell tumor is diagnosed intraoperatively, it should be staged and aggressively cytoreduced. Histologically, these lipid cell tumors lack the specific characteristics of stromal luteomas or Leydig cell tumors. Clinically, they present at an earlier age, with a mean age of 43 years, and are larger than the other steroid cell tumors with an average size of 8.5 cm. Bilaterality is not infrequent. In the largest series, 43% of patients demonstrated extraovarian disease either at diagnosis or during surveillance. Negative prognostic factors included age, size, increased mitosis, and presence of necrosis.113 The strongest prognostic factor other than stage is the number of mitotic figures, since over 90% of tumors with over two mitoses per 10 high-power fields are malignant. Stage IA tumors in women of reproductive age should be staged, and a unilateral salpingo-oophorectomy is appropriate. Patients not opting for future fertility require hysterectomy and bilateral salpingo-oophorectomy with staging, and patients with advanced disease require cytoreductive surgery. Although all reports are anecdotal, patients with tumors that are pleomorphic, have an increased mitotic count, are large, or are at an advanced stage should be treated with additional postoperative platinum-based chemotherapy.114 Radiation, melphalan, and hormonal therapy have also been used with variable outcomes.115 – 117
Summary Most ovarian stromal tumors are clinically indolent and are reported to have a good long-term prognosis. However,
464
GYNECOLOGICAL CANCERS
many occur in adolescent and reproductive-aged women, and therefore individualized treatment with consideration for fertility preservation is of great importance. Appropriate treatment guidelines based on current information have been presented in this chapter. Since this category represents a wide range of disease entities, broad generalizations regarding prognosis and treatment recommendations cannot be made. So, our approach to the management of these cases is to follow these general guidelines for appropriate treatment of patients with ovarian stromal tumors, and to undertake disease-specific therapy once the final pathology is identified. Knowledge remains limited regarding this rare group of diseases. However, with continuing commitment from cooperative groups in order to perform high-quality research, progressively better treatment will become available and more information will be known.
REFERENCES 1. Meyer R. Pathology of some special ovarian tumors and their relation to sex characteristics. Am J Obstet Gynecol 1931; 22: 697. 2. Morris JM, Scully RE. Endocrine Pathology of the Ovary. St. Louis, Missouri: CV Mosby Company, 1958. 3. Scully RE. Sex cord tumor with annular tubules: a distinctive ovarian tumor of the Peutz-Jeghers syndrome. Cancer 1970; 25: 1107. 4. Srivasta PJ, Keeney GL, Podratz KC. Disseminated cervical adenoma malignum and bilateral ovarian sex cord tumors with annular tubules associated with Peutz-Jeghers syndrome. Gynecol Oncol 1994; 53: 256. 5. Young RH, Path FRC, Scully RE. Sex cord-stromal, steroid cell, and other ovarian tumors with endocrine, paraendocrine, and paraneoplastic manifestations. In Kurman RJ (ed) Blaustein’s Pathology of the Female Genital Tract. New York: Springer-Verlag, 1994: 783. 6. Jemal A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55: 10 – 30. 7. International Histologic Classification of Tumors. Geneva, Switzerland: World Health Organization, 1973,: Vol. 9. 8. O’Leary M, et al. Female genital system cancers. In Bleyer A (ed) Adolescent and Young Adult Cancer Monograph. New York: SpringerVerlag, 2005. 9. Koonings PP, et al. Relative frequency of primary ovarian neoplasms: a 10-year review. Obstet Gynecol 1989; 74: 921 – 6. 10. Young RH, Scully RE. Endocrine tumors of the ovary. Curr Top Pathol 1992; 85: 113 – 64. 11. Gershenson DM, Hartmann LC, Young RH. Ovarian sex cord-stromal tumors. In Hoskins WJ, Perez CA, Young RC (eds) Principles and Practice of Gynecologic Oncology, 4th ed. Philadelphia, Pennsylvania: Lippincott Williams and Wilkins, 2004. 12. Unkila-Kallio L, et al. Reproductive features in women developing ovarian granulosa cell tumour at a fertile age. Hum Reprod 2000; 15: 589 – 93. 13. Robertson DM, McNeilage J. Inhibins as biomarkers for reproductive cancers. Semin Reprod Med 2004; 22: 219 – 25. 14. Choi K, et al. Ovarian granulosa cell tumor presenting as Meigs’ syndrome with elevated CA125. Korean J Intern Med 2005; 20: 105 – 9. 15. Gershenson DM. Fertility-sparing surgery for malignancies in women. J Natl Cancer Inst Monogr 2005; 34: 43 – 7. 16. Canis M, et al. A 12 year experience with long term follow-up. Obstet Gynecol 1994; 83: 707 – 12. 17. Mettler L, Semm K, Shive K. Endoscopic management of adnexal masses. J Soc Laparosc Surg 1997; 2: 103 – 12. 18. Schumer ST, Cannistra SA. Granulosa cell tumor of the ovary. J Clin Oncol 2003; 21: 1180 – 9. 19. Bjorkholm E, Silfversward C. Prognostic factors in granulosa-cell tumors. Gynecol Oncol 1981; 11: 261 – 74.
20. Brown J, Gershenson DM. Treatment of rare ovarian malignancies. In Gershenson DM (ed) Gynecologic Cancer, (in press). 21. Evans AT, et al. Clinicopathological review of 118 granulosa and 82 thecal cell tumors. Obstet Gynecol 1980; 55: 231. 22. Herbst AL. Neoplastic diseases of the ovary. In Mishell DR, et al. (eds) Comprehensive Gynecology, 3rd ed. New York: Mosby-Year Book Inc., 1997. 23. Schwartz PE, Smith JP. Treatment of ovarian stromal tumors. Am J Obstet Gynecol 1976; 125: 402 – 11. 24. Slayton RE. Management of germ cell and stromal tumors of the ovary. Semin Oncol 1984; 11: 299 – 313. 25. Jacobs AJ, Deppe G, Cohen CJ. Combination chemotherapy of ovarian granulosa cell tumor with cis-platinum and doxorubicin. Gynecol Oncol 1982; 14: 294 – 7. 26. Camlibel FT, Caputo TA. Chemotherapy of granulosa cell tumors. Am J Obstet Gynecol 1983; 145: 763 – 5. 27. Neville AJ, Gilchrist KW, Davis TE. The chemotherapy of granulosa cell tumors of the ovary: experience of the Wisconsin Clinical Cancer Center. Med Pediatr Oncol 1984; 12: 397 – 400. 28. Kaye SB, Davies E. Cyclophosphamide, adriamycin, and cisplatinum for the treatment of advanced granulosa cell tumors, using serum estradiol as a tumor marker. Gynecol Oncol 1986; 24: 261 – 4. 29. Gershenson DM, et al. Treatment of metastatic stromal tumors of the ovary with cisplatin, doxorubicin, and cyclophosphamide. Obstet Gynecol 1987; 70: 765 – 9. 30. Colombo N, et al. Cisplatin, vinblastine, and bleomycin combination chemotherapy in metastatic granulosa cell tumor of the ovary. Obstet Gynecol 1986; 37: 265 – 8. 31. Gershenson DM, et al. Treatment of poor-prognosis sex cord-stromal tumors of the ovary with the combination of bleomycin, etoposide, and cisplatin. Obstet Gynecol 1996; 87: 527 – 31. 32. Homesley HD, et al. Bleomycin, etoposide, and cisplatin combination chemotherapy of ovarian granulosa cell tumors and other stromal malignancies: a Gynecologic Oncology Group study. Gynecol Oncol 1999; 72: 131 – 7. 33. Brown J, et al. The activity of taxanes compared with bleomycin, etoposide, and cisplatin in the treatment of sex cord-stromal ovarian tumors. Gynecol Oncol 2005; 97: 489 – 96. 34. Brown J, et al. The activity of taxanes in the treatment of sex cordstromal ovarian tumors. J Clin Oncol 2004; 22: 3517 – 23. 35. Fishman A, et al. Leuprolide acetate for treating refractory or persistent ovarian granulosa cell tumor. J Reprod Med 1996; 41(6): 393 – 6. 36. Powell JL, Connor GP, Henderson GS. Management of recurrent juvenile granulosa cell tumor of the ovary. Gynecol Oncol 2001; 81: 113 – 6. 37. Tresukosol D, et al. Recurrent ovarian granulosa cell tumor: a case report of a dramatic response to Taxol. Int J Gynecol Cancer 1995; 5: 156 – 9. 38. Wolf JK, et al. Radiation treatment of advanced or recurrent granulosa cell tumor of the ovary. Gynecol Oncol 1999; 73: 35 – 41. 39. Savage P, et al. Granulosa cell tumors of the ovary: demographics, survival, and the management of advanced disease. Clin Oncol (R Coll Radiol) 1998; 10: 242 – 5. 40. Lee IW, et al. Radiotherapy for the treatment of metastatic granulosa cell tumor in the mediastinum: a case report. Gynecol Oncol 1999; 73: 455 – 60. 41. Wessalowski R, et al. Successful liver treatment of a juvenile granulosa cell tumor in a 4-year-old child by regional deep hyperthermia, systemic chemotherapy, and irradiation. Gynecol Oncol 1995; 57: 417 – 22. 42. Stenwig JT, Hazekamp JT, Beecham JB. Granulosa cell tumors of the ovary: a clinicopathologic study in 118 cases with long term followup. Gynecol Oncol 1979; 7: 136. 43. Bjorkholm E, Silfversward C. Granulosa and theca cell tumors: incidence and occurrence of second primary tumors. Acta Radiol Oncol 1980; 19: 161. 44. Ohel G, Kaneti H, Schenker JG. Granulosa cell tumors in Israel: a study of 172 cases. Gynecol Oncol 1983; 15: 278. 45. Nakashima N, Young RH, Scully RE. Androgenic granulosa cell tumors of the ovary. Arch Pathol Lab Med 1984; 108: 786.
STROMAL TUMORS OF THE OVARY 46. Gusberg SB, Kardon P. Proliferative endometrial response to thecagranulosa cell tumors. Am J Obstet Gynecol 1971; 111: 633. 47. Tanyi J, et al. Trisomy 12 in juvenile granulosa cell tumor of the ovary during pregnancy. A report of 2 cases. J Reprod Med 1999; 44: 826 – 32. 48. Young RH, Dudley AG, Scully RE. Granulosa cell, Sertoli-Leydig cell, and unclassified sex cord stromal tumors associated with pregnancy: a clinicopathologic analysis of thirty six cases. Gynecol Oncol 1984; 18: 181. 49. Norris HJ, Taylor HB. Prognosis of granulosa-theca cell tumors of the ovary. Cancer 1968; 21: 255. 50. Fox H, Agrawal K, Langley FA. A clinicopathologic study of 92 cases of granulosa cell tumor of the ovary with special reference to the factors influencing prognosis. Cancer 1975; 35: 231. 51. Dempster J, Geirsson RT, Duncan ID. Survival after ovarian granulosa and theca cell tumors. Scott Med J 1987; 32: 38 – 9. 52. Bjorkholm E. Granulosa cell tumors: a comparison of survival in patients and matched controls. Am J Obstet Gynecol 1980; 138: 329. 53. Malmstrom H, et al. Granulosa cell tumors of the ovary; prognostic factors and outcome. Gynecol Oncol 1994; 52: 50 – 5. 54. Miller BE, et al. Prognostic factors in adult granulosa cell tumor of the ovary. Cancer 1997; 79: 1951 – 5. 55. King LA, et al. Mitotic count, nuclear atypia, and immunohistochemical determination of Ki-67, c-myc, p21-ras, c-erbB2, and p53 expression in granulosa cell tumors of the ovary: mitotic count and Ki67 are indicators of poor prognosis. Gynecol Oncol 1996; 61: 227 – 32. 56. Evans MP, et al. DNA ploidy of ovarian granulosa cell tumors: lack of correlation between DNA index or proliferative index and outcome in 40 patients. Cancer 1995; 75: 2295 – 8. 57. Costa MJ, et al. Transformation in recurrent ovarian granulosa cell tumors: Ki67 (MIB-1) and p53 immunohistochemistry demonstrates a possible molecular basis for the poor histologic prediction of clinical behavior. Hum Pathol 1996; 27: 274 – 81. 58. Roush GR, El-Naggar AK, Abdul-Karim FW. Granulosa cell tumor of ovary: a clinicopathologic and flow cytometric DNA analysis. Gynecol Oncol 1995; 56: 430 – 4. 59. Liu FS, et al. Overexpression of p53 is not a feature of ovarian granulosa cell tumors. Gynecol Oncol 1996; 61: 50 – 3. 60. Fletcher JA, et al. Ovarian granulosa-stromal cell tumors are characterized by trisomy 12. Am J Pathol 1991; 138: 515 – 20. 61. Persons DL, et al. Fluorescence in situ hybridization analysis of trisomy 12 in ovarian tumors. Am J Clin Pathol 1994; 102: 775 – 9. 62. Lindgren V, Waggoner S, Rotmensch J. Monosomy 22 in two ovarian granulosa cell tumors. Cancer Genet Cytogenet 1996; 89: 93 – 7. 63. Teyssier JR, et al. Chromosomal changes in an ovarian granulosa cell tumor: similarity with carcinoma. Cancer Genet Cytogenet 1985; 14: 147. 64. Lin YS, et al. Molecular cytogenetics of ovarian granulosa cell tumors by comparative genomic hybridization. Gynecol Oncol 2005; 97: 68 – 73. 65. Dhillon VS, Aslam M, Husain SA. The contribution of genetic and epigenetic changes in granulosa cell tumors of ovarian origin. Clin Cancer Res 2004; 10: 5537 – 45. 66. Gershenson DM. Management of early ovarian cancer: germ cell and sex cord stromal tumors. Gynecol Oncol 1994; 55(Suppl): S62 – 72. 67. Lappohn RE, et al. Inhibin as a marker for granulosa cell tumors. N Engl J Med 1989; 321: 790. 68. Jobling T, et al. A prospective study of inhibin in granulosa cell tumors of the ovary. Gynecol Oncol 1994; 55: 285. 69. Boggess JF, et al. Serum inhibin and disease status in women with ovarian granulosa cell tumors. Gynecol Oncol 1997; 64: 64. 70. Stevens TA, et al. Adult granulosa cell tumors of the ovary in two first-degree relatives. Gynecol Oncol 2005; 98: 502 – 5. 71. Pankratz E, et al. Granulosa cell tumors: a clinical review of 61 cases. Obstet Gynecol 1978; 52: 718. 72. Martikainen H, et al. Gonadotropin-releasing hormone agonist analog therapy effective in ovarian granulosa cell malignancy. Gynecol Oncol 1989; 35: 406. 73. Gribbon M, Ein SH, Mancer K. Pediatric malignant ovarian tumors: a 43-year review. J Pediatr Surg 1992; 27: 480. 74. Young RH, Dickersin GR, Scully RE. Juvenile granulosa cell tumor of the ovary. Am J Surg Pathol 1984; 8: 575.
465
75. Lack EE, et al. Granulosa theca cell tumors in premenarchal girls: a clinical and pathological study of ten cases. Cancer 1981; 48: 1846. 76. Schoefield DE, Fletcher JA. Trisomy 12 in pediatric granulosa-stromal cell tumors. Am J Pathol 1992; 141: 1265. 77. Rodriguez E, Rao P, Reuter V. Cytogenetic analysis of a juvenile granulosa cell tumor. Cancer Genet Cytogenet 1992; 61: 207. 78. Zaloudek C, Norris HJ. Granulosa cell tumors of the ovary in children: a clinical and pathologic study of 32 cases. Am J Surg Pathol 1982; 6: 503. 79. Calaminus G, et al. Juvenile granulosa cell tumors of the ovary in children and adolescents: results from 33 patients registered in a prospective cooperative study. Gynecol Oncol 1997; 65: 447 – 52. 80. Vassal G, et al. Juvenile granulosa cell tumor of the ovary in children: a clinical study of 15 cases. J Clin Oncol 1988; 6: 990 – 5. 81. Bjorkholm E, Silfversward C. Theca cell tumors: clinical features and prognosis. Acta Radiol 1980; 19: 241. 82. Stage AH, Grafton WD. Thecomas and granulosa-theca cell tumors of the ovary: an analysis of 51 tumors. Obstet Gynecol 1977; 50: 21. 83. Zhang J, et al. Ovarian stromal tumors containing lutein or Leydig cells – a clinicopathologic analysis of fifty cases. Int J Gynecol Pathol 1982; 1: 270. 84. Bazot M, et al. Fibrothecomas of the ovary: CT and US findings. J Comput Assist Tomogr 1993; 17: 754. 85. Yaghoobian J, Pinck RL. Ultrasound findings in thecoma of the ovary. J Clin Ultrasound 1983; 11: 91. 86. Waxman M, et al. Ovarian low grade stromal sarcoma with thecomatous features. Cancer 1979; 44: 2206. 87. Izutsu T, et al. Numerical and structural chromosome abnormalities in an ovarian fibrothecoma. Cancer Genet Cytogenet 1995; 83: 84. 88. Dockherty MB, Masson JC. Ovarian fibromas: a clinical and pathologic study of two hundred and eighty three cases. Am J Obstet Gynecol 1944; 47: 741. 89. Meigs JV. Fibroma of the ovary with ascites and hydrothorax – Meigs syndrome. Am J Obstet Gynecol 1954; 67: 962. 90. Tsuji T, et al. Fibrosarcoma versus cellular fibroma of the ovary: a comparative study of their proliferative activity and chromosome aberrations using MIB-1 immunostaining, NDA flow cytometry, and fluorescence in situ hybridization. Am J Surg Pathol 1997; 21: 52. 91. Prat J, Scully RE. Cellular fibromas and fibrosarcomas of the ovary. Cancer 1981; 47: 2663. 92. Zaloudek C, Norris HJ. Sertoli-Leydig tumors of the ovary: a clinicopathologic study of 64 intermediate and poorly differentiated neoplasms. Am J Surg Pathol 1984; 8: 405. 93. Roth LM, et al. Sertoli-Leydig cell tumors: a clinicopathologic study of 34 cases. Cancer 1981; 48: 187. 94. O’Hern TM, Neubecker RD. Arrhenoblastoma. Obstet Gynecol 1962; 19: 758. 95. Emons G, Schally AV. The use of luteinizing hormone releasing hormone agonists and antagonists in gynaecological cancers. Hum Reprod 1994; 9: 1364. 96. Gordon MD, Ireland K. New developments in sex cord-stromal and germ cell tumors of the ovary. Clin Lab Med 1995; 15: 595. 97. O’Brien PK, Wilansky DL. Familial thyroid nodulation and arrhenoblastoma. Am J Clin Pathol 1981; 75: 578. 98. Goldstein DD, Lamb EJ. Arrhenoblastoma in first cousins: report of 2 cases. Obstet Gynecol 1970; 35: 444. 99. Puls LE, et al. Recurrent ovarian sex cord tumor with annular tubules: tumor marker and chemotherapy experience. Gynecol Oncol 1994; 54: 396. 100. Anderson MC, Rees DA. Gynandroblastoma of the ovary. Br J Obstet Gynecol 1975; 82: 68. 101. Neubecker RD, Breen JL. Gynandroblastoma. Am J Clin Pathol 1962; 38: 60. 102. Emig OR, Hertig AT, Rowe FJ. Gynandroblastoma of the ovary. Am J Pathol 1943; 19: 633. 103. Emig OR, Hertig AT, Rowe FJ. Gynandroblastoma of the ovary. Review and report of a case. Obstet Gynecol 1959; 13: 135. 104. Novak ER. Gynandroblastoma of the ovary: review of 8 cases from the ovarian tumor registry. Obstet Gynecol 1967; 30: 709. 105. Chalvardigjian A, Derzko C. Gynandroblastoma: its ultrastructure. Cancer 1982; 50: 710.
466
GYNECOLOGICAL CANCERS
106. Soules MR, Abraham GE, Bossen EH. The steroid profile of a virilizing ovarian tumor. Obstet Gynecol 1978; 52: 73. 107. Laatikainen T, Pelkonen R, Vihko R. Plasma steroids in two subjects with ovarian androgen producing tumors, arrhenoblastoma, gynandroblastoma. J Clin Endocrinol Metab 1972; 34: 580. 108. Luca V, et al. Gynandroblastoma of the ovary. Morphol Embryol 1983; 29: 117. 109. Hayes MC, Scully RE. Stromal luteoma of the ovary: a clinicopathologic analysis of 25 cases. Int J Gynecol Pathol 1987; 6: 313. 110. Dunnihoo DR, Grieme DL, Woolf RB. Hilar cell tumors of the ovary. Obstet Gynecol 1966; 27: 713. 111. Paraskevas M, Scully RE. Hilus cell tumor of the ovary. Int J Gynecol Pathol 1989; 8: 299. 112. Roth LM, Sternberg WH. Ovarian stromal tumors containing Leydigcells. II. Pure Leydig-cell tumor, non-hilar type. Cancer 1973; 32: 952.
113. Hayes MC, Scully RE. Ovarian steroid-cell tumors (not otherwise specified): a clinicopathological analysis of 63 cases. Am J Surg Pathol 1987; 11: 835. 114. Khoo SK, Buntine D. Malignant stromal tumor of the ovary with virilizing effects in an XXX female with streak ovaries. Aust N Z J Obstet Gynaecol 1980; 20: 123. 115. Echt GR, Hadd HE. Androgen excretion patterns in a patient with a metastatic hilus tumor of the ovary. Am J Obstet Gynecol 1968; 100: 1055. 116. Montag TW, Murphy RE, Belinson JL. Virilizing malignant lipoid cell tumor producing erythropoietin. Gynecol Oncol 1984; 19: 98. 117. Peng-Hui W, Hsiang-Tai C, Wen-Ling L. Use of a long-acting gonadotropin-releasing hormone agonist for treatment of steroid cell tumors of the ovary. Fertil Steril 1998; 69: 353.
Section 7 : Gynecological Cancers
42
Germ Cell Tumors of the Ovary Daniela E. Matei, Jeanne M. Schilder and Helen Michael
INTRODUCTION Ovarian germ cell tumors (OGCT) account for 2 to 3% of all ovarian cancers and usually occur in young women. The peak age incidence for development of these tumors is the early twenties. Accurate diagnosis, evaluation, and treatment are essential for cure in these women. Significant improvements in the management of OGCTs have been achieved during the past two decades. The leading cause for improved outcome is the development of more effective chemotherapy regimens, based on cisplatin. Other advances in management include a more precise surgical staging system, improved radiographic imaging, more sophisticated pathology techniques, and improved supportive care. Most women with OGCTs are long-term survivors and suffer minimal morbidity from treatment.
HISTORICAL BACKGROUND Germ cell tumors have been described in the medical literature for many years, dating back at least to 1911 when Chenot described a malignant nonepithelial neoplasm of the ovary.1 This particular tumor is now known to be the most common of the germ cell tumors, and was first termed dysgerminoma by Meyer in 1931.2 Several papers over the subsequent years characterized ovarian tumors with similar histological features, leading to their inclusion into a new class called germ cell tumors. Teilum first introduced this concept3 and OGCT classification was refined and formally introduced in 1973 by the World Health Organization (WHO).
PATHOLOGY The current WHO classification of OGCTs includes benign tumors, almost all of which are accounted for by dermoid cysts; malignant tumors arising from constituents of dermoid cysts; and primitive malignant germ cell tumors, which recapitulate normal embryonic and extraembryonic cells and structures. The malignant OGCTs account for 2–3% of all ovarian cancers and are classified as dysgerminoma, yolk sac tumor, embryonal carcinoma, polyembryoma, nongestational
choriocarcinoma, mixed germ cell tumors, and teratomas (immature, mature, and monodermal types).4
Dermoid Cysts Mature cystic teratomas, which account for one-fourth to one-third of all ovarian tumors, are most common in young women but are also found in children and occasionally in elderly women. Teratomas are neoplasms composed of tissue that is derived from two or three embryonic layers. Most often they present as cysts lined by epidermis with skin appendages. The cyst lumen contains sebaceous material and hair. In 10% of cases, dermoid cysts occur bilaterally. In two-thirds of cases, mature elements reflecting differentiation of tissues normally derived from all three embryonic germ layers (ectoderm, mesoderm, endoderm) are present. Any of these constituents has the potential for undergoing benign or malignant neoplastic transformation leading to the formation of a tumor within a tumor. Malignant tumors arising from constituents of dermoid cysts account for 2 to 3% of ovarian cancers. In patients older than 40 years, malignant transformation of dermoid cysts should always be excluded. Malignant lesions in dermoid cysts are rarely encountered before this age. The malignant areas present as small nodules in the wall of the cyst and are recognized after removal of the entire content of the cyst.5 The most common secondary tumor is squamous carcinoma,6,7 which is found in about 1% of dermoid cysts. This tumor appears grossly as an eccentric solid mass in the cyst wall or as a polypoid mass within the lumen. The natural history of squamous carcinoma arising in a dermoid cyst mimics that of squamous carcinoma arising in other primary sites. The spread may be by direct extension or regional lymphatic metastases (para-aortic lymph nodes) and peritoneal dissemination may occur following cyst rupture.5,8 Curative management can be accomplished with aggressive local therapies (surgery, radiation, chemoradiation), in a manner analogous to the management of squamous cancer arising in other anatomic sites. Other tumors arising in dermoid cysts include basal cell carcinomas, sebaceous tumors, malignant melanomas, adenocarcinomas, sarcomas, and neuroectodermal tumors.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
468
GYNECOLOGICAL CANCERS
Endocrine-type tumors include struma ovarii and carcinoid tumors;9 such tumors are malignant in less than 5% of cases. The struma is rarely functional, but carcinoid tumors can induce carcinoid syndrome in one-third of cases. The syndrome is almost always curable by removal of the tumor.
Malignant Ovarian Germ Cell Tumors Dysgerminoma
Dysgerminoma represents the most common ovarian malignant germ cell tumor.10 Five to ten percent are associated with gonadoblastomas in sexually maldeveloped patients, but most occur in normal females who present with abdominal enlargement, a mass, or pain due to torsion. About 10% of dysgerminomas are bilateral on gross examination and another 10% have microscopic involvement of the contralateral ovary. On gross examination, dysgerminomas are usually large, white to gray, fleshy lobulated masses (see Figure 1a) with focal areas of hemorrhage or necrosis on cut section. Abundant hemorrhage, necrosis, or cystic areas should raise the question of a mixed germ cell tumor. Microscopically, dysgerminomas display nests and cords of primitive-appearing germ cells with clear-to-eosinophilic cytoplasm and prominent cytoplasmic borders. Nuclei are enlarged but not pleomorphic. Mitoses may be numerous. Nests of tumor cells are separated by fibrous trabeculae that contain lymphocytes (see Figure 1b). Syncytiotrophoblast cells are present in about 3% of dysgerminomas. Dysgerminomas contain cytoplasmic glycogen, which is seen with a periodic acid-Schiff (PAS) stain. They show diffuse staining for placenta-like alkaline phosphatase (PLAP), usually with membrane accentuation. They stain positively for the c-KIT gene product CD117 and for OCT 3/4, a nuclear transcription factor expressed in human embryonic and stem cells. Syncytiotrophoblast cells present in dysgerminomas stain with human chorionic gonadotropin (HCG). In contrast to choriocarcinoma, the syncytiotrophoblast cells seen in dysgerminoma are not admixed with cytotrophoblast cells.
(a)
(b) Figure 1 Dysgerminoma. The tumor is large, tan, and fleshy without significant hemorrhage or necrosis (a). It is composed of nests of cells with clear cytoplasm; fibrous septae separate the tumor nests (b).
Yolk Sac Tumor
Yolk sac tumor (endodermal sinus tumor) is the second most common OGCT, accounting for 22% of the OGCTs studied at the Armed Forces Institute of Pathology (AFIP).11 These tumors grow very rapidly, often becoming clinically evident in less than 1 month. Ovarian yolk sac tumors are typically large and unilateral, although metastasis to the opposite ovary may occur. On gross examination, the tumor is round, oval, or lobulated, with a smooth external surface, unless rupture or invasion into surrounding structures has occurred. On cut section, these neoplasms are tan or gray, with abundant hemorrhage and necrosis (see Figure 2a). They are partially solid, but they contain cysts that vary in size from a few millimeters to several centimeters in diameter. The cut surface appears mucoid, slimy, or gelatinous. There are many histologic patterns of yolk sac tumor. The most common is the reticular or microcystic pattern, in which a myxoid stroma contains a meshlike network of flattened or cuboidal epithelial cells with varying degrees of
atypia (see Figure 2b). The festoon pattern contains SchillerDuvall bodies composed of a central capillary surrounded by connective tissue and a peripheral layer of columnar cells; this structure is located in a cavity lined by flattened cells (see Figure 2b). When present, Schiller-Duvall bodies are diagnostic of the yolk sac tumor. Other less common variants of yolk sac tumor include hepatoid, polyvesicular vitelline, enteric, endometrioid, solid, parietal, and mesenchymal patterns. Various patterns of yolk sac tumor may contain eosinophilic hyaline globules that are PAS positive and diastase resistant. These globules may be seen in nongerm cell tumors and are not specific for yolk sac tumor. The globules do not contain α-fetoprotein (AFP). Yolk sac tumors generally display cytoplasmic staining for cytokeratin and AFP, although the parietal pattern generally does not stain for AFP. Serum AFP is a useful marker for this tumor, although a negative serum AFP does not exclude disease. Chemotherapy has
GERM CELL TUMORS OF THE OVARY
469
Some embryonal carcinomas may display focal AFP positivity, which may represent partial transformation to yolk sac tumor. Syncytiotrophoblast cells may be present and display HCG staining, but are not accompanied by admixed cytotrophoblast cells. Polyembryoma
Polyembryoma is a very rare malignant ovarian tumor. In the few cases reported, the embryoid bodies have often coexisted with other germ cell tumor types. These unilateral tumors display microscopic evidence of embryoid bodies.14 Well-developed embryoid bodies display a yolk sac and an amniotic cavity that are separated by an embryonic disc. Choriocarcinoma (a)
(b) Figure 2 Yolk sac tumor. This neoplasm displays areas of hemorrhage and cystic degeneration (a). Both microcystic pattern and Schiller-Duvall bodies are present (b).
Primary nongestational ovarian choriocarcinoma is rare.15 It is most often seen as a component of mixed germ cell tumors of the ovary.1,16 These tumors display abundant hemorrhage and necrosis on gross examination (see Figure 3a). Microscopically, they display an admixture of syncytiotrophoblast and cytotrophoblast in a haphazard, often plexiform pattern (see Figure 3b). Syncytiotrophoblastic giant cells have abundant eosinophilic or amphophilic cytoplasm that contains several hyperchromatic nuclei. Cytotrophoblastic cells are round and often have fairly well-defined cell borders and clear or lightly eosinophilic, vacuolated cytoplasm with atypical nuclei. The mitotic rate may be very high. Vascular invasion is often present. Cytotrophoblast cells do not produce HCG. Syncytiotrophoblast is formed from cytotrophoblast and does produce HCG. Choriocarcinoma may also stain for PLAP, epithelial membrane antigen, and carcinoembryonic antigen. Nongestational choriocarcinoma must be distinguished from gestational choriocarcinoma because the former has a worse prognosis and requires more aggressive therapy. Identification of paternal genetic material indicates that the tumor is of gestational origin. Mixed Germ Cell Tumors
resulted in the appearance of AFP-negative parietal yolk sac tumor after eradication of AFP-positive patterns.12 Enteric glands may be carcinoembryonic antigen (CEA) positive. Embryonal Carcinoma
Embryonal carcinoma is rarely seen in the ovary, in contrast to its frequent occurrence in the testis. Only 14 cases were identified in a 30-year period at the AFIP.13 Embryonal carcinoma of the ovary is usually seen as a component of mixed germ cell tumors. It is often associated with areas of hemorrhage and necrosis on gross examination. Microscopically, this tumor is composed of crowded cells that often display overlapping nuclei in paraffin sections. The nuclei are pleomorphic and contain large, prominent nucleoli. The mitotic rate is high. Glandular, solid, and papillary patterns may be seen. Vascular invasion is common in this tumor. Embryonal carcinoma stains positively for PLAP, pancytokeratin (AE1/AE3 and CAM 5.2), CD30 and OCT 3/4.
Mixed germ cell tumors of the ovary contain two or more different types of germ cell neoplasms, either intimately admixed or as separate foci within the tumor.1,16 They are much less common in the ovary than in the testis, and they account for only 8% of malignant OGCTs assessed at the AFIP over a period of 30 years.16 Malignant mixed germ cell tumors are large, unilateral neoplasms, but the gross appearance on cut surface depends on the particular types of germ cell tumor present. The most common germ cell element in the AFIP series was dysgerminoma (80%), followed by yolk sac tumor (70%), teratoma (53%), choriocarcinoma (20%), and embryonal carcinoma (13%).16 The most frequent combination reported has been dysgerminoma and yolk sac tumor. Syncytiotrophoblast may occur either as a component of choriocarcinoma or as isolated cells in other germ cell tumor types. The diagnosis and prognosis of malignant mixed germ cell tumors depends on adequate tumor sampling to reveal small foci of different types of germ cell neoplasms, which may alter therapy and prognosis.
470
GYNECOLOGICAL CANCERS
Figure 4 Immature teratoma. This neoplasm is largely solid and encephaloid in appearance.
(a)
(b) Figure 3 Choriocarcinoma. This tumor displays abundant hemorrhage (a). Microscopically, there is a plexiform pattern consisting of both syncytiotrophoblast and cytotrophoblast cells (b).
display more immaturity, but immature neural tissue occupies no more than the area of three low-power fields in any slide. Immature grade 3 neoplasms have immature neural tissue occupying an area greater than three low-power fields in at least one slide.17 Mature tissue elements are easily identified in grade 1 lesions, are present to a lesser extent in grade 2 neoplasms, and may be absent altogether in grade 3 immature teratomas. The amount of mitotic activity and immature neural tissue with rosettes also increases with the grade. Some authors prefer classifying immature teratomas as either low grade (grade 1) or high grade (grades 2 and 3).18 It is clinically important to distinguish grade 1 tumors from highergrade neoplasms, because the latter require chemotherapy even in patients with stage I disease. In patients whose neoplasm has disseminated beyond the ovary, it is the grade of the tumor metastases that is important in predicting survival and determining treatment. Occasionally, patients may have implants, which contain only mature tissue. Others
Immature Teratomas
Immature teratomas are uncommon. They represent about 3% of all ovarian teratomas, but immature teratomas are the third most common form of malignant OGCTs. Most immature ovarian teratomas are unilateral, although they may metastasize to the opposite ovary and can be associated with mature teratoma in the opposite ovary. They are predominantly solid tumors, but they may contain scattered cystic areas. The cut surface is soft and fleshy and encephaloid in appearance, and areas of hemorrhage and necrosis are common (see Figure 4). Microscopically, these tumors contain a variety of mature and immature tissue elements. The immature elements almost always consist of immature neural tissue in the form of small round blue cells with rosettes and scattered tubules. There is a correlation between the prognosis and the degree of immaturity. Grade 1 neoplasms display some immaturity, but on an aggregate, immature neural tissue do not exceed the area of one low-power field (40x) in any slide. Grade 2 teratomas
A rare entity is the small cell carcinoma of the ovary. This type of tumor is poorly differentiated and carries a poor prognosis. It is encountered in young women and is associated with malignant hypercalcemia. Because pathologically this carcinoma may be derived from primitive germ cells of the ovary, it is frequently treated with a chemotherapy regimen analogous to that used for OGCTs (platinum, etoposidebased).
CLINICAL FEATURES Malignant germ cell tumors of the ovary occur mainly in girls and young women. In the UT MD Anderson Cancer Center (UTMDACC) series, the age of the patients ranged from 6 to 40 years, with a median age of 16–20 years.1 Abdominal pain associated with a palpable pelvic-abdominal mass is present in approximately 85% of patients.1,19 Ten percent of patients present with acute abdominal pain caused by rupture, hemorrhage, or torsion of these tumors. This
GERM CELL TUMORS OF THE OVARY
finding may be more common in patients with endodermal sinus tumor or mixed germ cell tumors and is frequently misdiagnosed as acute appendicitis. Less commonly, patients present with abdominal distension (35%), fever (10%), or vaginal bleeding (10%). Isosexual precocity can occur because of HCG production by the tumor. OGCTs can be diagnosed during pregnancy or in the immediate postpartum period.19 In a case series reported by Gordon, 20 of 158 patients with dysgerminoma were diagnosed during pregnancy or after delivery.20 Nondysgerminomatous ovarian tumors occur less frequently during pregnancy.21 – 24 By and large, patients with ovarian tumors diagnosed during pregnancy can be treated successfully, without compromising the health of the fetus. Surgical resection and chemotherapy have been performed safely in mid and third trimesters. However rapid disease progression or pregnancy termination/miscarriage can occur.25 Germ cell tumors possess the unique property of producing biologic markers. Accurate measurements of HCG and AFP in serum are useful for monitoring the results of treatment and for detecting subclinical recurrences. Endodermal sinus tumor and choriocarcinoma secrete AFP and HCG, respectively. Embryonal carcinoma can secrete both HCG and AFP, but most commonly produces HCG. Mixed tumors may produce either, both, or none of the markers, depending on the type and quantity of elements present. Dysgerminoma is commonly devoid of hormonal production, although a small percentage of tumors produce low levels of HCG if multinucleated syncytiotrophoblastic giant cells are present within the tumor tissue. An elevated level of AFP or high level of HCG (>100 units mL−1 ) denotes the presence of tumor elements other than dysgerminoma. Therapy should be adjusted accordingly. Although immature teratomas are associated with negative markers, a few tumors can produce AFP. Levels of lactic dehydrogenase (LDH) and CA125 can be elevated, however, they are less specific than HCG or AFP.26
SURGICAL EVALUATION OF OVARIAN GERM CELL TUMORS Once an adnexal mass is identified, other potential etiologies have been excluded, and an attempt has been made to classify the mass with tumor markers, surgical evaluation is indicated for both diagnosis and treatment. The staging and extent of surgical extirpation depends on intraoperative findings. Appropriate surgical evaluation requires an adequate vertical midline incision followed by careful inspection to determine the extent of disease. Ascites is present in 20% of patients; this should be evacuated and sent for cytologic evaluation. If no ascites is present, cytologic washings are obtained from the pelvis and bilateral paracolic gutters. Both ovaries should be carefully inspected. Bilaterality ranges from 5% with yolk sac tumors and immature teratomas to 10% with dysgerminomas.20 Surgical staging is similar to that described for epithelial ovarian cancer (see Table 1). Disease is most often grossly confined to one ovary; 60–70% of patients will have stage I disease. In patients with apparent early stage disease, unilateral salpingo-oophorectomy with complete surgical staging is
471
Table 1 FIGO staging of ovarian germ cell tumors.
Stages I IA IB IC
II IIA IIB IIC
III
IIIA IIIB IIIC IV
Description Tumor limited to ovaries Tumor limited to one ovary, no ascites, intact capsule Tumor limited to both ovaries, no ascites, intact capsule Tumor either stage IA or IB, but with ascites present containing malignant cells; or with ovarian capsule involvement or rupture; or with positive peritoneal washings Tumor involving one or both ovaries with extension to the pelvis Extension to uterus or tubes Involvement of both ovaries with pelvic extension Tumor either stage IIA or IIB, but with ascites present containing malignant cells or with ovarian capsule involvement or rupture or with positive peritoneal washings Tumor involving one or both ovaries with tumor implants outside the pelvis or with positive retroperitoneal or inguinal lymph nodes. Superficial liver metastases qualify as stage III Tumor limited to the pelvis with negative nodes but with microscopic seeding of the abdominal-peritoneal surface Negative nodes, tumor implants in the abdominal cavity smaller than 2 cm Positive nodes or tumor implants in the abdominal cavity larger than 2 cm Distant metastases present
FIGO, International Federation of Gynecology and Obstetrics.
indicated. Comprehensive surgical staging should then be undertaken as described in Table 2. In cases of pure dysgerminoma, biopsy of a normal appearing contralateral ovary can be considered, as microscopic involvement can be as high as 10%. Otherwise, ovarian biopsy is not indicated and in fact is discouraged, as biopsy can result in ovarian failure or infertility as a result of adhesion formation. Approximately 10% of patients will have a mature cystic teratoma in the contralateral ovary. These patients can be managed with cystectomy, sparing the normal ovarian tissue.27 Fortunately, preservation of the uterus and contralateral ovary is usually possible.28,29 One notable exception is when an abnormal karyotype is identified preoperatively, or if a dysgenetic gonad is identified on frozen section, in which case, bilateral salpingo-oophorectomy is indicated. Patients with gross metastatic disease should undergo cytoreductive surgery in a manner similar to that performed for epithelial ovarian cancer. Tumor is debulked to reduce volume to the minimal residual disease possible. The surgical approach should be tailored to the individual patient, taking operative risks into account. While the benefit of a maximum cytoreductive surgical effort is well established for patients with epithelial ovarian cancer, such benefit remains to be proven for patients with OGCTs, and less radical surgery with adjuvant chemotherapy often effects cure. If a patient has not completed childbearing, a normal ovary and/or uterus can remain in situ even in the face of extensive metastatic disease elsewhere, as these tumors are typically highly sensitive to chemotherapy, and cure is often achievable.30 Advances in artificial reproductive techniques can allow a patient to conceive via donor eggs if both ovaries are involved, or her genetic offspring can be carried by a surrogate mother if hysterectomy is necessary but one ovary is retained.
472
GYNECOLOGICAL CANCERS
Table 2 Comprehensive surgical staging.
Examination under anesthesia Appropriate surgical incision (usually midline vertical) Cytologic evaluation: ascites or peritoneal washings Pelvic cul-de-sac Bilateral paracolic gutters Left hemidiaphragm Comprehensive intra-abdominal inspection and palpation, including: Small bowel: ileocecal junction to ligament of Treitz Large bowel: ileocecal junction to rectum Peritoneal surfaces and mesentery including diaphragm Solid organs: kidneys, liver, spleen, gall bladder, and pancreas Total abdominal hysterectomya Unilateral or bilateral salpingo-oophorectomya Infracolic omentectomy Appendectomya Pelvic and aortic lymph node biopsies Resection of enlarged or suspicious nodes If no gross abnormalities, lymph node sampling is performed Peritoneal biopsies Pelvic cul-de-sac Bladder Bilateral sidewalls Bilateral paracolic gutters Diaphragm a
Tailor to individual.
Although historically second-look laparotomy (SLL) was once implemented in the assessment of chemotherapy efficacy in patients with epithelial ovarian cancer, a benefit has not been noted in the evaluation of patients with OGCTs. Gershenson et al. found that 52 of 53 patients undergoing SLL were negative for disease. Elevated tumor markers were present in the one patient with disease noted at SLL; furthermore, she responded well to salvage chemotherapy. SLL did not provide benefit in this study in terms of management or prognosis.1 The Gynecologic Oncology Group (GOG) has determined that the only group of patients that derives a clinical benefit from SLL is that with teratoma elements, who initially undergo incomplete resection.31 A well-described phenomenon noted at the time of SLL for OGCTs is called chemotherapeutic retroconversion. Implants noted at SLL often demonstrate evidence of residual mature teratoma.32 – 34 Additional chemotherapy is not indicated for this finding.
CHEMOTHERAPY FOR OVARIAN GERM CELL TUMORS Chemotherapy: from VAC to BEP One of the great successes of cancer treatment in the 70s and 80s has been the development of effective chemotherapy for testicular germ cell tumors.35,36 The lessons learned from prospective, randomized trials in testis cancer have been applied to OGCTs. Currently, the majority of patients with OGCTs are long-term survivors, when treated with cisplatinbased combination chemotherapy. Historically, the first regimen used successfully for women with OGCTs was VAC (vincristine, dactinomycin, and
cyclophosphamide). Although VAC therapy had curative potential for early stage disease, long-term survival was less than 50% in women with advanced disease. In the series from UTMDACC, 86% of patients with stage I tumors were cured with VAC,1 but only 57% of patients with stage II and 50% of patients with stage III tumors achieved longterm control. Two patients with stage IV tumors succumbed to their disease. In a GOG study, 39 of 54 patients with complete surgical resection and 7 of 22 patients with incompletely resected tumors achieved long-term disease control with VAC.37 In this report, 11 of 15 patients with stage III and both patients with stage IV disease failed within 12 months of follow-up. These data suggested that VAC chemotherapy is insufficient for the treatment of advanced stage and/or incompletely resected OGCTs. Along with the development of cisplatin-based regimens in testis cancer, these therapies were tested in women with OGCTs. The vinblastine, bleomycin, and cisplatin (PVB) combination was evaluated prospectively in GOG protocol 45 in patients with previously treated and untreated OGCTs.38 The 4-year overall survival was 70% and 47 of 89 (53%) patients were disease-free at 52 months. Twentynine percent of patients enrolled in this trial had received prior radiation or chemotherapy. Among the 30 patients with nonmeasurable disease and without prior treatment, there were eight treatment failures. Subsequent experience in testicular cancer documented that etoposide is at least equivalent to vinblastine and produces improved survival in patients with high tumor volume.36 Furthermore, the use of etoposide in place of vinblastine led to decreased neurologic toxicity, abdominal pain, and constipation. These observations led to the evaluation of the combination of cisplatin, etoposide, and bleomycin (BEP; see Table 3) in patients with OGCTs. In a series from UTMDACC, long-term remissions were recorded in 25 of 26 patients treated with BEP.1 In a GOG protocol, 91 of 93 patients were disease-free after BEP chemotherapy.30 The inclusion of cisplatin in the treatment of ovarian tumors resulted in a significant improvement in survival and disease control.39 – 41 On the basis of these data, although BEP and VAC have not been compared prospectively in patients with OGCTs, BEP emerged as the preferred regimen. Women with OGCT should receive three to four cycles of BEP chemotherapy after cytoreductive surgery.
Management of Residual or Recurrent Disease The majority of patients with OGCTs are cured with surgery and platinum-based chemotherapy. However, a small percentage of patients have persistent or progressive disease during treatment or recurrence after completion of treatment. Most recurrences occur within 24 months of primary Table 3 BEP regimen.
Cisplatin Etoposide (VP16) Bleomycin
20 mg m−2 100 mg m−2 30 Units
Note: Three to four courses given at 21-day interval.
Day 1 – 5 Day 1 – 5 Weekly
GERM CELL TUMORS OF THE OVARY
treatment. Like in testis cancer, treatment failures are categorized as platinum resistant (progression during or within 4–6 weeks of completing treatment) or platinum sensitive (recurrence beyond 6 weeks from platinum-based therapy).42 Given the high curability rate of OGCTs with primary treatment, the management of recurrent disease represents a complex and often difficult issue, and should be performed in a specialized center. Data to guide the management of patients with recurrent OGCTs are scant and are by and large extrapolated from the treatment of testis cancer patients. Approximately 30% of patients with recurrent platinum-sensitive testis cancer can be salvaged with second-line chemotherapy (VeIP: vinblastine, ifosfamide, platinum).43 In patients with recurrent or persistent testicular germ cell tumors, there is now presumptive but strong evidence that high-dose therapy with carboplatin, etoposide with or without cyclophosphamide or ifosfamide, and stem cell rescue is superior to standard dose salvage therapy.44 – 46 The single most important prognostic factor in patients with testis cancer is whether they are refractory to cisplatin. In patients who are truly cisplatin refractory, the likelihood of long-term survival and cure following high-dose therapy is less than 5% and high-dose therapy is of debatable appropriateness.47 On the other hand, the likelihood of cure with high-dose salvage therapy in patients who relapse from a complete remission after initial therapy is 50%. While this approach has not been and most probably will never be tested prospectively in women with recurrent platinumsensitive OGCTs, because of the small numbers of patients, the concepts are very similar and support the use of high-dose therapy in this setting. Referral to a specialized center for management of recurrent disease is desirable. Patients with platinum-refractory disease cannot be cured. Active agents in this setting include ifosfamide, taxanes, and gemcitabine48 – 50 and referral for treatment with investigational agents is appropriate.
Immediate Toxicity of Chemotherapy Acute adverse effects of chemotherapy can be substantial. Approximately 25% of patients experience febrile neutropenia and require hospitalization and broad-spectrum antibiotics.51 Cisplatin-induced nephrotoxicity can be prevented by adequate hydration and avoidance of nephrotoxic drugs. Bleomycin can cause pulmonary fibrosis.51 The most effective method for monitoring germ cell tumor patients is careful physical examination of the chest. Findings of early bleomycin lung disease are a lag or diminished expansion of one hemi-thorax or fine basilar rales that do not clear with cough. These findings can be subtle but if present, mandate discontinuation of bleomycin. As shown by randomized trials in testis cancer, bleomycin is an important component of the treatment regimen and should not be omitted in the absence of lung toxicity, particularly if only three courses of therapy are given.52,53 Modern antiemetic therapy has greatly reduced chemotherapy-induced emesis, common in the early days of cisplatin use. Patients with advanced OGCTs should receive three to four courses of treatment given in full dose and on schedule. There is presumptive evidence in testis cancer that
473
the timeliness of chemotherapy is associated with outcome. Thus, treatment is given regardless of hematological parameters on the scheduled day of treatment. By following these guidelines and by providing supportive care as indicated, virtually all patients can be treated on schedule, in full or nearly full dose. Chemotherapy-related mortality should be less than 1%.
Late Sequelae of Chemotherapy Given that a high percentage of women with germ cell tumors are cured, attention should be paid to the long-term effects of treatment. A recognized late effect of chemotherapy is the risk for secondary malignancies. Etoposide is associated with the development of acute myelogenous leukemia (AML) with a characteristic chromosomal translocation at the 11q23 locus. This treatment complication occurs within 2–3 years and appears to be dose54,55 and schedule dependent.56 In the GOG protocol testing BEP, there was one case of AML among 91 patients.30 An additional case of lymphoma was diagnosed, yet a correlation between chemotherapy and lymphoproliferative disorders has not been established. Chemotherapy also has long-term effects on gonadal function and leads to sterility.57 – 59 Older age at initiation of therapy, greater cumulative drug dose,60 and longer duration of therapy59 favor premature ovarian dysfunction. However, successful pregnancies after combination chemotherapy have been documented in patients with malignant OGCTs.1,61 – 64 In a review of the UTMDACC series,1 27 (68%) of 40 patients who had retained a normal contralateral ovary and uterus maintained regular menses consistently after completion of chemotherapy and 33 (83%) were having regular menses at the time of follow-up. Twelve patients had successful pregnancies. In a series from Milan, 138 of 196 patients underwent fertility-sparing surgery, and of these, 81 underwent adjuvant chemotherapy.64 After treatment, all but one woman recovered menstrual function and 55 conceptions were recorded. Limited reports are available concerning other late effects of chemotherapy in patients with OGCTs,65,66 but there are several articles on this topic about patients with testicular cancer.67 – 72 In male patients who received cisplatinbased combination regimens, principally PVB, late toxicities include high-tone hearing loss,67 neurotoxicity,67,70,72 Raynaud’s phenomena,68,72 ischemic heart disease,72,73 hypertension,72 renal dysfunction,72 and pulmonary toxicity.69,71 Fortunately, the majority of patients preserve an excellent overall health and functional status.71 The GOG recently completed an analysis evaluating the quality of life and psychosocial characteristics of survivors of OGCTs compared with matched controls. In this analysis, the survivors appeared to be well adjusted, were able to develop strong relationships, and were free of significant depression. The impact on fertility was modest or none, in those patients who underwent fertility-sparing surgeries. Overall, these women appeared to be free of any major physical illnesses at a median follow-up of 10 years (SD Williams, personal communication).
474
GYNECOLOGICAL CANCERS
DYSGERMINOMA Dysgerminoma is the female equivalent of seminoma. This disease differs from nondysgerminomatous tumors in several respects. First, it is more likely to be localized to the ovary at the time of diagnosis (stage I). Bilateral involvement is more common, as it spreads to the retroperitoneal lymph nodes. While less relevant now than before the era of modern chemotherapy, dysgerminoma is very sensitive to radiation.20,74,75
Observation for Stage I Tumors As many as two-thirds of dysgerminoma patients are in stage I at diagnosis.76,77 In the past, most of these women received postoperative radiotherapy. Given that pelvic radiotherapy leads to gonadal dysfunction and sterility, an alternative for low-risk patients is postsurgical clinical surveillance.78 Several case series report that 80–85% of patients with stage IA dysgerminoma are cured with surgery alone (see Table 4). Careful follow-up is required, because 15–25% of patients will have recurrence. However, given the exquisite chemosensitivity, virtually all dysgerminoma patients can be salvaged at the time of recurrence, if adequate follow-up and early detection have been accomplished.
Radiation Therapy In the past, many stage I patients and all patients with higher stage tumors received radiotherapy. Radiation therapy was delivered to the ipsilateral hemipelvis (with shielding of the contralateral ovary and the head of the femur) and to the para-aortic nodes. A single field incorporating these areas was used. In either case, the upper limit of the field was set at T10–T11. The lower limit of the spinal field was at L4–L5 level. For stage III retroperitoneal disease, a prophylactic field including the mediastinum and supraclavicular nodes was included. In the presence of peritoneal involvement, the whole abdomen and pelvis, mediastinum, and supraclavicular nodes were irradiated. The usual prophylactic dose was 30 Gy (7.5–9 Gy/week). For curative irradiation, 35–40 Gy total dose was given and a boost (10 Gy) was delivered to the
involved nodes. Prophylactic mediastinal radiation for highrisk patients used 30 Gy, given 3–6 weeks after completion of irradiation below the diaphragm. Results of radiation therapy were reasonably favorable. DePalo reports that all 13 stage I patients (12 stage IA and one with stage IB) are alive and free of disease with a median follow-up of 77 months.76 The 5-year relapse-free survival for 12 stage III patients was 61.4% and the overall survival was 89.5%. Lawson and Adler reported that 10 of 14 stages I –III patients were alive with a median followup of 54 months.81 Others have reported similar results with overall progression-free rates varying between 70 and 90% when radiotherapy followed surgical resection75,82 (see Table 5). However, despite the remarkable radiosensitivity of dysgerminoma, radiotherapy is rarely performed nowadays, as chemotherapy is equally or more effective, less toxic, and allows preservation of gonadal function.
Chemotherapy Dysgerminoma is very responsive to cisplatin-based chemotherapy, even more than nondysgerminomatous tumors.1,84 A recent analysis of dysgerminoma patients treated on GOG protocols revealed that 19 of 20 patients were disease-free with a median follow-up of 22 months.85 All had stages III or IV disease and most of them had suboptimal (greater than 2 cm) residual tumor. This suggests that nearly all patients with advanced dysgerminoma treated with chemotherapy are durable complete responders. A GOG study showed that carboplatin-etoposide is an alternative regimen for dysgerminoma patients.86 However, at present cisplatin should not be substituted for carboplatin, given the more robust data accumulated from studies on testis cancer, where the two approaches are not equivalent. Considering that patients with stage III dysgerminoma would require extensive radiation and still carry a risk of failure and that such patients probably fare worse with subsequent chemotherapy, it is clear that these patients should be treated primarily with chemotherapy. In summary, the majority of dysgerminoma patients have stage I disease at diagnosis. These patients can usually be treated with unilateral salpingo-oophorectomy and if fertility is an issue, they can be observed carefully with regular pelvic exams, abdominal computerized tomography, and Table 5 Results of radiotherapy in dysgerminoma.
Table 4 Results of clinical surveillance after surgery in patients with stage IA dysgerminoma.
Institution AFIP19 Hopkins20 Mayo Clinic74 Iowa Hospitals77 MD Anderson79 Mount Vernon Hospital80
Period
Progression free/total number (%)
Overall survival/ total number (%)
– 1969 1930 – 1981 1950 – 1984 1935 – 1985 – 1976 1973 – 1995
46/57 (80%) 58/72 (80%) 9/14 (64%) 7/7 (100%) 5/5 (100%) 6/9 (66%)
52/57 (91%) 67/72 (94%) 14/14 (100%) 7/7 (100%) 5/5 (100%) 9/9 (100%)
Reprinted from Principles and Practice of Gynecologic Oncology, Fourth Edition, with permission from Lippincott, Williams & Wilkins.
Institution 19
AFIP Mayo Clinic74 MD Anderson79 Florence82 NCI Milan76 Iowa Hospitals77 Sweden75 Egypt83 Prince of Whales Hospital81
Period
Stages
– 1969 1950 – 1984 – 1976 1960 – 1983 1970 – 1982 1935 – 1985 1927 – 1984 1978 – 1989 1969 – 1983
I – III I – IV I – III IC – III I – III I – III I – IV II – III II – III
Progression free/ total number of patients (%) 12/14 16/20 26/31 21/26 21/25 12/13 49/60 10/15 10/14
(85%) (80%) (84%) (80%) (84%) (92%) (83%) (66%) (72%)
Reprinted from Principles and Practice of Gynecologic Oncology, Fourth Edition, with permission from Lippincott, Williams & Wilkins.
GERM CELL TUMORS OF THE OVARY
tumor markers including LDH. Fifteen to twenty-five percent of patients treated with surveillance will experience a recurrence and will require chemotherapy. In patients with more advanced but resected disease, the risk of recurrence is significant enough to warrant adjuvant treatment. Alternatives are chemotherapy or radiotherapy. For the majority of patients, chemotherapy is the clear choice because of ease of administration, predictable, and minimal toxicity and fertility-sparing properties. Chemotherapy is also recommended for patients with metastatic or incompletely resected tumor and for patients who have recurrence after previous radiotherapy. Radiation might be considered as initial treatment in unusual circumstances, such as in older patients or in those with serious concomitant illness that would preclude the use of systemic chemotherapy.
14.
15. 16.
17.
18.
19. 20.
SUMMARY Virtually all patients with early stage, completely resected OGCTs will survive after careful surgical staging and three courses of adjuvant BEP. Furthermore, 70–80% of patients with incompletely resected or advanced tumors will be longterm survivors. Current and future clinical trials should address the latter group of patients in an effort to improve therapeutic results. Acute toxicity of treatment is relatively modest. An important, but fortunately unusual late complication of treatment is etoposide-induced leukemia. Otherwise, late consequences of chemotherapy are limited. Efforts should concentrate on fertility preservation for patients who desire subsequent pregnancies.
21.
22.
23. 24. 25. 26. 27.
REFERENCES 28. 1. Gershenson DM, et al. Mixed germ cell tumors of the ovary. Obstet Gynecol 1984; 64(2): 200 – 6. 2. Meyer R. The pathology of some special ovarian tumors and their relation to sex characteristics. Am J Obstet Gynecol 1931; 22: 697 – 713. 3. Teilum G. Classification of endodermal sinus tumour (mesoblastoma vitellinum) and so-called “embryonal carcinoma” of the ovary. Acta Pathol Microbiol Scand 1965; 64(4): 407 – 29. 4. Tavassoli FA, Devilee P (eds). Pathology and Genetics of Tumours of the Brease and Female Genital Organs. World Health Organization Classification of Tumours. Lyon, France: IARC Press, 2003. 5. Pins MR, et al. Primary squamous cell carcinoma of the ovary. Report of 37 cases. Am J Surg Pathol 1996; 20(7): 823 – 33. 6. Powell JL, et al. Squamous cell carcinoma arising in a dermoid cyst of the ovary. Gynecol Oncol 2003; 89(3): 526 – 8. 7. Mayer C, Miller DM, Ehlen TG. Peritoneal implantation of squamous cell carcinoma following rupture of a dermoid cyst during laparoscopic removal. Gynecol Oncol 2002; 84(1): 180 – 3. 8. Rose PG, Tak WK, Reale FR. Squamous cell carcinoma arising in a mature cystic teratoma with metastasis to the paraaortic nodes. Gynecol Oncol 1993; 50(1): 131 – 3. 9. Takemori M, et al. Ovarian strumal carcinoid with markedly high serum levels of tumor markers. Gynecol Oncol 1995; 58(2): 266 – 9. 10. Kurman RJ, Norris HJ. Malignant germ cell tumors of the ovary. Hum Pathol 1977; 8(5): 551 – 64. 11. Kurman RJ, Norris HJ. Endodermal sinus tumor of the ovary: a clinical and pathologic analysis of 71 cases. Cancer 1976; 38(6): 2404 – 19. 12. Damjanov I, Amenta PS, Zarghami F. Transformation of an AFPpositive yolk sac carcinoma into an AFP-negative neoplasm. Evidence for in vivo cloning of the human parietal yolk sac carcinoma. Cancer 1984; 53(9): 1902 – 7. 13. Kurman RJ, Norris HJ. Embryonal carcinoma of the ovary: a clinicopathologic entity distinct from endodermal sinus tumor
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
475
resembling embryonal carcinoma of the adult testis. Cancer 1976; 38(6): 2420 – 33. Beck JS, Fulmer HF, Lee ST. Solid malignant ovarian teratoma with “embryoid bodies” and trophoblastic differentiation. J Pathol 1969; 99(1): 67 – 73. Vance RP, Geisinger KR. Pure nongestational choriocarcinoma of the ovary. Report of a case. Cancer 1985; 56(9): 2321 – 5. Kurman RJ, Norris HJ. Malignant mixed germ cell tumors of the ovary. A clinical and pathologic analysis of 30 cases. Obstet Gynecol 1976; 48(5): 579 – 89. Norris HJ, Zirkin HJ, Benson WL. Immature (malignant) teratoma of the ovary: a clinical and pathologic study of 58 cases. Cancer 1976; 37(5): 2359 – 72. O’Connor DM, Norris HJ. The influence of grade on the outcome of stage I ovarian immature (malignant) teratomas and the reproducibility of grading. Int J Gynecol Pathol 1994; 13(4): 283 – 9. Asadourian LA, Taylor HB. Dysgerminoma. An analysis of 105 cases. Obstet Gynecol 1969; 33(3): 370 – 9. Gordon A, Lipton D, Woodruff JD. Dysgerminoma: a review of 158 cases from the Emil Novak Ovarian Tumor Registry. Obstet Gynecol 1981; 58(4): 497 – 504. Christman JE, et al. Delivery of a normal infant following cisplatin, vinblastine, and bleomycin (PVB) chemotherapy for malignant teratoma of the ovary during pregnancy. Gynecol Oncol 1990; 37(2): 292 – 5. Farahmand SM, et al. Ovarian endodermal sinus tumor associated with pregnancy: review of the literature. Gynecol Oncol 1991; 41(2): 156 – 60. Horbelt D, et al. Mixed germ cell malignancy of the ovary concurrent with pregnancy. Obstet Gynecol 1994; 84(4 Pt 2): 662 – 4. Rajendran S, Hollingworth J, Scudamore I. Endodermal sinus tumour of the ovary in pregnancy. Eur J Gynaecol Oncol 1999; 20(4): 272 – 4. Bakri YN, et al. Malignant germ cell tumors of the ovary. Pregnancy considerations. Eur J Obstet Gynecol Reprod Biol 2000; 90(1): 87 – 91. Sekiya S, Seki K, Nagai Y. Rise of serum CA 125 in patients with pure ovarian yolk sac tumors. Int J Gynaecol Obstet 1997; 58(3): 323 – 4. Williams SD, et al. Ovarian germ-cell tumors. In Hoskins WJPCYR (ed) Principles and Practice of Gynecologic Oncology, 2nd ed. Philadelphia, PA: Lippincot-Raven, 1997. S. PE. Surgery of germ cell tumours of the ovary. Forum (Genova) 2000; 10: 355 – 65. Peccatori F, et al. Surgical management of malignant ovarian germcell tumors: 10 years’ experience of 129 patients. Obstet Gynecol 1995; 86(3): 367 – 72. Williams S, et al. Adjuvant therapy of ovarian germ cell tumors with cisplatin, etoposide, and bleomycin: a trial of the Gynecologic Oncology Group. J Clin Oncol 1994; 12(4): 701 – 6. Williams SD, et al. Second-look laparotomy in ovarian germ cell tumors: the gynecologic oncology group experience. Gynecol Oncol 1994; 52(3): 287 – 91. Geisler JP, et al. Growing teratoma syndrome after chemotherapy for germ cell tumors of the ovary. Obstet Gynecol 1994; 84(4 Pt 2): 719 – 21. Itani Y, et al. Growing teratoma syndrome after chemotherapy for a mixed germ cell tumor of the ovary. J Obstet Gynaecol Res 2002; 28(3): 166 – 71. Kattan J, et al. The growing teratoma syndrome: a woman with nonseminomatous germ cell tumor of the ovary. Gynecol Oncol 1993; 49(3): 395 – 9. Einhorn LH, Donohue J. Cis-diamminedichloroplatinum, vinblastine, and bleomycin combination chemotherapy in disseminated testicular cancer. Ann Intern Med 1977; 87(3): 293 – 8. Williams SD, et al. Treatment of disseminated germ-cell tumors with cisplatin, bleomycin, and either vinblastine or etoposide. N Engl J Med 1987; 316(23): 1435 – 40. Slayton RE, et al. Vincristine, dactinomycin, and cyclophosphamide in the treatment of malignant germ cell tumors of the ovary. A Gynecologic Oncology Group Study (a final report). Cancer 1985; 56(2): 243 – 8. Williams SD, et al. Cisplatin, vinblastine, and bleomycin in advanced and recurrent ovarian germ-cell tumors. A trial of the Gynecologic Oncology Group. Ann Intern Med 1989; 111(1): 22 – 7.
476
GYNECOLOGICAL CANCERS
39. Culine S, et al. Cisplatin-based chemotherapy in the management of germ cell tumors of the ovary: The Institut Gustave Roussy Experience. Gynecol Oncol 1997; 64(1): 160 – 5. 40. Segelov E, et al. Cisplatin-based chemotherapy for ovarian germ cell malignancies: the Australian experience. J Clin Oncol 1994; 12(2): 378 – 84. 41. Dimopoulos MA, et al. Favorable outcome of ovarian germ cell malignancies treated with cisplatin or carboplatin-based chemotherapy: a Hellenic Cooperative Oncology Group study. Gynecol Oncol 1998; 70(1): 70 – 4. 42. Loehrer PJ Sr, et al. Salvage therapy in recurrent germ cell cancer: ifosfamide and cisplatin plus either vinblastine or etoposide. Ann Intern Med 1988; 109(7): 540 – 6. 43. Einhorn LH. Salvage therapy for germ cell tumors. Semin Oncol 1994; 21(4Suppl 7): 47 – 51. 44. Broun ER, et al. Early salvage therapy for germ cell cancer using high dose chemotherapy with autologous bone marrow support. Cancer 1994; 73(6): 1716 – 20. 45. Broun ER, et al. Tandem high dose chemotherapy with autologous bone marrow transplantation for initial relapse of testicular germ cell cancer. Cancer 1997; 79(8): 1605 – 10. 46. Lotz JP, et al. High dose chemotherapy with ifosfamide, carboplatin, and etoposide combined with autologous bone marrow transplantation for the treatment of poor-prognosis germ cell tumors and metastatic trophoblastic disease in adults. Cancer 1995; 75(3): 874 – 5. 47. Nichols CR, et al. Dose-intensive chemotherapy in refractory germ cell cancer – a phase I/II trial of high-dose carboplatin and etoposide with autologous bone marrow transplantation. J Clin Oncol 1989; 7(7): 932 – 9. 48. Loehrer PJ Sr, et al. Vinblastine plus ifosfamide plus cisplatin as initial salvage therapy in recurrent germ cell tumor. J Clin Oncol 1998; 16(7): 2500 – 4. 49. Hinton S, et al. Phase II study of paclitaxel plus gemcitabine in refractory germ cell tumors (E9897): a trial of the Eastern Cooperative Oncology Group. J Clin Oncol 2002; 20(7): 1859 – 63. 50. Nichols CR, et al. Salvage chemotherapy for recurrent germ cell cancer. Semin Oncol 1994; 21(5Suppl 12): 102 – 8. 51. Mann JR, et al. The United Kingdom Children’s Cancer Study Group’s second germ cell tumor study: carboplatin, etoposide, and bleomycin are effective treatment for children with malignant extracranial germ cell tumors, with acceptable toxicity. J Clin Oncol 2000; 18(22): 3809 – 18. 52. Loehrer PJ Sr, et al. Importance of bleomycin in favorable-prognosis disseminated germ cell tumors: an Eastern Cooperative Oncology Group trial. J Clin Oncol 1995; 13(2): 470 – 6. 53. de Wit R, et al. Importance of bleomycin in combination chemotherapy for good-prognosis testicular nonseminoma: a randomized study of the European Organization for Research and Treatment of Cancer Genitourinary Tract Cancer Cooperative Group. J Clin Oncol 1997; 15(5): 1837 – 43. 54. Nichols CR, et al. Secondary leukemia associated with a conventional dose of etoposide: review of serial germ cell tumor protocols. J Natl Cancer Inst 1993; 85(1): 36 – 40. 55. Pedersen-Bjergaard J, et al. Increased risk of myelodysplasia and leukaemia after etoposide, cisplatin, and bleomycin for germ-cell tumours. Lancet 1991; 338(8763): 359 – 63. 56. Pui CH, et al. Acute myeloid leukemia in children treated with epipodophyllotoxins for acute lymphoblastic leukemia. N Engl J Med 1991; 325(24): 1682 – 7. 57. Horning SJ, et al. Female reproductive potential after treatment for Hodgkin’s disease. N Engl J Med 1981; 304(23): 1377 – 82. 58. Byrne J, et al. Effects of treatment on fertility in long-term survivors of childhood or adolescent cancer. N Engl J Med 1987; 317(21): 1315 – 21. 59. Siris ES, Leventhal BG, Vaitukaitis JL. Effects of childhood leukemia and chemotherapy on puberty and reproductive function in girls. N Engl J Med 1976; 294(21): 1143 – 6. 60. Nicosia SV, Matus-Ridley M, Meadows AT. Gonadal effects of cancer therapy in girls. Cancer 1985; 55(10): 2364 – 72.
61. Brewer M, et al. Outcome and reproductive function after chemotherapy for ovarian dysgerminoma. J Clin Oncol 1999; 17(9): 2670 – 5. 62. Pektasides D, et al. Fertility after chemotherapy for ovarian germ cell tumours. Br J Obstet Gynaecol 1987; 94(5): 477 – 9. 63. Rustin GJ, et al. Fertility after chemotherapy for male and female germ cell tumours. Int J Androl 1987; 10(1): 389 – 92. 64. Zanetta G, et al. Survival and reproductive function after treatment of malignant germ cell ovarian tumors. J Clin Oncol 2001; 19(4): 1015 – 20. 65. Hale GA, et al. Late effects of treatment for germ cell tumors during childhood and adolescence. J Pediatr Hematol Oncol 1999; 21(2): 115 – 22. 66. Swenson MM, et al. Quality of life after among ovarian germ cell cancer survivors: a narrative analysis. Oncol Nurs Forum 2003; 30(3): 380. 67. Hansen SW, Helweg-Larsen S, Trojaborg W. Long-term neurotoxicity in patients treated with cisplatin, vinblastine, and bleomycin for metastatic germ cell cancer. J Clin Oncol 1989; 7(10): 1457 – 61. 68. Hansen SW, Olsen N. Raynaud’s phenomenon in patients treated with cisplatin, vinblastine, and bleomycin for germ cell cancer: measurement of vasoconstrictor response to cold. J Clin Oncol 1989; 7(7): 940 – 2. 69. Hansen SW, et al. Enhanced pulmonary toxicity in smokers with germcell cancer treated with cis-platinum, vinblastine and bleomycin: a long-term follow-up. Eur J Cancer Clin Oncol 1989; 25(4): 733 – 6. 70. Roth BJ, et al. Cisplatin-based combination chemotherapy for disseminated germ cell tumors: long-term follow-up. J Clin Oncol 1988; 6(8): 1239 – 47. 71. Boyer M, et al. Lack of late toxicity in patients treated with cisplatincontaining combination chemotherapy for metastatic testicular cancer. J Clin Oncol 1990; 8(1): 21 – 6. 72. Stoter G, et al. Ten-year survival and late sequelae in testicular cancer patients treated with cisplatin, vinblastine, and bleomycin. J Clin Oncol 1989; 7(8): 1099 – 104. 73. Nichols CR, et al. No evidence of acute cardiovascular complications of chemotherapy for testicular cancer: an analysis of the Testicular Cancer Intergroup Study. J Clin Oncol 1992; 10(5): 760 – 5. 74. Buskirk SJ, et al. Ovarian dysgerminoma: a retrospective analysis of results of treatment, sites of treatment failure, and radiosensitivity. Mayo Clin Proc 1987; 62(12): 1149 – 57. 75. Bjorkholm E, et al. Dysgerminoma. The Radiumhemmet series 1927 – 1984. Cancer 1990; 65(1): 38 – 44. 76. De Palo G, et al. Germ cell tumors of the ovary: the experience of the National Cancer Institute of Milan. I. Dysgerminoma. Int J Radiat Oncol, Biol, Phys 1987; 13(6): 853 – 60. 77. LaPolla JP, et al. Dysgerminoma of the ovary. Obstet Gynecol 1987; 69(6): 859 – 64. 78. Mitchell MF, et al. The long-term effects of radiation therapy on patients with ovarian dysgerminoma. Cancer 1991; 67(4): 1084 – 90. 79. Krepart G, et al. The treatment for dysgerminoma of the ovary. Cancer 1978; 41(3): 986 – 90. 80. Dark GG, et al. Surveillance policy for stage I ovarian germ cell tumors. J Clin Oncol 1997; 15(2): 620 – 4. 81. Lawson AP, Adler GF. Radiotherapy in the treatment of ovarian dysgerminomas. Int J Radiat Oncol, Biol, Phys 1988; 14(3): 431 – 4. 82. Santoni R, et al. Dysgerminoma of the ovary: a report on 29 patients. Clin Radiol 1987; 38(2): 203 – 6. 83. Zaghloul MS, Khattab TY. Dysgerminoma of the ovary: good prognosis even in advanced stages. Int J Radiat Oncol, Biol, Phys 1992; 24(1): 161 – 5. 84. Culine S, et al. Cisplatin-based chemotherapy in dysgerminoma of the ovary: thirteen-year experience at the Institut Gustave Roussy. Gynecol Oncol 1995; 58(3): 344 – 8. 85. Williams SD, et al. Chemotherapy of advanced dysgerminoma: trials of the Gynecologic Oncology Group. J Clin Oncol 1991; 9(11): 1950 – 5. 86. Williams SD, et al. Adjuvant therapy of completely resected dysgerminoma with carboplatin and etoposide: a trial of the Gynecologic Oncology Group. Gynecol Oncol 2004; 95(3): 496 – 9.
Section 7 : Gynecological Cancers
43
Fallopian Tube Cancer Destin Black and Richard R. Barakat
Primary malignant tumors of the fallopian tube account for 0.3 to 1.0% of all gynecologic cancers.1 – 5 This number may be higher as some tumors are mistaken for advanced ovarian malignancies. The majority of patients are 40 to 60 years of age with a mean age of 55,6 although it has been reported to occur in patients as young as 18 years of age.1 This rarity has precluded prospective randomized trials and hindered our knowledge of this disease. Our current understanding of fallopian tube cancer consists of relatively small retrospective series with fewer than 2000 cases reported in literature.
Anatomy The fallopian tube extends from the posterosuperior aspect of the uterine fundus to the ovary. At the ovary, the tube is composed of approximately 25 irregular fingerlike extensions called fimbriae. The fimbriae attach to the infundibulum, which is about 1 cm long and 1 cm in diameter. The infundibulum narrows gradually to about 4 mm in diameter and merges medially to the ampullary portion of the tube, which is the widest and longest section. At a point characterized by a thickening of the muscular wall, the isthmic portion begins and extends 2 cm to the uterus. Within the myometrium, the tube communicates with the endometrial cavity at the uterotubal junction.7 The fallopian tube has an internal mucosal layer, an intermediate muscular layer, and an external serosal layer. The serosa is lined by mesothelial cells that are continuous with the serosa covering the uterus. The epithelial layer of the mucosa is composed of three cell types: ciliated, secretory, and intercalary.8 The arterial blood supply originates from the tubal branch of the uterine artery and the tubal branch of the ovarian artery, which anastomose within the mesosalpinx. Venous drainage consists of anastomosing tubal branches of the ovarian and uterine veins. The lymphatics accompany the tubal vessels draining into the para-aortic and pelvic lymph nodes.
Staging Because of the clinical, therapeutic, and prognostic similarities with ovarian cancer, Dodson et al.,9 in 1970, proposed
applying the ovarian staging system to tubal carcinoma. In 1991, the International Federation of Gynecology and Obstetrics (FIGO) Committee established an official staging system for fallopian tube cancer (see Table 1), which is similar to that of ovarian cancer. In a review of 558 patients, 33, 33, and 33% were stages I, II, and III –IV, respectively.10 These statistics are more favorable than ovarian carcinoma patient statistics; 70% of ovarian carcinoma patients have stages III and IV disease. In 1967, Erez et al.11 suggested a staging system based on the Dukes’ colonic cancer staging, which used both depth of invasion and distant disease. This staging system would have been based on tumor penetration through the layers of the tube. In data from 18 institutions, pathologic stage was closely related to survival in 76 patients with tubal cancer.12 More recently, Peters et al.6 analyzed stage I patients and found a 50% depth of tubal muscularis invasion to be the only significant prognostic variable. However, other authors have not shown the depth of tubal invasion to be prognostic of survival.13,14
PRIMARY ADENOCARCINOMA Biology and Epidemiology The etiology of fallopian tube cancer remains poorly understood. A history of infertility is common; one study reported a 40% incidence of nulliparity in a series of 47 patients.15 Chronic inflammation and tuberculous salpingitis have also been suggested as possible predisposing factors. Although chronic inflammation often presents with fallopian tube carcinoma, it is doubtful that it is a causative factor. Chronic salpingitis usually involves both tubes, while the inflammatory response associated with fallopian tube cancer is usually limited to the tube involved with the carcinoma. In addition, tubal carcinoma occurs in postmenopausal women, a group with a low prevalence of chronic salpingitis. In Sedlis’16 review, the rate of pelvic tuberculosis in patients with fallopian tube cancer was not higher than the rate of pelvic tuberculosis in the general population. The rarity of this malignancy makes any of these relatively common conditions as unlikely etiologic factors. Alterations of the p53 tumor suppressor gene have been reported in 59 to 81% of cases.17 – 20 Although p53 mutations
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
478
GYNECOLOGICAL CANCERS
Table 1 FIGO fallopian tube staging.
Stage 0 Stage I Stage IA
Stage IB
Stage IC
Stage II Stage IIA Stage IIB Stage IIC
Stage III
Stage IIIA
Stage IIIB
Stage IIIC
Stage IV
Carcinoma in situ (limited to tubal mucosa) Growth limited to the fallopian tubes Growth is limited to one tube with extension into the submucosa and/or muscularis but not penetrating the serosal surface; no ascites Growth is limited to both tubes with extension into the submucosa and/or muscularis but not penetrating the serosal surface; no ascites Tumor either stage IA or IB with tumor extension through or onto the tubal serosa; or with ascites present containing malignant cells or with positive peritoneal washings Growth involving one or both fallopian tubes with pelvic extension Extension and/or metastasis to the uterus and/or ovaries Extension to other pelvic tissues Tumor either stage IIA or IIB and with ascites present containing malignant cells or with positive peritoneal washings Tumor involves one or both fallopian tubes with peritoneal implants outside of the pelvis and/or positive retroperitoneal or inguinal nodes. Superficial liver metastases equals stage III. Tumor appears limited to the true pelvis but with histologically proven malignant extension to the small bowel or omentum Tumor is grossly limited to the true pelvis with negative nodes but with histologically confirmed microscopic seeding of abdominal peritoneal surfaces Tumor involving one or both tubes with histologically confirmed implants of abdominal peritoneal surfaces, none exceeding 2 cm in diameter. Lymph nodes are negative Abdominal implants greater than 2 cm in diameter and/or positive retroperitoneal or inguinal nodes Growth involving one or both fallopian tubes with distant metastases. If pleural effusion is present, there must be positive cytology to be stage IV. Parenchymal liver metastases equals stage IV
Note: Staging for fallopian tube is by the surgical pathological system. Operative findings designating stage are determined prior to tumor debulking.
were seen in all stages of tubal carcinoma, patients with p53 mutations had a significantly shorter survival rate compared to patients without mutations. There is molecular and population-based evidence that fallopian tube cancer may be a component of the BRCArelated breast and ovarian cancer syndrome.21 – 25 Aziz24 and colleagues reported that BRCA mutations accounted for 16% of fallopian tube cancers in an unselected population. Levine et al.26 found germline BRCA mutations in 17% of Ashkenazi patients with fallopian tube cancer.
Pathology The rarity of fallopian tube cancer occurrence is partially related to the physicians’ convention of attributing tubal carcinoma involving the ovary or endometrium as an ovarian or endometrial primary.
Figure 1 Carcinoma of the fallopian tube. This intraoperative photograph depicts the fusiform dilation of the affected tube.
Macroscopic Appearance The tube is usually enlarged by the growth of the intraluminal tumor and appears fusiform (see Figure 1). In approximately 50% of cases, the fimbriated end of the tube is occluded with the development of pyosalpinx or hematosalpinx. On opening the tube, the lumen is occupied by a solid mass, frequently with hemorrhagic and necrotic areas. The tumor may arise from any portion of the tube, but it most often originates in the ampullary portion. Fallopian tube cancer affects both the right and left tubes with similar frequency and is bilateral in 10 to 26% of cases.1,12,13,16,27 In early stages, this bilaterality may represent an independent occurrence in both tubes if the intervening endometrium is free of cancer.16 With more advanced cases, the tumor penetrates the tubal serosa and may involve the ovaries and uterus or other pelvic or abdominal organs.
Spread Pattern The disease reaches the peritoneal cavity and its viscera through the tubal fimbria or through transmural invasion of the tubal wall. The mode of metastasis according to most clinical studies is intraperitoneal via the tubal ostia. Sedlis16 found the peritoneum to be the most frequent site of metastasis followed by the ovaries and the uterus. Evaluation for nodal involvement has not been done routinely by most authors. Tamimi et al.28 reported a 33% frequency of para-aortic nodal metastasis in a series of 15 patients. Gadducci et al.29 found histologically proven metastatic nodes in 50% of 22 patients submitted to paraaortic node dissection and in 24.2% of 33 patients submitted to pelvic node dissection. Recently, Deffieux et al.30 reported that the left para-aortic chain above the level of the inferior mesenteric artery is the most frequently involved nodal package. In addition, nodal metastasis has been found in patients who have had no other evidence of disease.28,31 This potential for nodal metastasis may explain the poor survival rate even when the disease is apparently limited to the tube. Nodal involvement has been shown to negatively
FALLOPIAN TUBE CANCER
impact survival.31,32 Klein et al.33 reported improved median survival when radical lymphadenectomy was performed at the time of initial staging. Tamimi et al.28 and Asmussen et al.14 found histologic evidence of vascular-lymphatic space involvement to be associated with nodal metastasis and to be a poor prognostic factor in survival.
Histology The most frequent histologic type is adenocarcinoma, similar to the ovarian serous variety. Less common histologic types include clear cell carcinoma, endometrioid carcinoma, adenosquamous carcinoma, squamous cell carcinoma, sarcoma, choriocarcinoma, and malignant teratoma. The histologic differential diagnosis of primary malignant tumors of the fallopian tube should include a wide variety of benign tumors and the possibility of metastasis from other primary sites. Various criteria to determine the definite diagnosis of primary fallopian tube carcinoma have been suggested. In 1949, Finn and Javert34 proposed the following criteria: Gross criteria 1. The tubes, at least in the distal portion, are abnormal. The fimbriated ends may be dilated and occluded, resembling chronic salpingitis. 2. There is a papillary growth in the endosalpinx. 3. The uterus and ovaries are either grossly normal or affected by a lesion other than cancer. Microscopic criteria 1. The epithelium of the endosalpinx is replaced in whole or in part by adenocarcinoma, and the histologic character of the cells resembles the epithelium of the endosalpinx (see Figure 2). 2. The endometrium and ovaries are normal or contain a malignant lesion that is secondary to a tubal primary in its size, distribution, and histologic appearance. 3. Tuberculosis has been clearly excluded.
479
In 1950, Hu et al.35 proposed additional criteria: 1. Grossly, the main tumor is in the tube. 2. Microscopically, the mucosa should be chiefly involved and show a papillary pattern. 3. If the tubal wall is found to be involved to a great extent, the transition between benign and malignant tubal epithelium should be demonstrable. More recently, gynecologic pathologists have thoroughly evaluated fallopian tubes removed at exploration for abdominal carcinomatosis. The transformation of in situ to invasive carcinoma in the tube has been identified and considered diagnostic of primary fallopian tube malignancy. Numerous studies of fallopian tube cancer have also noted multifocal upper genital tract tumors.4,6,13,36 In one study, 37% (24 of 64 patients) of patients had multiple primary tumors occurring in the ovary (31%), uterus (11%), and cervix (3%).13 Field neoplastic changes to the M¨ullerian epithelium were proposed to explain the existence of multiple upper genital tract neoplasias. Bannatyne and Russell37 reported seven cases of in situ or invasive tubal carcinoma after reviewing 251 cases of epithelial ovarian cancer. These authors stress that careful sectioning of the tube may identify tubal neoplasia in 5 to 10% of ovarian tumors. In one series, among 1592 ovarian cancers pathologically evaluated over the study period, tubal neoplasia was present in 1.3% of cases.13 Adenocarcinoma in situ has been described as an entity, and some authors do not distinguish it from adenomatous hyperplasia; it is usually focal, containing abnormal mitotic figures and nuclear pleomorphism with large nucleoli. Hu et al.35 divided fallopian tumors into three histologic classifications: grade I, papillary lesions; grade II, papillaryalveolar lesions; grade III, alveolar-medullary lesions. Most pathologists no longer use this system, and the grade of the tumor is classified as well, moderately, or poorly differentiated. Recently, Uehira et al.38 have described a transitional cell pattern. The disease-free interval was markedly improved for patients with a transitional cell pattern compared to those with a nontransitional cell pattern. The authors suggest that this may be the result of an improved response to chemotherapy, as has been demonstrated for transitional cell ovarian carcinoma.39 Endometrioid fallopian tube carcinomas have also been identified as a less invasive and less malignant histologic type.40
Clinical Presentation and Diagnostic Considerations Symptoms and Signs
Figure 2 Carcinoma of the fallopian tube. In this photomicrograph, the epithelium of the endosalpinx is replaced by adenocarcinoma.
The most common presenting symptoms are vaginal bleeding, abdominal pain, and watery discharge. The abdominal pain is classically colicky in nature due to the peristaltic activity of the fallopian tube. A dull, constant pain may occur due to chronic distention of the tubal wall and serosa. Latzko41 described the classic syndrome of “hydrops tubae profluens” in 1916. This is characterized by an adnexal mass and colicky lower abdominal pain that is relieved by the discharge of copious serous fluid from the vagina. Although said to be pathognomonic, it is uncommonly encountered.
480
GYNECOLOGICAL CANCERS
The most common finding on physical examination is an elongated pelvic mass. Ascites may also be present. Rarely, watery vaginal discharge or bleeding after hysterectomy has resulted in a diagnosis of fallopian tube carcinoma.42,43 Diagnostic Tests
Pelvic imaging studies usually demonstrate an adnexal mass that is cystic, complex, or solid in character. No specific pattern has been defined that could differentiate a tubal neoplasm from hydrosalpinx or pyosalpinx. Additionally, this finding is usually interpreted to be an ovarian neoplasm, which is much more common. Computed tomography or magnetic resonance imaging may be helpful for evaluating spread to other intra-abdominal or retroperitoneal structures. Positron-emission tomography with fluorine-18-2deoxyglucose has been reported to correlate with findings of recurrent disease.44 Although an elevated CA-125 is not diagnostic of fallopian tube cancer, more than 80% of patients have increased CA-125 levels and 87% of tumors stain for CA-125.45,46 Diagnosis
Fallopian tube cancer diagnosis is seldom made prior to surgery. In one review of 780 patients, only 10 (1.3%) were diagnosed preoperatively.47 Because of the rarity of the condition, the presence of a pelvic mass in a perimenopausal or postmenopausal woman usually suggests a primary ovarian neoplasm. Positive Papanicolaou smears have been reported in approximately 10% of cases.27,36,48 Patients with recurrent postmenopausal vaginal bleeding or an abnormal Papanicolaou smear (and for whom cervical and endometrial cancer have been ruled out by negative dilatation and curettage) should be suspected of having a tubal carcinoma. Laparoscopy may be helpful if the diagnosis is suspected and no pelvic mass is found by less invasive methods. Rarely does fallopian tube carcinoma enter the differential diagnosis of pelvic masses preoperatively. Intraoperatively, it should be distinguished from benign conditions that enlarge and affect the fallopian tubes such as endometriosis, ectopic pregnancy, hydrosalpinx, and tuboovarian abscess.
Treatment Surgery
Surgical therapy for fallopian tube cancer should be the same as it is for ovarian cancer. In cases grossly confined to the tube, a careful staging procedure should be performed that includes peritoneal washings and systematic inspection and palpation of the peritoneal surfaces. Omentectomy, bilateral pelvic and para-aortic lymphadenectomy, and peritoneal biopsies should be performed. Maxson et al.49 reported two of five patients who had nodal sampling at primary surgery that were positive. A limited number of patients with stage I disease have been treated with surgery alone without evidence of recurrence.13,50 However, one study examining routine lymphadenectomy found no cases of nodal metastasis in a small series of patients with disease grossly confined to the adnexa.51 Positive peritoneal washings carry prognostic
significance, with a 5-year survival rate of 20% compared to 67% for cases with negative peritoneal washings.15 If extratubal spread is found at the time of surgery, a maximal tumor reductive effort should be performed with the main objective of leaving minimal (<1 cm) residual disease. Significant improvement in survival for patients with residual tumors less than 1 cm compared to those with larger residual tumor has been demonstrated.6,15,52,53 Barakat et al.53 evaluated 38 patients with stages II, III, or IV disease and found a 5-year survival rate of 83% for patients with no residual disease compared to 29% for those with gross residual disease. Radiotherapy
The efficacy of radiation therapy for tubal carcinoma is difficult to determine from the current literature due to the lack of uniformity in staging criteria, treatment fields, dosage, fraction size, and type of radiation employed. A previous report9 showed that tubal carcinoma is radiosensitive. Phelps and Chapman3 reported survival in eight of nine stages I and II patients treated with both radioisotope and pelvic radiation (n = 5) or pelvic radiation alone (n = 3). Podratz et al.15 reported four of six patients developing recurrence after pelvic radiation for stage IA, IB, and IIA disease. The use of radiation therapy directed solely to the pelvis is of doubtful benefit owing to the potential of the disease for intraperitoneal dissemination. As in ovarian cancer, the entire peritoneal cavity may be at risk, and curability is limited by our inability to deliver therapeutic doses to the entire abdomen. Brown et al.54 reported a patient with stage III disease and long-term survival after abdominopelvic radiation. Again, in the series reported by Podratz et al.,15 five of five patients with stages IC, IIB, or III disease who received whole-abdominal radiation recurred. Recently, Kojs et al.55 reported 32 patients treated with whole-abdominal radiation. Five-year survival was achieved in 10 of 13 (76.9%) stage I patients, 5 of 9 (55.6%) stage II patients, and 2 of 10 (20%) stage III patients. Whole-abdominal radiation was ineffective in the setting of gross (>2 cm) residual disease. Few studies have used radioisotopes alone for tubal carcinoma. In one study, 4 early stages I or II patients received adjuvant 32 P, and three patients developed recurrence.13 Schray et al.56 reported two patients with stages I and II disease treated with 15 mCi of 32 P without recurrence. As in ovarian carcinoma, a subset of surgically staged patients without metastasis may benefit from radioisotope therapy.
Chemotherapy Hormonal Therapy
The tubal epithelium is hormonally sensitive. Both estrogen and progesterone receptors have been identified in fallopian tube carcinomas.54,57,58 Johnston58 reported a patient with stage I disease treated by surgery and adjuvant progesterone without recurrence at 29 months. However, numerous authors have reported no response to hormonal therapy.9,13,27,59,60 Although responses were noted with cytotoxic regimes that included progesterone,15,27,61,62 the role of progesterone in these regimens is unclear.
FALLOPIAN TUBE CANCER
481
Single-Agent Chemotherapy
Prognosis
A variety of alkylating agents including nitrogen mustard, thio-triethylenephosphamide (TEPA), chlorambucil, cyclophosphamide, and melphalan have been utilized for fallopian tube carcinoma.61,63 Favorable responses have been seen with cisplatin or doxorubicin used as single agents.13,59 More recently, paclitaxel and liposomal doxorubicin has demonstrated response in platinum-resistant disease.64 – 67 There is modest activity using weekly docetaxel in heavily pretreated, platinum-resistant patients.68,69
The most important prognostic factor that correlates with survival is the stage of the disease. Most authors have found significantly different survival rates between patients with disease confined to the pelvis and those with disease beyond the pelvis.6,15,36 Survival figures range from 40 to 60% for localized disease (stages I and II) and from 0 to 16% for more advanced disease (stages III and IV).4,36,50,59,74 In a multicenter retrospective study of 68 patients with stages I and II fallopian tube carcinomas, patients with grade I tumors had a significantly longer survival than patients with grade II or grade III tumors.5 Approximately 15% of patients who undergo a negative second-look laparotomy develop recurrent disease. In many cases, recurrences after negative second-look laparotomy have been at distant sites such as the supraclavicular nodes lungs, brain, kidneys, and axilla.50,54,76
Combination Chemotherapy
Although cisplatin-based combination chemotherapy has been reported in fallopian tube cancers since 1980,62 the cumulative experience is limited (see Table 2). The overall response rate is 61%, with 50% complete response rate. Peters et al.60 reported 46 patients evaluable for chemotherapy response. The response rate for 12 patients receiving cisplatin-containing multiagent chemotherapy was 81%, 75% of which were complete responses. This was significantly different than the response rates seen with multiagent chemotherapy (29%) or single-agent therapy (9%) without cisplatin. Barakat et al.53 reported 38 patients treated with cisplatin, doxorubicin, and cyclophosphamide (n = 24) or cisplatin and cyclophosphamide (n = 14).53 No difference in survival was noted, questioning the role of doxorubicin. Cormio et al.70 also reported 38 patients treated with cisplatin, doxorubicin, and cyclophosphamide. In their study, the median survival was 38 months, with a 5-year survival rate of 35%. Pectasides et al.71 reported 14 patients treated with platinum-based combination therapy, four of whom received carboplatin with a similar response rate. Other examples of response to carboplatin have also been reported.72 Recently, Gemignani et al.73 reported the use of paclitaxel-based chemotherapy after initial surgery in 24 patients with primary fallopian tube adenocarcinoma. The overall median progression-free survival was 27 months for the entire group. The 3-year progression-free survival rate was 67% for optimally cytoreduced compared with 45% in the suboptimally debulked group. Based on randomized trials in ovarian cancer, current therapy should consist of a combination of paclitaxel and a platinum compound. Table 2 Platinum-based therapy for fallopian tube carcinoma.
N Jacobs et al.72 Maxson et al.49 Peters et al.60 Rose et al.13 Morris et al.73 Muntz et al.74 Barakat et al.53 Pectasides et al.75 Cormio et al.76 Total a b c
9 12 16 14 9a 7 26 11 38 144
Evaluated at second-look laparotomy. Partial response. Complete response.
PRb
CRc
Total
– 2 1 1 4 2 – 2 4 16
4 9 12 2 1 3 11 8 22 72
4 11 13 3 5 5 11 10 28 88
Response to Therapy Like ovarian cancer, in the absence of clinically evident or progressive disease, disease status is difficult to assess. Response to therapy and recurrence have been correlated with CA-125 levels.75 Second-look laparotomy has been shown to be of prognostic importance in ovarian cancer.77 Because of the rarity of fallopian tube cancer, second-look laparotomy was not reported in this disease until 1980.62 Overall, 64% of patients undergoing the procedure have no evidence of disease. The likelihood of having a negative second look is related to the amount of residual tumor after initial surgery.74,78 – 81 Nodal evaluation is essential, and para-aortic nodal metastases have been detected as the only evidence of persistent disease at second-look laparotomy.28 Survival for the patients who were pathologically diseasefree was significantly different than those who were clinically disease-free.13,81 Second-look laparoscopy is a less invasive means to determine disease status, and, if positive, may be useful. However, in ovarian carcinoma the procedure is less sensitive and, if negative, some authors recommend confirmation by second-look laparotomy.82,83
SARCOMAS Malignant mixed mesodermal tumors of the fallopian tube are uncommon with slightly more than 50 cases reported.84 Although uncommon, they comprise a greater percentage of fallopian cancers than does ovarian sarcoma compared to ovarian carcinoma. The mean age at diagnosis, clinical features, stage at presentation, and spread pattern do not appear different from the more commonly encountered adenocarcinoma. An equal number of homologous and heterologous fallopian tube tumors have been reported.85 The diagnosis is usually not established until the final pathology is completed. The prognosis of sarcoma patients is poor: most patients survive less than 2 years, although long-term survival has been reported.13,86 – 88 Muntz et al.87 reported improved survival when the disease is confined to the muscularis. Carlson et al.88 reviewed 35 cases with sufficient treatment and follow-up data. Nine patients (26%) were disease-free after 36 months. In each case disease was limited to the
482
GYNECOLOGICAL CANCERS
pelvis, all disease was resected and postoperative treatment was employed (chemotherapy and radiation therapy, n = 5; chemotherapy alone, n = 2; and radiation therapy alone, n = 2). In another review of 13 survivors, 9 had stage I, 1 had stage II, and 3 had stage III disease.85 As in ovarian carcinoma, negative second-look laparotomy has been associated with long-term survival.13,85 – 87
TROPHOBLASTIC TUMORS Tubal molar pregnancy occurs in 1 per 5333 ectopic gestations or 1 per 1.6 million normal intrauterine pregnancies.89 Primary choriocarcinoma of the tube is an even rarer entity and may be gestational or nongestational. Choriocarcinoma involving the tube should be distinguished from a primary intrauterine tumor or a malignant ovarian germ cell tumor. Preoperatively, the diagnosis mimics an ectopic pregnancy. A study from the New England Trophoblastic Disease Center reported 16 cases of tubal gestational trophoblastic disease (GTD).90 Tubal GTD accounted for 0.8% of GTD cases managed at the referral center. The frequency of partial moles, complete moles, and choriocarcinoma were similar; 31, 31, and 38%, respectively. None of the patients presented with symptoms of hyperemesis, toxemia, theca lutein cysts, hyperthyroidism, respiratory insufficiency, or markedly elevated human chorionic gonadotropin (hCG) (>30000 U ml−1 ). The surgical approach may be conservative (i.e. unilateral adnexectomy). All of the partial and four of the five complete molar pregnancies reported by the New England Trophoblastic Disease Center responded to partial or complete salpingooophorectomy. One patient with a complete molar pregnancy developed metastatic disease requiring chemotherapy. Pregnancy after conservative therapy consisting of unilateral salpingo-oophorectomy and chemotherapy has been reported.91 Chemotherapy is an essential component in the management of tubal choriocarcinoma. Ober and Maier92 reviewed 76 cases of tubal choriocarcinoma, and found that of 59 patients, 46 treated in the prechemotherapy era died in contrast to 1 of 17 diagnosed in the postchemotherapy era. Treatment should be monitored by serum hCG titers. Nongestational choriocarcinoma has a poorer prognosis. Nongestational choriocarcinoma can be diagnosed only if germ cell elements other than choriocarcinoma are present.
METASTATIC TUMORS Approximately 80% of tubal malignancies are metastatic from other sites, most commonly from the ovary and endometrium.93,94 Extragenital primary cancers metastasizing to the tube are much rarer, and other primary sites include the breast and gastrointestinal tract. With metastatic disease to the tube, the mucosa is intact. There is serosal involvement, or nests of metastatic deposits are seen in the lymphatics underneath the epithelium.
REFERENCES 1. Hanton EM, et al. Primary carcinoma of the fallopian tube. Am J Obstet Gynecol 1966; 94(6): 832 – 9.
2. Momtazee S, Kempson RL. Primary adenocarcinoma of the fallopian tube. Obstet Gynecol 1968; 32(5): 649 – 56. 3. Phelps HM, Chapman KE. Role of radiation therapy in treatment of primary carcinoma of the uterine tube. Obstet Gynecol 1974; 43(5): 669 – 73. 4. Roberts JA, Lifshitz S. Primary adenocarcinoma of the fallopian tube. Gynecol Oncol 1982; 13(3): 301 – 8. 5. Rosen AC, et al. A comparative analysis of management and prognosis in stage I and II fallopian tube carcinoma and epithelial ovarian cancer. Br J Cancer 1994; 69(3): 577 – 9. 6. Peters WA III, et al. Prognostic features of carcinoma of the fallopian tube. Obstet Gynecol 1988; 71(5): 757 – 62. 7. James E, Wheeler MD. Disease of the Fallopian Tube, 4th ed. SpringerVerlag, 1994. 8. Pauerstein CJ, Woodruff JD. The role of the “indifferent” cells of the tubal epithelium. Am J Obstet Gynecol 1967; 98: 121 – 5. 9. Dodson MG, Ford JH Jr, Averette HE. Clinical aspects of fallopian tube carcinoma. Obstet Gynecol 1970; 36(6): 935 – 9. 10. Markman M, et al. Carcinoma of the Fallopian Tube, 4th ed. Philadelphia, Pennsylvania: Lippincott, 2005. 11. Erez S, Kaplan AL, Wall JA. Clinical staging of carcinoma of the uterine tube. Obstet Gynecol 1967; 30(4): 547 – 50. 12. Schiller HM, Silverberg SG. Staging and prognosis in primary carcinoma of the fallopian tube. Cancer 1971; 28(2): 389 – 95. 13. Rose PG, Piver MS, Tsukada Y. Fallopian tube cancer. The Roswell Park experience. Cancer 1990; 66(12): 2661 – 7. 14. Asmussen M, et al. Primary adenocarcinoma localized to the fallopian tubes: report on 33 cases. Gynecol Oncol 1988; 30(2): 183 – 6. 15. Podratz KC, et al. Jr Primary carcinoma of the fallopian tube. Am J Obstet Gynecol 1986; 154(6): 1319 – 26. 16. Sedlis A. Primary carcinoma of the fallopian tube. Obstet Gynecol Surv 1961; 16: 209 – 26. 17. Lacy MQ, et al. c-erbB-2 and p53 expression in fallopian tube carcinoma. Cancer 1995; 75(12): 2891 – 6. 18. Zheng W, et al. Early occurrence and prognostic significance of p53 alteration in primary carcinoma of the fallopian tube. Gynecol Oncol 1997; 64(1): 38 – 48. 19. Rosen AC, et al. p53 expression in fallopian tube carcinomas. Cancer Lett 2000; 156(1): 1 – 7. 20. Chung TK, et al. Overexpression of p53 and HER-2/neu and c-myc in primary fallopian tube carcinoma. Gynecol Obstet Invest 2000; 49(1): 47 – 51. 21. Bandera CA, et al. BRCA1 gene mutations in women with papillary serous carcinoma of the peritoneum. Obstet Gynecol 1998; 92(4 Pt 1): 596 – 600. 22. Schorge JO, et al. BRCA1-related papillary serous carcinoma of the peritoneum has a unique molecular pathogenesis. Cancer Res 2000; 60(5): 1361 – 4. 23. Zweemer RP, et al. Molecular evidence linking primary cancer of the fallopian tube to BRCA1 germline mutations. Gynecol Oncol 2000; 76(1): 45 – 50. 24. Aziz S, et al. A genetic epidemiological study of carcinoma of the fallopian tube. Gynecol Oncol 2001; 80(3): 341 – 5. 25. Menczer J, et al. Frequency of BRCA mutations in primary peritoneal carcinoma in Israeli Jewish women. Gynecol Oncol 2003; 88(1): 58 – 61. 26. Levine DA, et al. Fallopian tube and primary peritoneal carcinomas associated with BRCA mutations. J Clin Oncol 2003; 21(22): 4222 – 7. 27. Yoonessi M. Carcinoma of the fallopian tube. Obstet Gynecol Surv 1979; 34(4): 257 – 70. 28. Tamimi HK, Figge DC. Adenocarcinoma of the uterine tube: potential for lymph node metastases. Am J Obstet Gynecol 1981; 141: 132 – 7. 29. Gadducci A, et al. Analysis of treatment failures and survival of patients with fallopian tube carcinoma: a cooperation task force (CTF) study. Gynecol Oncol 2001; 81(2): 150 – 9. 30. Deffieux X, et al. Anatomy of pelvic and para-aortic nodal spread in patients with primary fallopian tube carcinoma. J Am Coll Surg 2005; 200(1): 45 – 8. 31. di Re E, et al. Fallopian tube cancer: incidence and role of lymphatic spread. Gynecol Oncol 1996; 62(2): 199 – 202. 32. Cormio G, et al. Primary carcinoma of the fallopian tube. A retrospective analysis of 47 patients. Ann Oncol 1996; 7(3): 271 – 5.
FALLOPIAN TUBE CANCER 33. Klein M, et al. Lymphadenectomy in primary carcinoma of the fallopian tube. Cancer Lett 1999; 147(1 – 2): 63 – 6. 34. Finn WF, Javert CT. Primary and metastatic cancer of the fallopian tube. Cancer 1949; 2: 803 – 14. 35. Hu CY, Taymor ML, Hertig AT. Primary carcinoma of the fallopian tube. Am J Obstet Gynecol 1950; 59(1): 58 – 67. 36. Eddy GL, et al. Fallopian tube carcinoma. Obstet Gynecol 1984; 64(4): 546 – 52. 37. Bannatyne P, Russell P. Early adenocarcinoma of the fallopian tubes. A case for multifocal tumorigenesis. Diagn Gynecol Obstet 1981; 3(1): 49 – 60. 38. Uehira K, et al. Transitional cell carcinoma pattern in primary carcinoma of the fallopian tube. Cancer 1993; 72(8): 2447 – 56. 39. Robey SS, et al. Transitional cell carcinoma in high-grade high-stage ovarian carcinoma. An indicator of favorable response to chemotherapy. Cancer 1989; 63(5): 839 – 47. 40. Navani SS, et al. Endometrioid carcinoma of the fallopian tube: a clinicopathologic analysis of 26 cases. Gynecol Oncol 1996; 63(3): 371 – 8. 41. Latzko W. Linkseitiges Tubenkarzinom Rechtsietige Karzinomatose tuboovarial Cyste. Zentralbl Gynakol 1916; 40: 599. 42. Muntz HG, et al. Post-hysterectomy carcinoma of the fallopian tube mimicking a vesicovaginal fistula. Obstet Gynecol 1992; 79(5 Pt 2): 853 – 6. 43. Ehlen T, et al. Posthysterectomy carcinoma of the fallopian tube presenting as vaginal adenocarcinoma: a case report. Gynecol Oncol 1989; 33(3): 382 – 5. 44. Karlan BY, et al. Whole-body positron emission tomography with (fluorine-18)-2-deoxyglucose can detect metastatic carcinoma of the fallopian tube. Gynecol Oncol 1993; 49(3): 383 – 8. 45. Hefler LA, et al. The clinical value of serum concentrations of cancer antigen 125 in patients with primary fallopian tube carcinoma: a multicenter study. Cancer 2000; 89(7): 1555 – 60. 46. Puls LE, et al. Immunohistochemical staining for CA-125 in fallopian tube carcinomas. Gynecol Oncol 1993; 48(3): 360 – 3. 47. Jones OV. Primary carcinoma of the uterine tube. Obstet Gynecol 1965; 26: 122 – 9. 48. Takashina T, Ito E, Kudo R. Cytologic diagnosis of primary tubal cancer. Acta Cytol 1985; 29(3): 367 – 72. 49. Maxson WZ, et al. Primary carcinoma of the fallopian tube: evidence for activity of cisplatin combination therapy. Gynecol Oncol 1987; 26(3): 305 – 13. 50. McMurray EH, et al. Carcinoma of the fallopian tube. Management and sites of failure. Cancer 1986; 58(9): 2070 – 5. 51. Klein M, et al. Radical lymphadenectomy in the primary carcinoma of the fallopian tube. Arch Gynecol Obstet 1993; 253(1): 21 – 5. 52. Rosen A, et al. Primary carcinoma of the fallopian tube – a retrospective analysis of 115 patients. Austrian Cooperative Study Group for Fallopian Tube Carcinoma. Br J Cancer 1993; 68(3): 605 – 9. 53. Barakat RR, et al. Cisplatin-based combination chemotherapy in carcinoma of the fallopian tube. Gynecol Oncol 1991; 42(2): 156 – 60. 54. Brown MD, et al. Fallopian tube carcinoma. Int J Radiat Oncol Biol Phys 1985; 11(3): 583 – 90. 55. Kojs Z, et al. Whole abdominal external beam radiation in the treatment of primary carcinoma of the fallopian tube. Gynecol Oncol 1997; 65(3): 473 – 7. 56. Schray MF, Podratz KC, Malkasian GD. Fallopian tube cancer: the role of radiation therapy. Radiother Oncol 1987; 10(4): 267 – 75. 57. Rosen AC, et al. Prognostic factors in primary fallopian tube carcinoma. Austrian Cooperative Study Group for Fallopian Tube Carcinoma. Gynecol Oncol 1994; 53(3): 307 – 13. 58. Johnston GA Jr. Primary malignancy of the fallopian tube: a clinical review of 13 cases. J Surg Oncol 1983; 24(4): 304 – 9. 59. Denham JW, Maclennan KA. The management of primary carcinoma of the fallopian tube. Experience of 40 cases. Cancer 1984; 53(1): 166 – 72. 60. Peters WA III, Andersen WA, Hopkins MP. Results of chemotherapy in advanced carcinoma of the fallopian tube. Cancer 1989; 63(5): 836 – 8. 61. Smith JP. Chemotherapy in gynecologic cancer. Clin Obstet Gynecol 1975; 18(4): 109 – 24.
483
62. Deppe G, Bruckner HW, Cohen CJ. Combination chemotherapy for advanced carcinoma of the fallopian tube. Obstet Gynecol 1980; 56(4): 530 – 2. 63. Boronow RC. Chemotherapy for disseminated tubal cancer. Obstet Gynecol 1973; 42(1): 62 – 6. 64. Tresukosol D, et al. Primary fallopian tube adenocarcinoma: clinical complete response after salvage treatment with high-dose paclitaxel. Gynecol Oncol 1995; 58(2): 258 – 61. 65. Markman M, et al. Phase 2 trial of liposomal doxorubicin (40 mg/m(2)) in platinum/paclitaxel-refractory ovarian and fallopian tube cancers and primary carcinoma of the peritoneum. Gynecol Oncol 2000; 78(3 Pt 1): 369 – 72. 66. Markman M, et al. Phase II trial of weekly single-agent paclitaxel in platinum/paclitaxel-refractory ovarian cancer. J Clin Oncol 2002; 20(9): 2365 – 9. 67. Rose PG, et al. Liposomal doxorubicin in ovarian, peritoneal, and tubal carcinoma: a retrospective comparative study of single-agent dosages. Gynecol Oncol 2001; 82(2): 323 – 8. 68. Berkenblit A, et al. A phase II trial of weekly docetaxel in patients with platinum-resistant epithelial ovarian, primary peritoneal serous cancer, or fallopian tube cancer. Gynecol Oncol 2004; 95(3): 624 – 31. 69. Markman M, et al. Phase 2 trial of single agent docetaxel in platinum and paclitaxel-refractory ovarian cancer, fallopian tube cancer, and primary carcinoma of the peritoneum. Gynecol Oncol 2003; 91(3): 573 – 6. 70. Cormio G, et al. Treatment of fallopian tube carcinoma with cyclophosphamide, adriamycin, and cisplatin. Am J Clin Oncol 1997; 20(2): 143 – 5. 71. Pectasides D, et al. Treatment of primary fallopian tube carcinoma with cisplatin-containing chemotherapy. Am J Clin Oncol 1994; 17(1): 68 – 71. 72. Muntz HG, et al. Combination chemotherapy in advanced adenocarcinoma of the fallopian tube. Gynecol Oncol 1991; 40(3): 268 – 73. 73. Gemignani ML, et al. Paclitaxel-based chemotherapy in carcinoma of the fallopian tube. Gynecol Oncol 2001; 80(1): 16 – 20. 74. Raju KS, Barker GH, Wiltshaw E. Primary carcinoma of the fallopian tube. Report of 22 cases. Br J Obstet Gynaecol 1981; 88(11): 1124 – 9. 75. Rosen AC, et al. Preoperative and postoperative CA-125 serum levels in primary fallopian tube carcinoma. Arch Gynecol Obstet 1994; 255(2): 65 – 8. 76. Semrad N, et al. Fallopian tube adenocarcinoma: common extraperitoneal recurrence. Gynecol Oncol 1986; 24(2): 230 – 5. 77. Schwartz PE, Smith JP. Second-look operations in ovarian cancer. Am J Obstet Gynecol 1980; 138(8): 1124 – 30. 78. Guthrie D, Cohen S. Carcinoma of the Fallopian tube treated with a combination of surgery and cytotoxic chemotherapy. Case report. Br J Obstet Gynaecol 1981; 88(10): 1051 – 3. 79. Eddy GL, Copeland LJ, Gershenson DM. Second-look laparotomy in fallopian tube carcinoma. Gynecol Oncol 1984; 19(2): 182 – 6. 80. Harrison CR, et al. Carcinoma of the fallopian tube: clinical management. Gynecol Oncol 1989; 32(3): 357 – 9. 81. Barakat RR, et al. Second-look laparotomy in carcinoma of the fallopian tube. Obstet Gynecol 1993; 82(5): 748 – 51. 82. Cormio G, et al. Second-look laparotomy in the management of fallopian tube carcinoma. Acta Obstet Gynecol Scand 1997; 76(4): 369 – 72. 83. Piver MS, et al. Second-look laparoscopy prior to proposed second-look laparotomy. Obstet Gynecol 1980; 55(5): 571 – 3. 84. Weber AM, et al. Malignant mixed mullerian tumors of the fallopian tube. Gynecol Oncol 1993; 50(2): 239 – 43. 85. Imachi M, et al. Malignant mixed Mullerian tumor of the fallopian tube: report of two cases and review of literature. Gynecol Oncol 1992; 47(1): 114 – 24. 86. Kahanpaa KV, Laine R, Saksela E. Malignant mixed Mullerian tumor of the fallopian tube: report of a case with 5-year survival. Gynecol Oncol 1983; 16(1): 144 – 9. 87. Muntz HG, et al. Carcinosarcomas and mixed Mullerian tumors of the fallopian tube. Gynecol Oncol 1989; 34(1): 109 – 15. 88. Carlson JA Jr, Ackerman BL, Wheeler JE. Malignant mixed mullerian tumor of the fallopian tube. Cancer 1993; 71(1): 187 – 92. 89. Hertig AT, Gore H. Hydatiform Mole and Choriocarcinoma. Washington, DC: Armed Forces Institute of Pathology, 1961.
484
GYNECOLOGICAL CANCERS
90. Muto MG, et al. Gestational trophoblastic disease of the fallopian tube. J Reprod Med 1991; 36(1): 57 – 60. 91. Dekel A, et al. Primary choriocarcinoma of the fallopian tube. Report of a case with survival and postoperative delivery. Review of the literature. Obstet Gynecol Surv 1986; 41(3): 142 – 8. 92. Ober WB, Maier RC. Gestational choriocarcinoma of the fallopian tube. Diagn Gynecol Obstet 1981; 3(3): 213 – 31.
93. Nordin AJ. Primary carcinoma of the fallopian tube: a 20-year literature review. Obstet Gynecol Surv 1994; 49(5): 349 – 61. 94. Rauthe G, Vahrson HW, Burkhardt E. Primary cancer of the fallopian tube. Treatment and results of 37 cases. Eur J Gynaecol Oncol 1998; 19(4): 356 – 62.
Section 7 : Gynecological Cancers
44
Uterine Sarcomas and Unusual Endometrial Carcinomas
Peter G. Rose, Pedro F. Escobar, Peter Fleming and Charles Biscotti
UTERINE SARCOMA Incidence and Etiology Uterine sarcoma is rare, accounting for only 3% of uterine cancers. However, its virulent course in advanced disease and its high recurrence rate, even when initially limited to the uterus, makes it one of the most lethal of gynecologic malignancies. Despite its rarity, uterine sarcoma accounts for 15% of uterine malignancy deaths. Little is known about the global incidence of uterine sarcomas, as most tumor registries do not specify the histologic type when reporting uterine malignancy. The National Cancer Institute has collected incidence data on uterine sarcoma since the 1970s, and the age-adjusted incidence has remained unchanged. However, in a Norwegian study the incidence and mortality of uterine sarcoma doubled from 1956 to 1992.1 Most of this increase was due to an increase in mixed mesodermal tumors. Patient age is varied, ranging from the second to the eighth decade with a mean age of 55.7 years. Leiomyosarcoma occurs at a younger age than other uterine sarcomas.2,3 Despite the threefold increased frequency of leiomyoma in black women, the incidence of leiomyosarcoma has not increased.4,5 Uterine sarcomas account for a higher percentage (10%) of uterine malignancies in black women.5 This figure is partly explained by the fact that endometrial carcinoma is twice as frequent in the white population as in blacks. However, the incidence of mixed mesodermal tumors is higher in black women than white women, being twice as common at age 60.5 years. A recent analysis of 2677 cases using Surveillance Epidemiology and End Results (SEER) databases indicated that blacks continue to be at greater risk than whites for uterine leiomyosarcoma and carcinosarcoma. The reasons for the differences in findings from previous reports may reflect adjustment of incidence figures to recent US female population census and categorizing carcinosarcoma and mixed M¨ullerian tumors into one group since these are currently considered to be of the same category.6
As with endometrial carcinoma, obesity, hypertension, and diabetes are frequently associated disorders, occurring in 18, 11, and 8% respectively, of patients with uterine sarcoma; their etiologic role is not defined.7 Marital status has also been reported to be a risk factor; with women who have never been married to be at a 50% increased risk.8 Press and Scully reported six endometrial sarcomas following unopposed estrogenic stimulation and suggested that it is an etiologic factor for uterine sarcoma.9 Subsequent cases of endometrial sarcomas following unopposed estrogenic stimulation and associated with tamoxifen have been reported.10 – 12 Schwartz et al., in a case –controlled study of 166 patients with uterine sarcoma, found that exogenous hormone use and increased body mass index increased the risk of uterine sarcoma.13 Conversely, cigarette smoking decreased the risk, presumably as a result of estrogen inactivation. Numerous authors have reported uterine sarcoma following radiation therapy for carcinoma of the cervix.14,15 Czesnin and Wronkowski reported three uterine sarcomas developing among 8043 cervical cancer patients treated with radiation therapy, placing them at a relative risk of 5.48.16 Mark et al. reported 13 cases of postradiation sarcoma and estimated a risk of 0.03–0.8%.17 In an international collaborative study involving 82 616 irradiated cervical cancer patients with 623 798 years of follow-up, no increased incidence of uterine malignancy was found, although uterine sarcomas were not reported separately.18 Malignant transformation of leiomyoma is rare. At the Johns Hopkins University a review of 13 000 leiomyomas revealed only 38 cases with malignant change (0.29%).19 However, this figure represented surgically treated patients, and a more accurate estimate is less than 0.1%.20 In the Mayo Clinic series only 3 of 105 cases of leiomyosarcoma were thought to represent malignant transformation of previously diagnosed leiomyoma.21
Molecular Biology At a molecular level, mutations of the p53 tumor suppressor gene commonly occur in uterine sarcoma.22,23 Additionally,
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
486
GYNECOLOGICAL CANCERS
molecular alterations of MSH2 and microsatellite instability have been identified in 25% of uterine sarcomas.24 Tyrosine Kinases
A wide variety of cancers including uterine sarcomas, demonstrate activation or overexpression of several tyrosine kinases. Several studies have investigated the expression of KIT (transmembrane tyrosine kinase receptor) in leiomyosarcomas, which ranges from 0 to 75%.25 – 27 More recent studies have demonstrated that although uterine sarcomas express c-kit, they lack mutations at exon 9, 11, and 13 and, therefore, are not expected to respond to imatinib.28,29 Matrix Metalloproteinase
Matrix metalloproteinase (MMPs) are endopeptidases, which are capable of degrading matrix proteins, and because these enzymes need zinc or calcium atoms to work properly, they are called metalloproteinase. According to their cellular localization, MMPs can be further subdivided into secretedand membrane-bound MMPs. The MMPs play an important role in tissue remodeling associated with various physiological and pathological processes such as morphogenesis, angiogenesis, tissue repair, and tumor invasion. BodnerAlder et al.30 analyzed the expression of MMP (MMP-1 and MMP-2) proteins in 21 patients with uterine leiomyosarcoma, with expression of MMP-1 in 86% and MMP-2 in 48% of the cases. Furthermore, a statistically significant positive correlation was found between MMP-2 expression and vascular space involvement. In contrast, prolonged, disease-free survival was noted in patients with no MMP-2 expression.30
(22%).36 Leiomyosarcomas and endometrial stromal sarcomas (ESS) are also less frequently diagnosed by preoperative uterine curettage (4%) than are mixed mesodermal tumors (91%).4 A radiograph of the chest is routinely obtained preoperatively and is a useful screening method for pulmonary metastasis. However, in patients who are finally diagnosed as having uterine sarcoma, computed tomography of the chest is necessary to exclude early metastatic disease.37 Isolated instances of the use of magnetic resonance imaging (MRI) for preoperative evaluation of leiomyosarcoma have been reported,38 although its real impact will require more extensive evaluation. Other radiographic studies and endoscopic procedures are performed as indicated by the patient’s symptoms and the results of preliminary laboratory studies.
Pathology Numerous histologic subtypes of uterine sarcoma have been described. Ober classified these tumors as pure or mixed with either homologous or heterologous components39 (see Table 1). Pure sarcomas contain a single recognizable element, whereas mixed sarcomas contain two or more elements. Sarcomatous elements are further classified as homologous, containing tissue native to the uterus (e.g. smooth muscle, endometrial stroma, and blood or lymph vessels), or heterologous, with tissue foreign to the uterus (e.g. bone, cartilage, skeletal muscle, or fat). Kempson and Bari elaborated on this scheme, further subdividing by histologic type.40 Table 2 presents the World Health Organization classification of uterine sarcomas. Leiomyosarcoma
Clinical Presentation Signs and Symptoms
Irregular vaginal bleeding is the most common presenting symptom (80%) for all uterine sarcomas. Vaginal bleeding is more frequent with endometrial sarcomas (94%) than with leiomyosarcomas (58%).4 The presence of gross tissue at the cervical os is common and in one series was present in 73% of patients with mixed mesodermal tumors.31 Other symptoms include pelvic or abdominal pain (16%), an enlarged uterus (12%), a pelvic or abdominal mass (9.5%), and vaginal discharge (9.5%).3 Leiomyosarcomas are classically thought to present as a rapidly enlarging pelvic mass. However, one recent review found leiomyosarcoma to occur in only 0.23% of such patients.32 Gonadotropin-releasing hormone analogs have been used in patients with presumed leiomyoma who are subsequently found to have leiomyosarcoma.33,34 The failure of involution or the presence of increased vaginal bleeding has been suggested as a means of identifying the leiomyosarcoma.34,35 Preoperative Evaluation
Because most of the patients present with uterine bleeding, routine evaluation including Papanicolaou smear and fractional endocervical and endometrial sampling are often performed. Papanicolaou smears are more frequently positive with endometrial sarcoma (46%) than with leiomyosarcoma
Although formerly the most frequently reported uterine sarcoma,41 studies now report leiomyosarcoma to be second in frequency after carcinosarcoma.3,7 Leiomyosarcoma is a pure tumor arising in the myometrium (see Figure 1). Table 1 Ober classification of uterine sarcomas.
Homologous
Heterologous
Pure Stromal sarcoma Leiomyosarcoma Angiosarcoma Liposarcoma
Pure Rhabdomyosarcoma Chondrosarcoma Osteosarcoma –
Mixed Carcinosarcoma
Mixed Mixed M¨ullerian tumor
Table 2 Classification of uterine sarcomas endorsed by the World Health Organization.
Leiomyosarcomas Endometrial stromal sarcoma, low grade Undifferentiated endometrial sarcoma Other mesenchymal tumors Mixed epithelial and mesenchymal tumors Carcinosarcoma Adenosarcoma Source: Tavassoli FA, Devilee P. Pathology and genetics. Tumors of the Breast and Female Genital Organs. World Heal Organization Classification of Tumors, IARC Press, 2003.
UTERINE SARCOMAS AND UNUSUAL ENDOMETRIAL CARCINOMAS
Figure 1 Uterine leiomyosarcomas typically have a soft variegated appearance often containing hemorrhage and necrosis.
Approximately 70% are intramural, 20% are submucosal, and 10% are subserosal.42 A higher percentage involve the cervix than do leiomyomas.43 Furthermore, intraoperative frozen section evaluation for leiomyosarcoma is poor, with only 3 of 16 cases (18%) correctly identified.35,44 The diagnosis depends on a constellation of pathological features including mitotic index, nuclear atypia, and necrosis. In one study, 90% of leiomyosarcomas had 10 or more mitoses per 10 hpf.44 Mitotic count alone is not sufficient to diagnose sarcoma. Some benign, smooth muscle tumors have more than 10 mitoses per 10 hpf and some sarcomas have fewer. Among the 27 uterine smooth muscle neoplasms with 5–9 mitoses per 10 hpf, 11 (40%) recurred. Even among 42 uterine smooth muscle tumors with 1–4 mitoses per 10 hpf, 5 (12%) recurred.45 Recently, there have been a number of favorable reports of patients with up to 15 mitoses per 10 hpf as having a benign course.46 – 48 On this basis, Hendrickson and Kempson have revised the classification of smooth muscle tumors (Table 3). Using these criteria, Peters et al. reviewed the prognosis of tumors with uncertain malignant potential,49 and found that 27% of these patients developed recurrence and had a protracted course. Bell et al. have further identified coagulative tumor necrosis as an important criterion of malignancy50 (see Figure 2). Myxoid smooth muscle tumors contain smooth muscle cells separated by abundant myxoid ground substance. Clinically malignant tumors can have low mitotic counts, 0–2 per 10 hpf.51,52 An infiltrative margin with invasion of adjacent myometrium and/or vessels is a better criterion of malignancy.
487
Figure 2 Microscopically, most leiomyosarcomas contain tumor cell necrosis (center right) characterized by sharp zones of necrosis lacking a healing margin.
Figure 3 Low-grade endometrial stromal sarcomas resemble proliferative endometrial stroma. This typical example has the characteristic cellular appearance. A mitosis is present (center).
Endometrial Stromal Sarcoma
Endometrial sarcomas are classified as low-grade endometrial stromal sarcoma, or undifferentiated endometrial stromal sarcoma (see Table 2).53 Endometrial stromal nodule is a benign neoplasm of endometrial stromal cells characterized by a circumscribed margin. Stromal nodules and low-grade stromal sarcomas closely resemble proliferative endometrial stroma (see Figure 3). Many delicate small arterioles are
Table 3 Hendrickson and Kempson revised classification system for uterine smooth muscle tumors.
Mitotic count per 10 hpf Cellular atypia
<5
None Mild to moderate
Leiomyoma Leiomyoma
Severe
Uncertain malignant potential
5 – 10 Leiomyoma Uncertain malignant potential Leiomyosarcoma
11 – 15
>15
Leiomyoma Leiomyosarcoma
Leiomyosarcoma Leiomyosarcoma
Leiomyosarcoma
Leiomyosarcoma
488
GYNECOLOGICAL CANCERS
characteristically present. Stromal nodules have an expansive circumscribed margin. In contrast, low-grade ESS have irregular, infiltrative margins54 (see Figures 4 and 5). The definitive diagnosis of stromal nodule requires excision of the entire lesion and examination of the tumor margin. Thus, the distinction between stromal nodule and low-grade ESS is rarely possible in biopsy or curetting specimens. Low-grade ESS, previously referred to as endolymphatic stromal myosis (ESM), are indolent malignancies with a high (100%) 5-year survival rate. With prolonged followup, 7 of 19 patients reported by Norris and Taylor developed recurrent disease, although it accounted for only one patient death.55 It has been demonstrated that low-grade ESS more commonly expresses estrogen and progesterone receptors.56 Undifferentiated endometrial sarcoma has recently been defined as an endometrial nonepithelial neoplasm without the features of endometrial stromal sarcoma or other distinct sarcoma.57 Undifferentiated endometrial sarcomas are high-grade sarcomas characterized by more unclear pleomorphism and atypia than low-grade ESS. By definition,
undifferentiated endometrial sarcoma lacks resemblance to proliferative endometrial stroma and lacks heterologous differentiation. These tumors behave aggressively and have a poor prognosis. Mixed Epithelial and Mesenchymal Tumors
Figure 4 This low-grade endometrial stromal sarcoma has the characteristic pattern of myometrial infiltration.
Carcinosarcomas contain both a sarcomatous element and a carcinomatous element (see Figure 6). While originally believed to be the result of two separate cancers colliding (collision theory), more recent cell culture and pathologic studies suggest that this specific carcinoma can convert to a sarcoma (conversion theory).58 – 62 Since the endometrial origin of carcinosarcoma has been well established, Amant et al. evaluated whether uterine carcinosarcomas have a different prognosis other than high-risk histologies including serous papillary carcinoma, clear cell carcinoma, and poorly differentiated endometrioid carcinoma.63 Carcinosarcoma and nonendometrioid carcinoma were more likely to spread to regional lymph nodes and carcinosarcomas were more likely to spread to the lungs. Among stage I or II cancers, carcinosarcomas had a poorer prognosis with 44% survival compared to 75% and 86% survival for nonendometrioid and grade 3 endometrioid carcinomas, respectively. Metastatic sites often demonstrate only the carcinomatous element.58 Sarcomatous elements are classified as homologous or heterologous. The term carcinosarcoma, originally used to specify a tumor with malignant epithelial and homologous sarcomatous elements, is now considered synonymous with malignant M¨ullerian mixed tumor, malignant mesodermal mixed tumor, and metaplastic carcinoma (see Table 2). Norris and Taylor reported that homologous carcinosarcomas had a better prognosis than heterologous tumors.60 This suggestion, however, was not supported in a number of subsequent series.61,64,65 Certain heterologous elements such as cartilage or skeletal muscle have been reported to affect survival adversely.66,67 In a more recent series, recurrence was statistically more frequent in stages I and II patients with rhabdomyosarcoma than in those with chondrosarcoma. Rhabdomyosarcomas, however, invaded deeper into the myometrium than chondrosarcomas, which may have
Figure 5 Low endometrial stromal sarcomas often invade blood vessels. This example has a tumor thrombus filling the uterine vein.
Figure 6 This carcinosarcoma has the characteristic bulky, polypoid, fleshy gross appearance.
UTERINE SARCOMAS AND UNUSUAL ENDOMETRIAL CARCINOMAS
accounted for their worse outcome.68 The tumor marker CA125 has been reported to correlate with the disease status.69 Adenosarcoma is composed of a sarcoma admixed with a benign epithelial component70,71 (see Figures 7 and 8). Compared to carcinosarcoma, adenosarcoma has a better prognosis. Clement and Scully reported long-term follow-up in 88 of 100 patients.70 Only 15 of the 100 patients had myometrial invasion, with only 4 having deep myometrial invasion. Twenty-six percent of patients developed recurrent tumor that was usually limited to the vagina, pelvis, and abdomen. Myometrial invasion was the only feature associated with recurrence. The term adenosarcoma with sarcomatous overgrowth identifies a tumor in which 25% or more of the tumor is composed of a high-grade sarcomatous component.70,71 Adenosarcomas with sarcomatous overgrowth are associated with a significantly higher risk of recurrence (44–70%) that is similar to that of carcinosarcoma. Across all histologic types, numerous studies have confirmed tumor ploidy to be a significant prognostic factor for survival.72 – 74
Figure 7 Uterine M¨ullerian adenosarcomas usually have a polypoid gross appearance, as seen in this example.
489
A variety of other sarcomas arise in the uterus. These include high-grade sarcomas resembling undifferentiated endometrial sarcoma, but having heterologous differentiation. Heterologous rhabdomyosarcoma occurs most commonly, but chondrosarcoma, osteosarcoma, liposarcoma, and mixed heterologous sarcomas also occur.75 Mixed stromal muscle tumors also occur, albeit rarely. A small amount of smooth muscle differentiation is common and irrelevant in a typical low-grade ESS.76 Similarly, an insignificant amount of stromal differentiation has no relevance in an otherwise typical smooth muscle tumor. A minimum 30% threshold has been established for the designation mixed stromal muscle tumor.77 Recently, perivascular epithelioid cell tumor (PECOMA) has been reported in the uterus.78 These tumors histologically resemble epithelioid smooth muscle tumors but strongly express HMB-45. These tumors should be regarded as those of uncertain malignant potential based on limited experience.
Staging and Pattern of Spread There is no accepted staging system for uterine sarcomas. The International Federation of Gynecology and Obstetrics (FIGO) staging for endometrial cancer is most commonly applied. Survival depends on stage, with significantly improved survival in stage I patients.2,4 Most patients (55–65%) present with clinical stage I disease,2 – 4 although surgical exploration demonstrates what appears to be a more advanced disease in as many as 253 to 55%79 of cases. The treatment of uterine sarcoma has been hampered by our lack of understanding of the biologic behavior and mode of metastasis of these tumors. Uterine sarcomas frequently metastasize to extrapelvic sites. Salazar et al. collected data from 235 patients with recurrent disease from the literature and noted that 85% of recurrences included extrapelvic sites80 among 45 patients with distant recurrence whose metastatic sites included the lung (69%), upper abdomen (60%), bone (24%), and brain (4%). However, both distant and local recurrences are common.81 The high recurrence rate of stage I disease despite hysterectomy (50–80%)4,7,82 implies that subclinical metastases are present. Tumor metastasis may occur from lymphatic, hematogenous or transperitoneal routes. Positive peritoneal cytology is frequently found in advanced stage disease. The significance of positive peritoneal cytology in stage I mixed mesodermal tumor has been reported.83,84 Among 18 pathologic stage I patients, 10 of 10 with positive cytology died of disease, in contrast to 6 of 23 dying with negative cytology. Positive cytology appears to be predictive of subclinical extrauterine intraperitoneal metastasis. Whether lymphatic and peritoneal spread is the only mode for distant metastases is not known. In an autopsy study, a high percentage of patients had pulmonary metastasis in the absence of retroperitoneal nodal or intraperitoneal involvement.85 This finding supports the hematogenous route of metastasis.
Primary Therapy Surgery Figure 8 Cellular stromal papillae protrude into gland spaces yielding a phyllodes pattern characteristic of adenosarcoma.
The traditional surgical therapy for uterine sarcoma has been total abdominal hysterectomy and bilateral salpingo
490
GYNECOLOGICAL CANCERS
oophorectomy (TAH/BSO). In a cumulative report of 90 stage I patients treated in this manner, the 2-year survival was 45%.86 The presence of extrauterine disease significantly affects prognosis.64 – 67 A number of studies have examined the role of surgical staging in uterine sarcoma.68,87,88 The GOG studied 453 patients with clinical stages I and II uterine sarcoma, who underwent comprehensive surgical staging.65 Lymph node involvement correlated strongly with histologic type. Among 287 eligible patients with mixed mesodermal tumors the frequency of pelvic or para-aortic node metastasis was 17.8%. Pelvic nodes were involved twice as frequently as para-aortic nodes. Factors associated with nodal involvement include adnexal metastasis, positive peritoneal cytology, more than 50% myometrial invasion, cervical or isthmic tumor location, and vascular lymphatic involvement. Other sites of known intra-abdominal recurrence, including the omentum, peritoneum, bowel, and liver, should be evaluated.80 In contrast, nodal involvement was seen in only 2 of 57 patients (3.5%) with leiomyosarcoma.65 Cytoreduction For patients with disease extending outside the uterus a TAH/BSO is recommended if it will effect significant tumor reduction or to control excessive uterine bleeding. The role of radical debulking surgery is not well documented, but isolated responses have been reported. Parente et al. reported a 10-month complete clinical response of a patient having leiomyosarcoma with pulmonary metastasis following primary hysterectomy and chemotherapy.89 Second-look Laparotomy Hannigan et al. reexplored patients with uterine sarcoma who had undergone primary surgery and subsequent chemotherapy and radiation therapy.90 Of eight patients with negative second-look laparotomy, there were no recurrences after a median follow-up of 72 months. However, only 11 of the original 106 patients treated during the period of the study underwent a second-look procedure. Furthermore, of the 11 patients reported, 46% had low-grade tumors. Two patients with gross disease at second look were redebulked and treated with chemotherapy, and they were alive and disease-free off treatment at 30 and 50 months. Myomectomy Van Dinh and Woodruff reported nine patients who were discovered to have leiomyosarcoma after myomectomy for infertility.91 Only one patient developed a recurrence during follow-up ranging from 1 to 13 years. Three patients subsequently became pregnant. O’Connor and Norris reported 14 patients treated by myomectomy. Only one patient had a recurrence of a low-grade leiomyosarcoma, 8 years after her original surgery.47 However, Berchuck et al. reported residual disease at hysterectomy in two of three patients treated originally by myomectomy.92 Although most authors have reported a good outcome with conservative therapy, a standard recommendation for such therapy cannot be made. The prognosis is largely dependent on the grade of the original histology. Radiation Therapy
Among medically inoperable patients, a small percentage can be effectively treated with primary irradiation. Among
37 patients reported by Badib et al., 5 (14%) achieved a 5-year survival.93 When evaluated by stage, four of nine (44%) stage I patients survived 5 years compared with one of nine stage II patients and no stage III (n = 10) or stage IV (n = 9) patients. DiSaia et al. reported 18 patients with mixed mesodermal sarcoma treated by irradiation alone.64 For patients with disease limited to the uterus, two of five were alive and disease-free at 2 years. However, if disease extended to the cervix, vagina, or parametrium (n = 6) or outside the pelvis (n = 7), there were no 2-year survivors. Perez et al. reported no 3-year survivors following irradiation alone for stage II or IV disease.94 Badib et al. reported that only 8% of leiomyosarcoma patients and 10% of ESS patients survived 5 years following radiation therapy compared to 17% of those with carcinosarcomas (homologous mixed mesodermal tumors) and 25% of those with heterologous mixed mesodermal tumors.93 The significance of this histologic survival difference is uncertain in view of the small number of patients treated and the inaccuracy of clinical staging. Chemotherapy
There are no reports of primary chemotherapy achieving a cure,4,65,95 although transient tumor reduction has been seen with several cytotoxics in common use, such as doxorubicin and cisplatin, depending on the predominant histological pattern.
Adjuvant Therapy for Stages I and II Disease Adjuvant Radiation Therapy
As demonstrated previously, surgery alone resulted in only a 45% 2-year survival. To decrease recurrence, many centers have utilized adjuvant radiation therapy after primary hysterectomy. The cumulative results are presented in Table 4. Improved survival following adjuvant radiation therapy has been suggested with mixed mesodermal tumors but not with leiomyosarcoma. In summary, most studies demonstrate that adjuvant radiation therapy increases pelvic disease control with only occasional improved survival. In the large surgical randomized trial of adjuvant doxorubicin by the GOG in which postoperative radiation therapy was left to the discretion of the physician, recurrences were noted in 53% of patients treated with radiation versus 57.5% of those not treated with radiation.96 The sites of first recurrence differ for leiomyosarcoma and other endometrial sarcoma histologies.97 In leiomyosarcoma, 83% of first recurrences were at distant sites while for other endometrial sarcoma histologies, among 54% of patients not receiving radiation, first recurrences occurred in the pelvis. Radiation therapy decreased pelvic recurrences for nonleiomyosarcoma histologies from 54 to 23%.97 Adjuvant Chemotherapy
In view of frequent distant recurrence and the proved, although limited, effectiveness of chemotherapy for advanced disease, there is great interest in prophylactic chemotherapy for early stage disease. The cumulative results of adjuvant
UTERINE SARCOMAS AND UNUSUAL ENDOMETRIAL CARCINOMAS
491
Table 4 Frequency of recurrence following adjuvant radiation therapy.
Stage I (number of patients)a Surgery
Surgery + irradiation
Surgery
Surgery + irradiation
5 (3) 12 (12) 54 (30) 24 (10) 6 (3) 44 (21) 32 (24) –
24 (10) 4 (1) 28 (7) 14 (7) 17 (6) 23 (14) 16 (9) 40 (6)
5 (5) – 7 (7) – 3 (3) 9 (6) – –
11 (8) – 3 (0) – 8 (5) 5 (2) – 9 (5)
177 (103), 58.2%
166 (60), 36.1%
24 (21), 87.5%
36 (20), 55.6%
References 64b,c
DiSaia et al. Gilbert et al.98c Vongtama et al.99d Salazar et al.80 Perez et al.94c,e Omura96 Rose et al.100 Knocke et al.101 Total Recurrence rate
Stage II (number of patients)a
a
Numbers in parentheses are the number of patients with recurrence. Reported at 2 years follow-up. Mixed mesodermal tumor or endometrial stomal sarcoma. d Reported at 5 years follow-up. e Reported at 3 years follow-up. b c
chemotherapy are presented in Table 5. The GOG performed a randomized study with 156 patients comparing doxorubicin for six monthly cycles to no further therapy.102 The recurrence rate was 41% for those treated with chemotherapy compared to 53% for those not receiving chemotherapy. The progression-free interval was 73.7 months for the chemotherapy arm versus 55 months for the untreated arm. These differences were not statistically different, although the size of the trial reduced its power to detect a small, but real, difference. At the time of the trial, the differential response rate of leiomyosarcoma and mixed mesodermal tumors was not known,103 and there were too few patients to allow subset analysis. A recent meta-analysis of 14 localized resectable adult soft tissue sarcomas totaling 1568 patients demonstrates a hazard ratio of 0.75 (CI 0.64–0.87) with adjuvant doxorubicin.104 Sutton et al. reported the largest trial of adjuvant chemotherapy for stages I and II carcinosarcomas in which patients were treated with three cycles
of cisplatin and ifosfamide and no radiation therapy. The 7-year disease-free survival rate was 52% with 10 of 19 single site recurrences in the pelvis.105 The distant spread of uterine sarcoma particularly implies the need for adjuvant systemic therapy. Whether current agents or combinations are effective enough to result in long-term cure requires further study. Adjuvant Chemotherapy and Radiation
Since the majority of first recurrences are in the pelvis, for carcinosarcoma and other nonleiomyosarcoma histologies, the combination of radiation therapy for local control and chemotherapy for systemic disease control might be more effective. Manolitsas et al. were the first to report the combined use of both localized radiation therapy and systemic chemotherapy for uterine carcinosarcoma.110 Thirty-eight patients with clinical Stage I or II who underwent surgical staging followed by platinum/anthracycline chemotherapy
Table 5 Frequency of recurrence following adjuvant chemotherapy.
Stage I (number of patients)a Regimen
Surgery
Surgery + chemotherapy
Surgery
Surgery + chemotherapy
VAC Doxorubicin Doxorubicin VACb Doxorubicind CYVADe,f Not specified Cisplatin/ifosfamide
– – 44 (21), 48% – 11 (7), 64% – 28 (21), 75% –
12 (7), 58% 29 (3), 10% 41 (16), 39% 7 (2), 29% 8 (2), 25% 11 (2), 18% 5 (1), 20% 50 (23h )
– – 9 (6), 67% – – – – –
– – 3 (3), 100% – – – – 15 (7h )
83 (49), 59.0%
163 (56), 34.4%
9 (6), 67%
18 (10), 56%
References 106
Buchsbaum et al. Kolstad102c Omura et al.102 VanNagell et al.107 Piver et al.108 Kohorn et al.109g Sutton et al.105
Stage II (number of patients)a
b
Total Recurrence rate a
–
Number in parentheses are the number of patients with recurrence. VAC, vincristine, dactinomycin, cyclophosphamide. Reported at 18 months follow-up. d Reported at 5 years follow-up. e Reported at 2 – 5 years follow-up. f CYVAD, cyclophosphamide, vincristine, doxorubicin, dacarbazine. g Stages I and II reported together. h Estimated. b c
492
GYNECOLOGICAL CANCERS
for six courses with tailored radiation therapy sandwiched between cycles two and three of the chemotherapy. The entire treatment program was not completed in 17 patients because of poor performance status or patient or investigator deviations. The survival rate for those patients who completed treatment according to the multimodality protocol was 95% (20 of 21 patients), with a disease-free survival rate of 90% (19 of 21 patients). Pautier et al. reported the use of adjuvant chemotherapy with cisplatin, ifosfamide, and doxorubicin for three cycles followed by radiotherapy in 18 patients with localized uterine sarcomas.111 At 3 years the recurrence-free survival was 76% for those treated with both modalities versus 43% for historical controls treated with radiation therapy alone, although the 95% confidence intervals overlapped. Further studies of radiation therapy combined with chemotherapy are needed to determine its role. Adjuvant Hormonal (Low-grade Endometrial Stromal Sarcomas [EES]) Therapy
Because of the high (50%) recurrence rate for stage I uterine ESM and objective response to progestational agents,112 – 115 a number of authors have advocated adjuvant progesterone therapy.114,116,117 In view of the high frequency of estrogen and progesterone receptors in uterine sarcomas, castration should be part of primary therapy.
Advanced Disease (Stages III and IV) The survival rate for patients with stages III and IV disease is poor, ranging from 0 to 10%.2,3 In patients with disease limited to the pelvis on laparotomy (stage III), radiation therapy has occasionally resulted in prolonged survival.96 However, these patients are at risk to have subclinical extrapelvic disease and treatment should include systemic therapy.118 As mentioned earlier, response to aggressive debulking and chemotherapy has been reported.89,119 In studies of patients with advanced or recurrent disease, the ability to excise all gross tumor has been shown to confer a prognostic benefit.120 Pulmonary metastases frequently respond to chemotherapy,110 and their presence is not an absolute contraindication of surgical abdominal debulking. Recurrent Disease
Most recurrences are at distant sites and require systemic therapy. Isolated late pulmonary recurrences have responded to local excision and have been associated with up to 35% 10-year survival.121,122 However, isolated pulmonary recurrences are rare, being seen in only 1 of 25 patients with stage I or II leiomyosarcoma in one series.92 This patient was treated by thoracotomy but had a quick recurrence and died. Rarely, isolated pelvic recurrences have been treated with exenterative surgery.123 The success of these local therapies may, in part, be due to the indolent nature of some low-grade sarcomas. Occasionally, recurrences in the pelvis have responded to radiation therapy alone. Smith et al.124 treated 38 patients with recurrent or advanced disease with pelvic radiotherapy and concomitant vincristine followed
by VAC (vincristine, actinomycin D, cyclophosphamide) chemotherapy for 2 years. Fourteen patients were alive for 10–70 months without disease, although the complications of therapy were severe.
Chemotherapy Chemotherapy for uterine sarcoma has been integrally based on the results of other soft tissue sarcoma regimens. Only through recent cooperative studies by the GOG has the chemosensitivity of this heterogeneous group of tumors been subdivided and studied. Because of these differences in response rates among the uterine sarcoma histologies, prior studies that fail to recognize this difference are difficult to interpret. Effective single agents used in the therapy for uterine sarcoma are listed in Table 6. In leiomyosarcoma, doxorubicin and ifosfamide have been considered the most active agents with response rates of 25 and 17% respectively.103,125 Recently, gemcitabine has also demonstrated significant activity with a response rate of 20.9%.126 Liposomal doxorubicin is also active with a response rate of 16.1%.127 However, cisplatin has no significant activity (3–5%).128,129 In contrast, in carcinosarcomas, ifosfamide, cisplatin, and paclitaxel are the most active agents, with response rates of 32, 20, and 18% respectively.129 – 133 Doxorubicin seems less active, with a 10% response rate.103 Combination chemotherapy, first utilized for soft tissue sarcoma in 1970, produced improved response rates. However, response rates in older studies are difficult to interpret since they often included patients with different histologies and primary gynecologic sites. Combination chemotherapy regimens used for uterine sarcoma are listed in Table 7. Two randomized studies, doxorubicin with Table 6 Single-agent activity for uterine sarcoma by histology.a
Agents
Complete Partial Overall response response response (%) References
Leiomyosarcoma Amonifide Cisplatin Doxorubicin Etoposide Gemcitabine Ifosfamide Liposomal doxorubicin Paclitaxel Piperazinedione Topotecan Trimetrexate
0/26 0/52 – 0/57 1/44 0/35 1/31 5/80 0/11 1/36 1/23
1/26 2/52 7/28a 2/57 8/44 6/35 4/31 2/80 1/11 3/36 0/23
4 3.8 25 3.5 20.5 17.1 16.1 8.8 9.1 11 4.3
134 128,129 103 135,136 126,137 125 127 138,139 140 141 142
Carcinosarcoma Amonifide Cisplatin Diaziquone (AZQ) Doxorubicin Etoposide Ifosfamide Paclitaxel Piperazinedione Trimetrexate Topotecan
0/16 6/91 0/22 0/56 0/31 8/49 4/44 0/6 0/21 5/48
1/16 12/91 1/22 7/56 2/31 8/49 4/44 0/6 1/21 0/48
6.3 19.8 4 12.5 6.5 32.7 18.2 0 4.8 10
143 129,130 144 103,145 146 131,132 133 140 147 148
a
Complete and partial responses not reported separately.
UTERINE SARCOMAS AND UNUSUAL ENDOMETRIAL CARCINOMAS
493
Table 7 Effective combination chemotherapy regimens used for uterine sarcomas.a
Regimen
Complete response
Partial response
Overall response (%)
References
Leiomyosarcoma Cisplatin, dacarbazine Cisplatin, vincristine, doxorubicin, dacarbazine Doxorubicin, dacarbazine Doxorubicin, ifosfamide Etoposide, cisplatin, doxorubicin Gemcitabine, docetaxel Mitomycin, cisplatin, doxorubicin Hydroxyurea, dacarbazine, etoposide Vincristine, dactinomycin, cyclophosphamide (VAC)
1/3 1/3 – 1/27 1/7 3/29 3/35 2/38 0/17
1/3 0/3 6/20a 9/27 1/7 13/29 5/35 5/38 1/17
33.3 16.7 30.0 37.0 28.6 55.2 22.9 18.4 5.9
151 152 103 153 154 155 156 157 158
Carcinosarcoma Carboplatin, paclitaxel Cisplatin, dacarbazine, Cisplatin, dacarbazine, doxorubicin Cisplatin, ifosfamide Cisplatin, vincristine, doxorubicin, dacarbazine Doxorubicin, cisplatin Doxorubicin, dacarbazine Hexamethylmelamine, cyclophosphamide, doxorubicin, cisplatin Hydroxyurea, dacarbazine, etoposide Vincristine, dactinomycin, cyclophosphamide (VAC)
4/5 1/6 1/6 29/92 2/13 3/6 – 2/7 2/33 0/19
0/5 1/6 1/6 21/92 1/13 2/6 7/31a 3/7 3/33 3/19
80.0 16.7 16.7 54.3 23.1 83.3 22.6 71.4 15.2 15.8
159 151 150 160 152 161 103 162 149 158
a
Complete and partial responses not reported separately.
or without dacarbazine and doxorubicin with or without cyclophosphamide, demonstrated no difference in activity in patients with a mixture of sarcoma histologies.103,149 Since those studies, the GOG has limited its studies in uterine sarcoma to either leiomyosarcoma or carcinosarcoma. The combination of ifosfamide and cisplatin was studied against ifosfamide alone in a phase III trial in mixed mesodermal tumors of the uterus.150 The response rate in the combination arm was 54 versus 36% with ifosfamide alone. However, the combination produced greater toxicity, a minimal impact on progression-free survival (median 6 vs 4 months), and no difference in survival. A subsequent GOG phase III trial compared the combination of ifosfamide and paclitaxel versus ifosfamide alone with final results pending. Currently the GOG is studying the combination of paclitaxel and carboplatin in patients with uterine carcinosarcoma. In leiomyosarcoma of the uterus, the combination of doxorubicin and ifosfamide has been studied with a 37% response rate. In view of the single-agent activity of gemcitabine, Hensley et al. reported the combination of gemcitabine and docetaxel with a response rate of 53% despite 50% of the patients receiving prior doxorubicin. This study is being repeated by the GOG.
Biologic Therapy Targeted Therapy
Despite the high frequency of estrogen and progesterone receptors in uterine sarcoma tissue, response to hormonal therapy is rare (3%).122 Consistent response to hormonal therapy has been limited to low-grade ESS.106 Hormonal therapy is the treatment of choice for these tumors.123 An ongoing GOG study is evaluating imatinib in leiomyosarcoma.
Immunotherapy
Thalidomide has been studied in 29 evaluable patients with no responses seen and only 2 patients remained progressionfree for >6 months, which was the primary measure of efficacy being evaluated.163 There has been limited experience with immunotherapy in this disease. Patients with uterine sarcomas have been included in two vaccine trials with peptides or autologous tumor antigen-pulsed dendritic cells with no responses observed.164,165
Other Rare Tumors Lymphomas
Primary lymphomas of the cervix and uterus are rare (see also Chapter 49, Rare Lymphomas). Only nine cases were reported among 9500 lymphoma patients at the Armed Forces Institute of Pathology.166 Involvement as part of a generalized process is well recognized, and in reported series varies from 16167 to 40%168 of patients. When disease is limited to the vagina, cervix, or uterus, a 5-year survival of 73% is reported,169 compared to 40% when the disease involves the ovaries.170 If discovered at laparotomy, assessment of splenic and lymph node disease status, as well as liver biopsy, is advocated.171 Granulocytic sarcoma, also known as chloroma because of its green color, has been reported in the female genital tract and may precede the diagnosis of acute myelogenous leukemia.172 Hemangiopericytomas
Hemangiopericytomas are rare vascular uterine sarcomas characterized by proliferation of capillaries. These tumors are best treated surgically and respond poorly to chemotherapy or radiation therapy.173 Gelfoam embolization preoperatively has aided in surgical excision.174
494
GYNECOLOGICAL CANCERS
Rare Mesenchymal Tumors Leiomyoblastoma
Leiomyoblastoma is a smooth muscle tumor with borderline malignant activity. Grossly, it is softer than a leiomyoma and contains areas of hemorrhage and necrosis. Only 3 of 26 patients reported by Kurman and Norris developed recurrent disease, of whom 2 responded to therapy.175 Mitotic count correlated with the outcome. Intravenous Leiomyomatosis
Intravenous leiomyomatosis is a smooth muscle tumor characterized by its direct growth into venous channels. It is believed to arise from the muscular wall of veins or from a leiomyoma with vascular invasion and subsequent extension beyond the leiomyoma from which it originated.176 Because of suspected estrogen dependence, oophorectomy is advocated.177 Therapy consists of local excision and includes resection of involved veins. In some cases, excision of lung metastasis and removal of tumor from the vena cava and right atrium have resulted in long-term survival.178 Benign Metastasizing Leiomyoma
Benign metastasizing leiomyoma is a smooth muscle tumor with multiple intrapulmonary nodules. It is thought to arise from a cellular leiomyoma or intravenous leiomyomatosis that gains access to the vascular system and is transplanted to the lungs, where it is implanted and grows. Some authors have suggested that it results after incomplete primary surgery such as curettage, myomectomy, or supracervical hysterotomy.179 Disseminated Peritoneal Leiomyomatosis
Disseminated peritoneal leiomyomatosis is important in that it must be differentiated from metastatic leiomyosarcoma. This tumor is characterized by multiple peritoneal implants, generally measuring less than 1 cm in diameter, which involve the pelvic and abdominal–peritoneal cavity. Leiomyosarcoma, in contrast, is characterized by fewer but larger metastatic lesions. On histologic examination, the lesion appears benign with few mitotic figures and minimal nuclear atypia.180 The lesion is generally associated with an excess of estrogen and is found in conjunction with pregnancy,181 granulosa cell tumor,182 and oral contraceptive use.183 Radical excision is unnecessary, as the lesion regresses following estrogen normalization.132 However, in two cases sarcomatous degeneration has been reported.184
with estrogen use and obesity, they are poorly differentiated, deeply invasive, and frequently have nodal metastasis. Collectively, these tumors usually are of advanced stage and have a poor prognosis. Since this original definition, studies of molecular biology of endometrial carcinomas have demonstrated significant differences between types I and II tumors. Type I tumors are estrogen and progesterone receptor positive, diploid, have a low frequency of allelic imbalance, and demonstrate genetic alterations in K-ras, MLH1, PTEN, and CTNNB1; while type II tumors are estrogen and progesterone receptor negative, aneuploid, have frequent allelic imbalance, and demonstrate genetic alterations in p53 and erb-B2.186 The risk factors for type I tumor have been identified as unopposed estrogen use, nulliparity, diabetes, and tamoxifen use. But since these tumors tend to be of low grade, cancer screening based on these risk factors will not impact overall mortality from endometrial cancer. Unfortunately, the risk factors for type II tumors remain elusive. Controversy exists over the prognosis of patients with uterine papillary serous carcinoma, clear cell carcinoma, and poorly differentiated endometrioid carcinoma, with a poorer prognosis for uterine papillary serous carcinoma and clear cell carcinoma in some but not all series.187 – 189
Uterine Papillary Serous Carcinoma and Clear Cell Carcinoma Endometrial serous adenocarcinoma, resembling ovarian serous adenocarcinoma, represents an unusual and aggressive variant accounting for 5–10% of endometrial carcinomas.186 Serous carcinomas affect older, usually postmenopausal, patients, and often arise in the setting of endometrial atrophy. Microscopically, serous carcinomas usually have complex branching papillae with prominent epithelial stratification yielding epithelial tufts and an apparently detached epithelial cell cluster (see Figure 9). The lining cells usually have marked nuclear atypia. Serous carcinomas tend to invade deeply, extending well beyond the gross extent of tumor, even in small uteri without grossly evident myometrial invasion. In one series, 40% of pathologic stage I tumors involved the outer half of the myometrium.190
UNUSUAL ENDOMETRIAL CARCINOMAS Endometrial carcinomas can be separated into two different pathologic types, as originally suggested by Bokhman et al.185 Type I endometrial carcinomas are indolent tumors that are associated with estrogen use and obesity. Tumors are of low histologic grade (grades 1, 2) with superficial myometrial invasion and infrequent nodal metastasis. Collectively, these tumors are of low stage and have an excellent prognosis. While type II endometrial carcinomas are not associated
Figure 9 Uterine papillary serous adenocarcinomas have a complex papillary architecture with marked cellular stratification, as seen in this example.
UTERINE SARCOMAS AND UNUSUAL ENDOMETRIAL CARCINOMAS
Almost 50% of cases have had vascular invasion. Clear cell adenocarcinoma accounts for approximately 4% of endometrial carcinomas.186,191 Unlike vaginal and cervical clear cell carcinomas, endometrial tumors are not related to in-utero diethylstilbestrol (DES) exposure. Microscopically, clear cell carcinomas appear similar regardless of site of origin in the female genital tract. Microscopically, epithelial cells with vacuolated or clear cytoplasm usually predominate (see Figure 10). Abundant glycogen causes the cytoplasmic clearing. The carcinoma cells are arranged in solid, papillary, or tubulocystic patterns, or often, in a mixture of these patterns.
Management of Uterine Papillary Serous Carcinoma and Clear Cell Carcinoma The only consensus regarding the management of uterine papillary serous carcinoma and clear cell carcinomas is that they should be very carefully staged surgically. Silva and Jenkins reported 16 patients with serous carcinoma limited to endometrial polyps.192 Six patients had evidence of extrauterine disease at exploration. Among 10 patients with disease confined to the uterus, 60% had a recurrence. Goff et al. demonstrated that careful staging identified metastatic disease in 72% of patients with clinical stage I disease.193 Most pronounced in their study was the fact that intraperitoneal disease or nodal metastasis did not correlate with the extent of myometrial invasion. Subsequent studies of carefully staged patients demonstrated a lower risk of recurrence of 14–36%.194 – 196 Throughout the 1980s and 1990s, radiation therapy was the predominant mode of adjuvant therapy utilized for endometrial carcinoma. Discrepant results in the efficacy of adjuvant pelvic radiation therapy appear to be related to the inclusion of comprehensively staged and nonstaged patients.197 In view of the potential for peritoneal spread, many investigators utilized whole abdominal radiation therapy. In the only prospective trial of whole abdominal radiation therapy for patients with stage I or II endometrial carcinoma, the 5year progression-free survival was only 35% with over half the recurrences within the radiation field.198 The benefit of
Figure 10 Vacuolated clear cytoplasm surrounds high-grade nucleus in this clear example of adenocarcinoma.
495
adjuvant radiation has been questioned by Grice et al. whose only stage I recurrence had received radiation therapy and Huh et al. who found no difference between patients treated with radiation or observed.194,195 In view of the minimal morbidity of vaginal brachytherapy and its possible benefit, its use has been advocated.196,197,199 Alternatively, adjuvant chemotherapy has also been advocated for uterine papillary serous carcinoma.200,201 Conflicting results of comparative studies have been reported.196,202 Low et al. reported recurrence in only one of 12 stage I or II patients after adjuvant platinum-based chemotherapy, pelvic radiation, and vaginal brachytherapy.201 For patients with advanced stage papillary serous carcinoma of the endometrium, the prognosis is poor. In a retrospective study of stage IV patients who underwent surgery, those with residual disease of diameter >1 cm, <1 cm, or no residual disease at completion of the surgery had median survivals of 9.6, 20.5, and 30.4 months respectively.203 In this study, patients who received a platinum and paclitaxel combination chemotherapy had a longer median survival (29.1 months) than those who received platinum-based chemotherapy alone (7.6 months). In the past, whole abdominal radiation therapy had been advocated for advanced stage endometrial carcinoma limited to the abdomen with <2 cm residual disease.204,205 The GOG recently reported a phase II trial evaluating whole abdominal radiation therapy in stages III and IV patients with <2 cm disease limited to the abdomen.206 Among 103 patients with uterine papillary serous carcinoma and clear cell carcinoma, the 3-year recurrence-free and overall survival rates were 27 and 35% respectively. No patient with gross residual disease following surgery survived. In view of the recent randomized GOG trial comparing whole abdominal radiation to chemotherapy in advanced stage endometrial carcinoma showing slightly improved outcomes with chemotherapy, we would favor adjuvant carboplatin and paclitaxel for patients with abdominal disease.207 The prognosis of clear cell carcinomas of the endometrium is more controversial. In a large series from Norway collected from 1970 to 1992, 181 patients with clear cell carcinoma of the endometrium were identified.208 The 5-year survivals for stages I, II, III, and IV patients were approximately 55, 28, 15, and 0%, respectively. Two-thirds of patients relapsed outside the pelvis, with many abdominal relapses. Other studies have reported better outcomes for low stage clear cell carcinoma, which is very similar to that of poorly differentiated endometrial carcinoma.209,210 In a more recent series by Murphy et al. of 38 patients with clear cell carcinoma, only 3.8% relapsed outside the pelvis.211 Other rare endometrial carcinoma variants include small cell undifferentiated carcinoma and squamous carcinoma. Small cell undifferentiated carcinoma is a high-grade neuroendocrine carcinoma, morphologically identical to those occurring in the lungs. Endometrial squamous carcinoma must be differentiated from endometrioid adenocarcinoma with extensive squamous differentiation and from cervical squamous carcinoma with extension to the uterine corpus. Limited data suggest a poor prognosis for endometrial squamous carcinoma. Forty
496
GYNECOLOGICAL CANCERS
percent of patients with pathologic stage I tumors died of disease within 3 years of diagnosis.212
CONCLUSION Uterine sarcomas, the most common type of gynecologic sarcoma, are virulent tumors. Their prognosis and subsequent treatment depend on an understanding of their biology. For early disease, therapy consists of primary surgery. The role of adjuvant therapy remains unclear at this time. For advanced disease, chemotherapy is the most effective therapy. Because of rapid tumor dissemination, significant advances in chemotherapy are needed to improve the current survival rates. Certain endometrial carcinomas including uterine papillary serous carcinoma, clear cell carcinoma, and other rare histologies are distinct in their behavior and warrant more aggressive treatment.
REFERENCES 1. Nordal RR, Thoresen SO. Uterine sarcomas in Norway 1956 – 1992: incidence, survival and mortality. Eur J Cancer 1997; 33: 907 – 11. 2. Schwartz Z, et al. Uterine sarcoma in Israel: a study of 104 cases. Gynecol Oncol 1985; 20: 354 – 63. 3. Covens AL, et al. Uterine sarcoma: an analysis of 74 cases. Am J Obstet Gynecol 1987; 156: 370 – 4. 4. Wheelock JB, et al. Uterine sarcoma: analysis of prognostic variables in 71 cases. Am J Obstet Gynecol 1987; 151: 1016 – 22. 5. Harlow BL, Weiss NS, Lofton S. The epidemiology of sarcomas of the uterus. J Natl Cancer Inst 1986; 76: 399 – 402. 6. Brooks SE, et al. Surveillance, epidemiology, and end results analysis of 2677 cases of uterine sarcoma 1989 – 1999. Gynecol Oncol 2004; 93: 204 – 8. 7. Marchese JM, et al. Uterine sarcomas: a clinicopathologic study, 1965 – 1981. Gynecol Oncol 1984; 18: 299 – 312. 8. Schwartz SM, Weiss NS. Martial status and the incidence of sarcomas of the uterus. Cancer Res 1990; 50: 1886 – 9. 9. Press MF, Scully RE. Endometrial sarcomas complicating ovarian thecoma, polycystic ovarian disease and estrogen therapy. Gynecol Oncol 1985; 21: 135 – 54. 10. Altaras MM, et al. Role of prolonged excessive estrogen stimulation in the pathogenesis of endometrial sarcomas: two cases and a review of the literature. Gynecol Oncol 1990; 38: 273 – 7. 11. Altaras MM, et al. Role of prolonged stimulation of tamoxifen therapy in the etiology of endometrial sarcomas. Gynecol Oncol 1993; 49: 255 – 8. 12. Eddy GL, Mazur MT. Endolymphatic stromal myosis associated with tamoxifen use. Gynecol Oncol 1997; 64: 262 – 4. 13. Schwartz SM, et al. Exogenous sex hormone use, correlates of endogenous hormone levels and the incidence of histologic types of sarcoma of the uterus. Cancer 1996; 77: 717 – 24. 14. Thomas WO, Harris HH, Enden JA. Postirradiation malignant neoplasms of the uterine fundus. Am J Obstet Gynecol 1969; 104: 209 – 19. 15. Fehr PE, Prem KA. Malignancy of the uterine corpus following irradiation therapy for squamous cell carcinoma of the cervix. Am J Obstet Gynecol 1974; 119: 685 – 92. 16. Czesnin K, Wronkowski Z. Second malignancies of the irradiated area in patients treated for uterine cervix cancer. Gynecol Oncol 1978; 6: 309 – 15. 17. Mark RJ, et al. Postirradiation sarcoma of the gynecologic tract. A report of 13 cases and a discussion of the risk of radiation-induced gynecologic malignancies. Am J Clin Oncol 1996; 19: 59 – 64. 18. Boice JD, et al. Second cancers following radiation treatment for cervical cancer: an international collaboration among cancer registries. J Natl Cancer Inst 1985; 74: 955 – 75.
19. Montague AC, Swartz DP, Woodruff JD. Sarcoma arising in a leiomyoma of the uterus. Am J Obstet Gynecol 1965; 92: 421 – 7. 20. Torpin R, Pund E, Peeples WJ. The etiologic and pathologic factors in a series of 1,741 fibromyomas of the uterus. Am J Obstet Gynecol 1942; 44: 569 – 74. 21. Aaro LA, Symmonds RE, Dockerty MB. Sarcoma of the uterus: A clinical and pathologic study of 177 cases. Am J Obstet Gynecol 1966; 94: 101 – 9. 22. Lui FS, et al. Mutation and overexpression of the p53 tumor suppressor gene frequently occurs in uterine and ovarian sarcomas. Obstet Gynecol 1994; 83: 118 – 24. 23. Costa MJ, Vogelsan J, Young LJT. p53 gene mutation in female genital tract carcinosarcomas (malignant mixed m¨ullerian tumors): a clinicopathologic study of 74 cases. Mod Pathol 1994; 7: 619 – 27. 24. Risinger JI, et al. Microsatellite instability in gynecological sarcomas and in hMSH2 mutant uterine sarcoma cell lines defective in mismatch repair activity. Cancer Res 1995; 55: 5664 – 9. 25. Wang L, et al. The proto-oncogene c-kit is expressed in leiomyosarcomas of the uterus. Gynecol Oncol 2003; 90: 402 – 6. 26. Winter WE 3rd, et al. Clinicopathological analysis of c-kit expression in carcinosarcomas and leiomyosarcomas of the uterine corpus. Gynecol Oncol. 2003; 91: 3 – 8. 27. Klein WM, Kurman RJ. Lack of expression of c-kit protein (CD117) in mesenchymal tumors of the uterus and ovary. Int J Gynecol Pathol 2003; 22: 181 – 4. 28. Rushing RS, et al. Uterine sarcomas express KIT protein but lack mutation(s) in exon 11 or 17 of c-KIT. Gynecol Oncol 2003; 91: 9 – 14. 29. Raspollini MR, et al. Uterine leiomyosarcomas express KIT protein but lack mutation(s) in exon 9 of c-KIT. Gynecol Oncol 2005; 98: 334 – 5. 30. Bodner-Adler B, et al. MMP-1 and MMP-2 expression in uterine leiomyosarcoma and correlation with different clinicopathologic parameters. J Soc Gynecol Investig 2003; 10: 443 – 6. 31. Doss LL, Llorens AS, Hernandez EM. Carcinosarcoma of the uterus: a 40 year experience from the state of Missouri. Gynecol Oncol 1984; 18: 43 – 53. 32. Parker WH, Fu Yao S, Berek JS. Uterine sarcoma in patients operated on for presumed leiomyoma and rapidly growing leiomyoma. Obstet Gynecol 1994; 83: 414 – 8. 33. Meyer WR, et al. Unsuspected leiomyosarcoma: treatment with a gonadotropin-releasing hormone analogue. Obstet Gynecol 1990; 75: 529 – 31. 34. Hitti IF, et al. Uterine leiomyosarcoma with massive necrosis diagnosed during gonadotropin-releasing hormone analog therapy for presumed uterine fibroid. Fertil Steril 1991; 56: 779 – 80. 35. Schwartz LB, Diamond MP, Schwartz PE. Leiomyosarcomas: Clinical presentation. Am J Obstet Gynecol 1993; 168: 180 – 3. 36. Massoni EA, Hajdu SI. Cytology of primary and metastatic uterine sarcomas. Acta Cytol 1984; 28: 93 – 100. 37. Muhm JR, et al. Comparison of whole lung tomography and computed tomography for detecting pulmonary nodules. Am J Roentgenol 1978; 131: 981 – 4. 38. Takemori M, Nishimura R, Sugimura K. Magnetic resonance imaging of uterine leiomyosarcoma. Arch Gynecol Obstet 1992; 251: 215 – 8. 39. Ober WB. Uterine sarcomas: histogenesis and taxonomy. Ann N Y Acad Sci 1959; 75: 568 – 85. 40. Kempson RL, Bari W. Uterine sarcomas: classification, diagnosis and prognosis. Hum Pathol 1970; 1: 331 – 49. 41. Silverberg SG. Leiomyosarcoma of the uterus: a clinicopathologic study. Obstet Gynecol 1971; 38: 613 – 28. 42. Taylor HB, Norris HJ. Mesenchymal tumors of the uterus. Arch Pathol 1966; 82: 40 – 4. 43. Leibsohn S, et al. Leiomyosarcoma in a series of hysterectomies performed for presumed uterine leiomyomas. Am J Obstet Gynecol 1990; 162: 968 – 76. 44. Barter JF, et al. Leiomyosarcoma of the uterus: clinicopathologic study of 21 cases. Gynecol Oncol 1985; 21: 220 – 7. 45. Hannigan EV, Gomez LG. Uterine leiomyosarcoma: a review of prognostic clinical and pathologic features. Am J Obstet Gynecol 1979; 134: 557 – 64.
UTERINE SARCOMAS AND UNUSUAL ENDOMETRIAL CARCINOMAS 46. Perrone T, Dehner LP. Prognostically favorable ‘mitotically active’ smooth-muscle tumors of the uterus. Am J Surg Pathol 1988; 12: 1 – 8. 47. O’Connor DM, Norris HJ. Mitotically active leiomyomas of the uterus. Hum Pathol 1990; 21: 223 – 7. 48. Prayson RA, Hart WR. Mitotically active leiomyomas of the uterus. Am J Clin Pathol 1992; 97: 14 – 20. 49. Peters WA, et al. Uterine smooth-muscle tumors of uncertain malignant potential. Obstet Gynecol 1994; 83: 1015 – 20. 50. Bell SW, Kempson RL, Hendrickson MR. Problematic uterine smooth muscle neoplasms: a clinicopathologic study of 213 cases. Am J Surg Pathol 1994; 18: 535 – 58. 51. King ME, Dickersin GR, Scully RE. Myxoid leiomyosarcoma of the uterus: a report of six cases. Am J Surg Pathol 1982; 6: 589 – 98. 52. Chen KTK. Myxoid leiomyosarcoma of the uterus. Int J Gynecol Pathol 1984; 3: 389 – 92. 53. Tavassoli FA, Devilee P. World Health Organization Classification of Tumors. Pathology and Genetics. Tumors of the Breast and Female Genital Organs Leon, France: IARC Press, 2003. 54. Chang KL, et al. Primary uterine endometrial stromal neoplasms. Am J Surg Pathol 1990; 14: 415 – 38. 55. Norris HJ, Taylor HB. Mesenchymal tumors of the uterus: a clinical and pathological study of 53 endometrial stromal tumors. Cancer 1966; 19: 755 – 66. 56. Sutton GP, et al. Estrogen and progesterone receptors in uterine sarcomas. Obstet Gynecol 1986; 68: 709 – 14. 57. Altrabulsi B, et al. Undifferentiated carcinoma of the endometrium. Am J Surg Pathol 2005; 29: 1316 – 21. 58. Silverberg SG, et al. A. Carcinosarcoma (malignant mixed mesodermal tumor) of the uterus. Int J Gynecol Pathol 1990; 9: 1 – 19. 59. Sreenan JJ, Hart WR. Carcinosarcomas of the female genital tract. Am J Surg Pathol 1995; 19: 666 – 74. 60. Norris HJ, Taylor HB. Mesenchymal tumors of the uterus: a clinical and pathologic study of 31 carcinosarcomas. Cancer 1966; 19: 1459 – 65. 61. Chuang JT, VanVelden DJJ, Graham JB. Carcinosarcoma and mixed mesodermal tumor of the uterine corpus: a review of 49 cases. Obstet Gynecol 1970; 35: 769 – 79. 62. Itsuo G, Chie D, Minaguchi H. Establishment and characterization of carcinosarcoma cell line of the human uterus. Cancer 1993; 71: 775 – 86. 63. Amant F, et al. Endometrial carcinosarcomas have a different prognosis and pattern of spread compared to high risk epithelial endometrial cancer. Gynecol Oncol 2005; 98: 274 – 80. 64. DiSaia PJ, Castro JR, Rutledge FN. Mixed mesodermal sarcoma of the uterus. Am J Roentgenol 1973; 117: 632 – 6. 65. Major FJ, et al. Prognostic factors in early-stage uterine sarcoma. Cancer 1993; 71: 1702 – 9. 66. Koss LG, Spiro RH, Brunschwig A. Endometrial stromal sarcoma. Surg Gynecol Obstet 1965; 121: 531 – 7. 67. Norris HJ, Roth E, Taylor HB. Mesenchymal tumors of the uterus: a clinical and pathologic study of 31 mixed mesodermal tumors. Obstet Gynecol 1966; 28: 57 – 63. 68. Peters WA, et al. Prognostic features of sarcomas and mixed tumors of the endometrium. Obstet Gynecol 1984; 63: 550 – 6. 69. Patsner B, Mann WJ. Use of serum CA-125 in monitoring patients with uterine sarcoma: a preliminary report. Cancer 1988; 61: 1355 – 8. 70. Clement PB, Scully RE. Mullerian adenosarcoma of the uterus: a clinicopathologic analysis of 100 cases with a review of the literature. Hum Pathol 1990; 21: 363 – 81. 71. Krivak TC, et al. Uterine adenosarcoma with sarcomatous overgrowth versus uterine carcinosarcoma: comparison of treatment and survival. Gynecol Oncol 2001; 83: 89 – 94. 72. Wolfson AH, et al. A multivariate analysis of clinicopathologic factors for predicting outcome in uterine sarcomas. Gynecol Oncol 1994; 52: 56 – 62. 73. Nola M, et al. Prognostic parameters for survival of patients with malignant mesenchymal tumors of the uterus. Cancer 1996; 78: 2543 – 50. 74. Lennart K, et al. Flow cytometric analysis of uterine sarcomas. Gynecol Oncol 1994; 55: 339 – 42.
497
75. August CZ, et al. Neoplasms of endometrial stroma: histopathologic and flow cytometric analysis with clinical correlation. Hum Pathol 1989; 20: 232 – 7. 76. Zoloudek C, Norris HJ. Mesenchymal tumors of the uterus. In Kurman RJ (ed) Blaostein’s Pathology of the Female Genital Tract, 4th ed. Springer-Verlag, Chap. 13, 519. 77. Oliva E, et al. Mixed endometrial stromal and smooth muscle tumors of the uterus: a clinicopathologic study of 15 cases. Am J Surg Pathol 1998; 22: 997 – 1005. 78. Vang R, Kempson RL. Perivascular epithelioid cell tumor (PEComa) of the uterus: a subset of HMB-45-positive epithelioid mesenchymal neoplasms with an uncertain relationship to pure smooth muscle tumors. Am J Surg Pathol 2002; 26: 1 – 13. 79. Macasaet MA, et al. Prognostic factors in malignant mesodermal (mullerian) mixed tumors of the uterus. Gynecol Oncol 1985; 20: 32 – 42. 80. Salazar OM, et al. Uterine sarcomas: analysis of failures with special emphasis on the use of adjuvant radiation therapy. Cancer 1978; 42: 1161 – 70. 81. Fleming WP, et al. Autopsy findings in patients with uterine sarcoma. Gynecol Oncol 1984; 19: 168 – 72. 82. Spanos WJ, Peters LJ, Oswald MJ. Patterns of recurrence in malignant mixed mullerian tumor of the uterus. Cancer 1986; 57: 155 – 9. 83. Geszler G, et al. Prognostic value of peritoneal washings in patients with malignant mixed mullerian tumors of the uterus. Am J Obstet Gynecol 1986; 155: 83 – 9. 84. Kanbour AI, et al. Peritoneal cytology in malignant mixed mullerian tumors of the uterus. Gynecol Oncol 1989; 33: 91 – 5. 85. Rose PG, et al. Patterns of metastasis in uterine sarcoma: an autopsy study. Cancer 1989; 63: 935 – 8. 86. Belgrad R, Elbadawi N, Rubin P. Uterine sarcoma. Radiology 1975; 114: 181 – 8. 87. Chen SS. Propensity of retroperitoneal lymph node metastasis in patients with stage I sarcoma of the uterus. Gynecol Oncol 1989; 32: 215 – 7. 88. Goff BA, et al. Uterine leiomyosarcoma and endometrial stromal sarcoma: lymph node metastases and sites of recurrence. Gynecol Oncol 1993; 50: 105 – 9. 89. Parente JT, et al. Leiomyosarcoma of the uterus with pulmonary metastases: a favorable response to operation and chemotherapy in a patient monitored with serial carcinoembryonic antigen. Am J Obstet Gynecol 1978; 131: 812 – 5. 90. Hannigan EV, et al. Reexploration after treatment for uterine sarcoma. Gynecol Oncol 1983; 16: 1 – 5. 91. Van Dinh T, Woodruff JD. Leiomyosarcoma of the uterus. AM J Obstet Gynecol 1982; 144: 817 – 23. 92. Berchuck A, et al. Treatment of uterine leiomyosarcoma. Obstet Gynecol 1988; 71: 845 – 50. 93. Badib AO, et al. Radiotherapy in the treatment of sarcomas of the corpus uteri. Cancer 1969; 24: 724 – 9. 94. Perez CA, Askin F, Baglan RJ. Effects of irradiation on mixed mullerian tumors of the uterus. Cancer 1979; 43: 1274 – 84. 95. Gershenson DM, et al. Cisplatin therapy for disseminated mixed mesodermal sarcoma of the uterus. J Clin Oncol 1987; 5: 618 – 21. 96. Omura GA, et al. A randomized clinical trial of adjuvant Adriamycin in uterine sarcomas: a Gynecologic Oncology Group study. J Clin Oncol 1985; 9: 1240 – 5. 97. Hornback NB, Omura G, Major FJ. Observations on the use of adjuvant radiation therapy in patients with stage I and II uterine sarcoma. Int J Radiat Oncol Biol Phys 1986; 12: 2127 – 30. 98. Gilbert HA, et al. The value of radiation therapy in uterine sarcoma. Obstet Gynecol 1975; 45: 84 – 8. 99. Vongtama V, et al. Treatment results and prognostic factors in stage I and II sarcomas of the corps uteri. Am J Roentgenol Radium Ther Nucl Med 1976; 126: 139 – 47. 100. Rose PG, Boutselis JG, Sachs L. Adjuvant therapy for stage I uterine sarcoma. Am J Obstet Gynecol 1987; 156: 660 – 2. 101. Knocke TH, et al. Results of postoperative radiotherapy in the treatment of sarcoma of the corpus uteri. Cancer 1998; 83: 1972 – 9. 102. Kolstad P. Adjuvant chemotherapy in sarcoma of the uterus: a preliminary report. In Morrow CP, et al. (eds) Recent Clinical
498
103.
104.
105.
106. 107.
108.
109. 110.
111.
112. 113.
114. 115.
116. 117.
118. 119.
120.
121.
122. 123. 124. 125.
126.
127.
GYNECOLOGICAL CANCERS Developments in Gynecologic Oncology. New York: Raven Press, 1983:: 123 – 129. Omura GA, et al. A randomized study of adriamycin with and without dimethyl triazenoimidazole carboxamide in advanced uterine sarcomas. Cancer 1983; 52: 626 – 32. Tierney JF, et al. Adjuvant chemotherapy for localized resectable softtissue sarcoma of adults: meta-analysis of individual data. Lancet 1997; 350(ii): 1647 – 54. Sutton G, et al. Adjuvant ifosfamide and cisplatin in patients with completely resected stage I or II carcinosarcomas (mixed mesodermal tumors) of the uterus: a Gynecologic Oncology Group study. Gynecol Oncol 2005; 96: 630 – 4. Buchsbaum HJ, Lifshitz S, Blythe JG. Prophylactic chemotherapy in stages I and II uterine sarcoma. Gynecol Oncol 1979; 8: 346 – 8. van Nagell JR, et al. Adjuvant vincristine, dactinomycin, and cyclophosphamide therapy in stage I uterine sarcomas. Cancer 1986; 57: 1451 – 4. Piver MS, et al. The effect of adjuvant chemotherapy on time to recurrence and survival of stage I uterine sarcomas. J Surg Oncol 1988; 38: 233 – 9. Kohorn EI, et al. Adjuvant therapy in mixed m¨ullerian tumors of the uterus. Gynecol Oncol 1986; 23: 212 – 21. Manolitsas TP, et al. Multimodality therapy for patients with clinical stage I and II malignant mixed M¨ullerian tumors of the uterus. Cancer 2001; 91: 1437 – 43. Pautier P, et al. Adjuvant chemotherapy with cisplatin, ifosfamide, and doxorubicin followed by radiotherapy in localized uterine sarcomas: results of a case-control study with radiotherapy alone. Int J Gynecol Cancer 2004; 14: 1112 – 7. Baggish MS, Woodruff JD. Uterine stomatosis: clinicopathologic features and hormone dependency. Obstet Gynecol 1972; 40: 487 – 90. Krumholts BA, Lobovsky FYI, Halitsky V. Endolymphatic stromal myosis with pulmonary metastasis, remission with progestin therapy: report of a case. J Reprod Med 1973; 10: 85 – 9. Thatcher SS, Woodruff JD. Uterine stromatosis: a report of 33 cases. Obstet Gynecol 1982; 59: 428 – 34. Gloor E, Schnyder P, Cikes M. Endolymphatic stromal myosis: surgical and hormonal treatment of extensive abdominal recurrence 20 years after hysterectomy. Cancer 1982; 50: 1888 – 93. Piver MS, et al. Uterine endolymphatic stromal myosis: a collaborative study. Obstet Gynecol 1984; 64: 173 – 8. Tsukamoti N, et al. Endolymphatic stromal myosis: a case with positive estrogen and progesterone receptors and good response to progestins. Gynecol Oncol 1985; 20: 120 – 8. Yazigi R, Piver MS, Barlow JJ. Stage III Uterine sarcoma: case report and literature review. Gynecol Oncol 1979; 8: 92 – 6. Azizi F, et al. Remission of uterine leiomyosarcomas treated with vincristine, adriamycin and dimethyltriazeno-imidazole carboxamide. Am J Obstet Gynecol 1979; 133: 379 – 81. Muss HB, et al. Treatment of recurrent or advanced uterine sarcoma: a randomized trial of doxorubicin versus doxorubicin and cyclophosphamide (a phase III trial of the Gynecologic Oncology Group). Cancer 1985; 55: 1648 – 53. Barber HRK, Brunschwig AA. Surgical approach to endometrial sarcoma: left inferior pulmonary lobectomy for metastases 8 1/2 years after anterior pelvic exenteration. Obstet Gynecol 1965; 26: 821 – 5. Levenback C, et al. Resection of pulmonary metastases from uterine sarcomas. Gynecol Oncol 1992; 45: 202 – 5. Reid GC, et al. The role of pelvic exenteration for sarcomatous malignancies. Obstet Gynecol 1989; 74: 80 – 4. Smith JP, et al. Combined irradiation and chemotherapy for sarcomas of the pelvis in females. Am J Roentgenol 1975; 123: 571 – 6. Sutton GP, et al. Phase II trial of ifosfamide and mesna in leiomyosarcoma of the uterus: a gynecologic oncology group study. Am J Obstet Gynecol 1992; 166: 556 – 9. Look KY, et al. Phase II trial of gemcitabine as second-line chemotherapy of uterine leiomyosarcoma: a Gynecologic Oncology Group (GOG) study. Gynecol Oncol 2004; 92: 644 – 7. Sutton G, et al. Phase II evaluation of liposomal doxorubicin (DOXIL) in recurrent or advanced leiomyosarcoma of the uterus: a Gynecologic Oncology Group Study. Gynecol Oncol 2005; 96: 749 – 52.
128. Thigpen JT, Blessing JA, Wilbanks GD. Cisplatin as second line chemotherapy in the treatment of advanced or recurrent leiomyosarcoma of the uterus. Am J Clin Oncol 1986; 9: 18 – 20. 129. Thigpen JT, et al. Phase II trial of cisplatin as first-line chemotherapy in patients with advanced or recurrent uterine sarcomas: a Gynecologic Oncology Group Study. J Clin Oncol 1991; 9: 1962 – 6. 130. Thigpen JT, et al. Phase II trial of cisplatin in the treatment of patients with advanced or recurrent mixed mesodermal tumor of the uterus: a Gynecologic Oncology Group study. Cancer Treat Rep 1986; 70: 271 – 4. 131. Sutton GP, et al. Phase II trial of ifosfamide and mesna in mixed mesodermal tumors of the uterus (a Gynecologic Oncology Group Study). Am J Obstet Gynecol 1989; 161: 309 – 12. 132. Sutton GP, et al. A phase II trial of ifosfamide and mesna in patients with advanced or recurrent mixed mesodermal tumors of the ovary previously treated with platinum-based chemotherapy: a gynecologic oncology group study. Gynecol Oncol 1994; 53: 24 – 6. 133. Curtin JP, et al. Paclitaxel in the treatment of carcinosarcoma of the uterus: a gynecologic oncology group study. Gynecol Oncol 2001; 83: 268 – 70. 134. Asbury R, et al. Amonifide in patients with leiomyosarcoma of the uterus: a phase II gynecologic oncology group study. Am J Clin Oncol 1998; 21: 145 – 6. 135. Slayton R, et al. Phase II trial of etoposide in the management of advanced and recurrent leiomyosarcoma of the uterus: a Gynecologic Oncology Group Study. Cancer Treat Rep 1987; 71: 1303 – 4. 136. Rose PG, et al. Prolonged oral etoposide in recurrent or advanced leiomyosarcoma of the uterus: a gynecologic oncology group study. Gynecol Oncol 1998; 70: 267 – 71. 137. Thigpen T, et al. Phase II trial of etoposide in leiomyosarcoma of the uterus: a Gynecologic Oncology Group study. Gynecol Oncol 1996; 63: 120 – 2. 138. Gallup DG, et al. Evaluation of paclitaxel in previously treated leiomyosarcoma of the uterus: A Gynecologic Oncology Group study. Gynecol Oncol 2003; 89: 48 – 51. 139. Sutton G, Blessing JA, Ball H. Phase II trial of paclitaxel in leiomyosarcoma of the uterus: a gynecologic oncology group study. Gynecol Oncol 1999; 74: 346 – 9. 140. Thigpen JT, et al. Phase II trial of piperazinedione with advanced or recurrent mixed mesodermal uterine sarcoma. Am J Clin Oncol 1985; 8: 350 – 2. 141. Miller DS, et al. Phase II trial of topotecan in patients with advanced, persistent or recurrent uterine leiomyosarcomas: A Gynecologic Oncology Group Study. Am J Clin Oncol 2000; 23: 355 – 7. 142. Smith HO, Blessing JA, Vacarello L. Trimetrexate in the treatment of recurrent or advanced leiomyosarcoma of the uterus: A Phase II study of the Gynecologic Oncology Group. Gynecol Oncol 2002; 84: 140 – 4. 143. Asbury R, et al. A phase II trial of amonifide in patients with mixed mesodermal tumors of the uterus: a gynecologic oncology group study. Am J Clin Oncol 1998; 21: 306 – 7. 144. Slayton RE, Blessing JA, Clarke-Pearson D. A phase II trial of diaziquone (AZQ) in mixed mesodermal sarcomas of the uterus: a Gynecologic Oncology Group Study. Invest New Drugs 1991; 9: 93 – 4. 145. Gershenson DM, et al. High-dose doxorubicin infusion therapy for disseminated mixed mesodermal sarcoma of the uterus. Cancer 1987; 59: 1264 – 7. 146. Slayton RE, et al. Phase II trial of etoposide in the management of advanced or recurrent mixed mesodermal sarcomas of the uterus: a Gynecologic Oncology Group Study. Cancer Treat Rep 1987; 71: 661 – 2. 147. Fowler JM, et al. Phase II evaluation of oral trimetrexate in mixed mesodermal tumors of the uterus: a gynecologic oncology group study. Gynecol 2002; 85: 311 – 4. 148. Miller DS, et al. Phase II evaluation of topotecan in carcinosarcoma of the uterus: a gynecologic oncology group study. Gynecol Oncol 2005; 98: 217 – 21. 149. Curry JL, et al. Phase II trial of hydroxyurea, dacarbazine (DTIC), and etoposide (VP-16) in mixed mesodermal tumors of the uterus: a gynecologic oncology group study. Gynecol Oncol 1996; 61: 94 – 6.
UTERINE SARCOMAS AND UNUSUAL ENDOMETRIAL CARCINOMAS 150. Baker TR, et al. Prospective trial of cisplatin adriamycin and dacarbazine in the metastatic mixed mesodermal sarcomas of the uterus and ovary. Am J Clin Oncol 1991; 14: 246 – 50. 151. Piver MS, Lele SB, Patsner B. Cisdiamminedichloroplatinum plus dimethyltriazenoimidazole carboxamide as second and third line chemotherapy for sarcomas of the female pelvis. Gynecol Oncol 1986; 23: 371 – 5. 152. Piver MS, et al. Cyclophosphamide, vincristine, adriamycin and dimethyl-triazeno imidazole carboxamide (CYVADIC) for sarcomas of the female genital tract. Gynecol Oncol 1982; 14: 319 – 23. 153. Sutton G, Blessing JA, Malfetano JH. Ifosfamide and doxorubicin in the treatment of advanced leiomyosarcomas of the uterus: a Gynecologic Oncology Group study. Gynecol Oncol 1996; 62: 226 – 9. 154. Resnik E, et al. Malignant uterine smooth muscle tumors: role of etoposide, cisplatin, and doxorubicin (EPA) chemotherapy. J Surg Oncol 1996; 63: 145 – 7. 155. Hensley ML, et al. Gemcitabine and docetaxel in patients with unresectable leiomyosarcoma: results of a phase II trial. J Clin Oncol 2002; 20: 2824 – 31. 156. Edmonson JH, et al. Phase II study of mitomycin, doxorubicin, and cisplatin in the treatment of advanced uterine leiomyosarcoma: a Gynecologic Oncology Group study. Gynecol Oncol 2002; 85: 507 – 10. 157. Currie J, et al. Combination chemotherapy with hydroxyurea, dacarbazine (DTIC), and etoposide in the treatment of uterine leiomyosarcoma: a gynecologic oncology group study. Gynecol Oncol 1996; 61: 27 – 30. 158. Hannigan EV, et al. Rutledge FN. Treatment of advanced uterine sarcoma with vincristine, actinomycin-D and cyclophosphamide. Gynecol Oncol 1983; 115: 224 – 9. 159. Toyoshima M, et al. Clinical experience with combination paclitaxel and carboplatin therapy for advanced or recurrent carcinosarcoma of the uterus. Gynecol Oncol 2004; 94: 774 – 8. 160. Sutton G, et al. A phase III trial of ifosfamide with or without cisplatin in carcinosarcoma of the uterus: A Gynecologic Oncology Group Study. Gynecol Oncol 2000; 79: 147 – 53. 161. Seltzer V, et al. Doxorubicin and cisplatin in the treatment of advanced mixed mesodermal uterine sarcoma. Cancer Treat Rep 1984; 68: 1389 – 90. 162. Jansen RL, et al. Cyclophosphamide, hexamethylmelamine, adriamycin and cisplatin combination chemotherapy in mixed mesodermal sarcoma of the female genital tract. Eur J Cancer Clin Oncol 1987; 23: 1131 – 3. 163. McMeekin S, et al. A phase II study of thalidomide in patients with recurrent or persistent leiomyosarcoma (LMS) of the uterus: a gynecologic oncologic group (GOG) study. Gynecol Oncol 2004; 92: 440 (abstract 103). 164. Tsuda N, et al. Vaccination with predesignated or evidence-based peptides for patients with recurrent gynecologic cancers. J Immunother 2004; 27: 60 – 72. 165. Hernando JJ, et al. Vaccination with autologous tumour antigenpulsed dendritic cells in advanced gynecological malignancies: clinical and immunological evaluation of a phase I trial. Cancer Immunol Immunother 2002; 51: 45 – 52. 166. Chorlton I, Karnei RF, Norris HJ. Primary malignant reticuloendothelial disease involving the vagina, cervix and corpus uteri. Obstet Gynecol 1974; 44: 735 – 48. 167. Lathrop JC. Malignant pelvic lymphomas. Obstet Gynecol 1967; 30: 137 – 45. 168. Rosenberg SA, et al. Lymphosarcoma: a review of 1269 cases. Medicine 1961; 40: 31 – 76. 169. Harris NL, Scully RE. Malignant lymphoma and granulocytic sarcoma of the uterus and vagina. Cancer 1984; 53: 2530 – 45. 170. Osborne BM, Robboy SJ. Lymphomas or leukemia presenting as ovarian tumors: an analysis of 42 cases. Cancer 1983; 52: 1933 – 43. 171. Crisp WE, et al. Malignant pelvic lymphoma. Am J Obstet Gynecol 1982; 143: 69 – 74. 172. Friedman HD, et al. Granulocytic sarcoma of the uterine cervix – literature review of granulocytic sarcoma of the female genital tract. Gynecol Oncol 1992; 46: 128 – 37.
499
173. Wilbanks GD, Szymanska Z, Miller AW. Pelvic hemangiopericytoma: report of 4 patients and review of the literature. Am J Obstet Gynecol 1975; 123: 555 – 69. 174. Smullens SN, et al. Preoperative embolization of retroperitoneal hemangiopericytomas as an aid of their removal. Cancer 1982; 50: 1870 – 5. 175. Kurman RJ, Norris HJ. Mesenchymal tumors of the uterus: epithelioid smooth muscle tumors including leiomyoblastoma and clear cell leiomyoma. Cancer 1976; 37: 1853 – 65. 176. Norris HJ, Parmley T. Mesenchymal tumors of the uterus: intravenous leiomyomatosis. Cancer 1975; 36: 2164 – 78. 177. Evans AT, Symmonds RE, Gaffey TA. Recurrent pelvic intravenous leiomyomatosis. Obstet Gynecol 1981; 57: 260 – 4. 178. Tierney WM, et al. Intravenous leiomyomatosis of the uterus with extension into the heart. Am J Med 1980; 69: 471 – 5. 179. Abell MR, Littler ER. Benign metastasizing uterine leiomyoma: multiple lymph node metastases. Cancer 1975; 36: 2206 – 13. 180. Goldberg MF, Hurt WG, Frable WJ. Leiomyomatosis peritonealis disseminata: report of a case and review of the literature. Obstet Gynecol 1977; 49: 46S – 52S. 181. Williams LJ, Pavlick FJ. Leiomyomatosis peritonealis disseminata: two case reports and a review of the medical literature. Cancer 1980; 45: 1726 – 33. 182. Wilson JR, Peale AR. Multiple peritoneal leiomyomas associated with a granulosa-cell tumor of the ovary. Am J Obstet Gynecol 1952; 64: 204 – 8. 183. Tavassoli FA, Norris HJ. Peritoneal leiomyomatosis (leiomyomatosis peritonealis disseminata): a clinicopathologic study of 20 cases with ultrastructural observations. Int J Gynecol Pathol 1982; 1: 59 – 74. 184. Akkersdijk GJM, et al. Malignant leiomyomatosis peritonealis disseminata. Am J Obstet Gynecol 1990; 163: 591 – 3. 185. Bokhman JV. Two pathogenetic types of endometrial carcinoma. Gynecol Oncol 1983; 15: 10 – 7. 186. Sidaway MK, Silverberg SG. Endometrial carcinoma: Pathologic factors of therapeutic and prognostic significance. Pathol Annu 1992; 27: 153 – 85. 187. Cirisano FD Jr, et al. The outcome of stage I-II clinically and surgically staged papillary serous and clear cell endometrial cancers when compared with endometrioid carcinoma. Gynecol Oncol 2000; 77: 55 – 65. 188. Creasman WT, et al. Prognosis of papillary serous, clear cell, and grade 3 stage I carcinoma of the endometrium. Gynecol Oncol 2004; 95: 593 – 6. 189. Alektiar KM, et al. Is there a difference in outcome between stage III endometrial cancer of papillary serous/clear cell and endometrioid FIGO Grade 3 cancer? Int J Radiat Oncol Biol Phys 2002; 54: 79 – 85. 190. Hendrickson M, et al. Uterine papillary serous carcinoma: a highly malignant form of endometrial adenocarcinoma. Am J Surg Pathol 1982; 6: 93 – 108. 191. Silverberg SG, DeGiorgi LS. Clear cell carcinoma of the endometrium. Clinical, pathologic, and ultrastructural findings. Cancer 1973; 31: 1127 – 40. 192. Silva EG, Jenkins R. Serous carcinoma in endometrial polyps. Mod Pathol 1990; 3: 120 – 8. 193. Goff BA, et al. Uterine papillary serous carcinoma: patterns of metastatic spread. Gynecol Oncol 1994; 54: 264 – 8. 194. Grice J, et al. Uterine papillary serous carcinoma: evaluation of longterm survival in surgically staged patients. Gynecol Oncol 1998; 69: 69 – 73. 195. Huh WK, et al. Uterine papillary serous carcinoma: comparisons of out outcomes of surgical stage I patients with and without adjuvant therapy. Gynecol Oncol 2003; 88: 209 (abstract 110). 196. Kelly MG, et al. Improved survival in surgical stage I patients with uterine papillary serous carcinoma (UPSC) treated with adjuvant platinum-based chemotherapy. Gynecol Oncol 2005; 98: 353 – 9. 197. Goff BA. Uterine papillary serous carcinoma: what have we learned over he past quarter century? Gynecol Oncol 2005; 98: 341 – 3. 198. Sutton G, et al. Adjuvant whole abdominal irradiation in clinical stages I and II papillary serous or clear cell carcinoma of the endometrium: a phase II study of the gynecologic oncology group. Gynecol Oncol 2005; (in press).
500
GYNECOLOGICAL CANCERS
199. DuBeshter B, et al. High-dose rate brachytherapy for stage I/II papillary serous or clear cell endometrial cancer. Gynecol Oncol 2004; 94: 383 – 6. 200. Zanotti KM, et al. The use of paclitaxel and platinum-based chemotherapy in uterine papillary serous carcinoma. Gynecol Oncol 1999; 74: 272 – 7. 201. Low JS, et al. Adjuvant sequential chemotherapy and radiotherapy in uterine papillary serous carcinoma. Gynecol Oncol 2005; 97: 171 – 7. 202. Elit L, et al. Optimal management for surgically stage 1 serous cancer of the uterus. Gynecol Oncol 2004; 92: 240 – 6. 203. Bristow RE, Duska LR, Montz JF. The role of cytoreductive surgery in the management of stage IV uterine papillary serous carcinoma. Gynecol Oncol 2001; 81: 92 – 9. 204. Greer BE, Hamburger AD. Treatment of intraperitoneal metastatic adenocarcinoma of the endometrium by whole-abdominal moving strip technique and pelvic boost irradiation. Gynecol Oncol 1983; 16: 365 – 73. 205. Martinez AA, et al. Improved outcome at 10 years for serouspapillary/clear cell or high-risk endometrial cancer patients treated with adjuvant high-dose abdomino-pelvic irradiation. Gynecol Oncol 2003; 90: 537 – 46.
206. Sutton G, et al. Whole abdominal radiotherapy in the adjuvant treatment of patients with stage III and IV endometrial cancer: a gynecologic oncology group study. Gynecol Oncol 2005; 97: 755 – 63. 207. Randall ME, et al. Whole abdominal radiotherapy versus combination doxorubicin-cisplatin chemotherapy in advanced endometrial carcinoma. A randomized phase III trial of the gynecologic oncology group. Proc Am Soc Clin Oncol 2003; 22: 2 (abstract #3). 208. Abeler VM, et al. Clear cell carcinoma of the endometrium: prognosis and metastatic pattern. Cancer 1996; 78: 1740 – 7. 209. Malpica An, et al. Low-stage clear-cell carcinoma of the endometrium. Am J Surg Pathol 1995; 19: 769 – 74. 210. Carcangiu ML, Chambers JT. Early pathologic stage clear cell carcinoma and uterine papillary serous carcinoma of the endometrium: comparison of clinicopathologic features and survival. Int J Gynecol Pathol 1995; 14: 30 – 8. 211. Murphy KT, et al. Outcome and patterns of failure in pathologic stages I-IV clear-cell carcinoma of the endometrium: implications for adjuvant radiation therapy. Int J Radiat Oncol Biol Phys 2003; 55: 1272 – 6. 212. Clement PB, Scully RE. Endometrial hyperplasia and carcinoma. In Clement PB, Young RH (eds) Contemporary Issues in Surgical Pathology, Churchill Livingstone, Chap. 5, 181 – 264.
Section 7 : Gynecological Cancers
45
Tumors of the Cervix Krishnansu S. Tewari and Bradley J. Monk
HISTORICAL NOTES In January 1878, Freund of Germany “extirpated a cancerous uterus through an abdominal incision”, but limited the procedure to removal of the parametrial tissues only as far lateral as the ureteral tunnel, without removal of lymph nodes.1 John Goodrich Clark (1927), a resident under Howard Atwood Kelly at The Johns Hopkins Hospital, studied 20 autopsy cases of cervical cancer and on April 26, 1895, performed what is considered to be the first true radical hysterectomy in which he removed all of the parametria to the lateral pelvic wall. His paper included illustrations by the eminent artist Max Br¨odel and was published in the July–August 1895 issue of The Bulletin of the Johns Hopkins Hospital.2 Clark credited Freund with the earliest attempt at an abdominal approach to cervical cancer and noted that because of the high primary mortality associated with abdominal hysterectomy for cancer, most gynecologists during Freund’s era moved to vaginal hysterectomy. Because the Americans would eventually recognize the need for a formal lymphadenectomy, the vaginal radical hysterectomy, commonly called the Schauta operation after the Austrian surgeon Friedrich Shauta (1849–1919), never became popular in the United States.3 Ernst Wertheim (1864–1920), a student of Shauta’s, developed an abdominal version of Shauta’s radical vaginal operation, but during his early experience from 1900 to 1912 he had sepsis to contend with.4 Of those who survived the operation, many would go on to fail at the pelvic sidewall, where the lymph nodes reside. While studying the refraction of light from a cathode tube, Wilhelm Konrad von R¨oentgen (1845–1923) accidentally discovered X rays and went on to win the first Nobel Prize in Physics in 1901 for this achievement and the era of diagnostic radiology commenced.5 Next, Madame Marie Sklowdowska Curie (1867–1934), from a French family of scientists, discovered radium while working with an extract from pitchblende and was given the Nobel Prize in Physics in 1903.6 However, it was the Scottish American inventor, Alexander Graham Bell, who suggested the vaginal insertion of radium to treat a patient with cervical cancer, in 1908 nearly three decades after having demonstrated the first
telephone apparatus.7 Thus, radiotherapeutics found a role in the management of cervical cancer through the transvaginal insertion of long radium needles. Only after World War II did particle accelerators powerful enough to generate a therapeutic beam of radiation become available. Meanwhile, surgeons continued to refine their approach toward cervical tumors. Frederick Taussig (1891–1943) developed the technique of transperitoneal pelvic lymphadenectomy,8 which was later combined with the Wertheim operation by Joseph Vincent Meigs (1892–1963) of the Massachusetts General Hospital in Boston. From 1940 to 1951, Meigs performed 100 consecutive, personalized radical hysterectomies without a single perioperative mortality.9 In Japan, Shuichi Okabayashi (1884–1953) pioneered the Japanese radical abdominal hysterectomy,10 and later, Alexander Brunschwig (1901–1974) in the United States became the architect of pelvic exenterative surgery for centrally recurrent disease.11 On the preventative front, George Nicolas Papanicoloau (1883–1962), working first with the hamster cervix, invented the screening procedure that bears his name.12 The widespread implementation of the Papanicoloau test has resulted in a substantial decrease in the incidence of and mortality from cervical cancer in developed countries. Henrietta Lacks (1920–1951) died of cervical cancer at Johns Hopkins University. The cells of her cancer, known as HeLa cells, were the first human cells discovered to thrive and multiply outside the body, seemingly indefinitely, allowing investigators to create the polio vaccine, unravel the mechanistics of malignant transformation, and study molecular therapeutics.13 Finally, during the 1980s Henrich zur Hausen performed cross-hybridization studies and identified two new human papillomavirus (HPV) subtypes, designated HPV subtypes 16 and 18, both of which have been implicated separately (i.e., one or the other) in the pathogenesis of the majority of cervical cancers.14,15
UNCOMMON HISTOLOGIES Cervical cancer is a disease of epidemic proportions. It is the third most common cancer diagnosis in women
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
502
GYNECOLOGICAL CANCERS
worldwide, with 480 000 new cases diagnosed annually, but is probably underreported. In developed nations, there has been a dramatic decrease in the incidence and death rates because of screening. The median age at diagnosis for preinvasive disease of the cervix is 29 years, and 47 years for invasive carcinoma. This section focuses on many of the uncommon and scarce tumors of the cervix (see Table 1). Because of the rarity of many of these lesions, their real nature and clinical course have not been fully clarified. Importantly, some tumors have a highly aggressive course (e.g. adenoid cystic carcinoma (ACC) and small cell carcinoma), while others actually have a better prognosis when they arise in the cervix as opposed to other sites (e.g. embryonal rhabdomyosarcoma (E-RMS)). For many of these entities, arriving at the correct diagnosis may be a complex process, and immunohistochemical studies may be required. The differential diagnosis can become frustrating for the pathologist. In some cases (e.g. warty carcinoma), careful cytologic examination may hint at the diagnosis when histologic material is equivocal. Finally, some tumor types may be indicative of systemic disease (e.g. granulocytic sarcoma (GS) and lymphoepithelioma-like carcinoma (LELC)).
Squamous Cell Carcinomas Verrucous Carcinoma
Pipe smoking, chewing tobacco, dipping snuff, and poor oral hygiene and dentures have been cited as etiologic factors for verrucous carcinoma of the oral cavity. Ackerman described the first case in 1948. Preexisting condyloma acuminata, poor hygiene, and an infected prepuce has been linked with the development of penile verrucous carcinoma. The first case of verrucous carcinoma of the cervix was reported by Jennings and Barclay in 1972. An intranuclear viruslike particle measuring 45–50 nm in diameter has been identified in some
cases of female genital verrucous carcinomas, suggesting that perhaps the human papillomavirus may play a role. Among 29 patients with verrucous carcinoma of the cervix reported by Degefu et al., the mean age was 51 years at diagnosis (range, 30–84 years). Most often, patients presented with persistent vaginal discharge and bleeding.16 Grossly, verrucous carcinoma of the cervix appears as cauliflower-like, exophytic, and warty, with a grayish-white to brick-red color. Because cellular differentiation appears to proceed in a normal pattern throughout the thickened layer of the epidermis, dysplastic cells are not usually evident, and cytology does not alert the clinician to the lesion. Additionally, histopathologic misinterpretation was identified by Degefu et al. as a second factor leading to a delay in diagnosis and treatment, with 12 of the lesions in their report being originally assigned a diagnosis of benign squamous papilloma of the cervix resembling condyloma acuminata.16 Patients who were initially treated with podophyllin, cautery, or trachelectomy generally had a shortened survival following accurate diagnosis and further therapy. Histologically, the lesions manifest as well-differentiated squamous epithelium with benign-appearing cells and an intact basement membrane. Nuclear atypia may be present, and the parabasal and prickle cell nuclei tend to be enlarged and pleomorphic. The rete ridges undergo cystic dilation, forming epithelial fronds and contain flakes of keratin. Unlike condyloma acuminata, the fibrovascular core is usually absent and the epithelium consists of bulbous fronds compressing the surrounding tissue and underlying stroma. In addition, while the deeper portions of condyloma acuminata are papillary, the deeper portions of verrucous carcinoma form a solid nest of epithelial cells with a wide base. A biopsy should incorporate the base of the tumor and underlying stroma since it is in the depths of these lesions that distinguishing characteristics of an infiltrating growth pattern
Table 1 Histologic classification of cervical cancer.
Epithelial tumors Nonglandular
Glandular
Other, including mixed
Squamous cell carcinoma Verrucous carcinoma Warty (condylomatous) carcinoma Papillary squamotransitional carcinoma Lymphoepithelioma-like carcinoma Sarcomatoid carcinoma
Adenocarcinoma, usual endocervical type Mucinous adenocarcinoma Endometrioid adenocarcinoma Well-differentiated villoglandular adenocarcinoma Adenoma malignum (minimal deviation) Intestinal-like adenocarcinoma Signet-ring cell adenocarcinoma Colloid adenocarcinoma Clear cell adenocarcinoma Serous papillary adenocarcinoma Mesonephric adenocarcinoma
Adenosquamous Glassy cell carcinoma Mucoepidermoid carcinoma Adenoid cystic carcinoma Adenoid basal carcinoma Small cell carcinoma Classical carcinoid tumor Gestational choriocarcinoma
Mesenchymal tumors
Germ cell tumors
Miscellaneous
Carcinosarcoma Leiomyosarcoma Epithelioid leiomyosarcoma Extrauterine endometrial stromal sarcoma Adenosarcoma Embryonal rhabdomyosarcoma Granulocytic sarcoma (chloroma)
Mature teratoma Immature teratoma Yolk sac tumor Nongestational choriocarcinoma
Melanoma Lymphoma Primitive neuroectodermal tumor
TUMORS OF THE CERVIX
extending into the stroma are found. The importance of direct communication to the pathologist about the clinical appearance and extent of the lesion cannot be overemphasized. The traditional view has held verrucous carcinomas to have an indolent course with little invasive potential, but in the advanced stages or in the recurrent setting, this tumor may be aggressive. Primary radical excision is recommended for both primary and recurrent disease. Inadequate resection at the outset has resulted in rapid (i.e. within 1 year) central pelvic or vaginal recurrence. Although the incidence of nodal metastases is very low, pelvic lymphadenectomy may identify the occasional high-risk patient that may benefit from closer surveillance or some form of adjuvant therapy. Radiotherapy adds little to the treatment program, and among patients with verrucous carcinomas of the larynx and of the oral cavity, this treatment modality has been associated with prompt and explosive anaplastic transformation of the tumor in 30% of cases. Although several previous reviews of female genital verrucous carcinoma have considered the cervical cases and vulvar cases collectively, it is interesting that there is a distinct difference in the uncorrected 2-year survival for verrucous carcinoma of the cervix (40%) as compared with that for vulvar lesions (75%). In both diseases, histologically confirmed pulmonary metastases have been observed. Warty (Condylomatous) Carcinoma
This tumor is characterized by marked condylomatous changes. It is a recently described variant of squamous cell carcinoma (SCCA) of the cervix, and is histologically identical to warty carcinoma of the vulva. Unlike its homolog in the vulva, these cervical cancers exhibit evidence of SCCA at the deep margin. Multiple subtypes of HPV (including low-risk strains) can be detected in the majority of tumors. Ng et al. have performed thin-layer cytologic preparations on three cases and noted small, cohesive clusters and syncytial sheets of tumor cells, with necrotic tumor debris (diathesis) in the background.17 The tumor cells were polygonal to elongated and contained oval nuclei, coarse chromatin, and sometimes distinct nuclei. Dyskeratotic tumor cells with bizarre shapes were also noted. These three lesions were also characterized by having many koilocytic cells possessed of pleomorphic nuclei, distinct nucleoli, and perinuclear cytoplasmic halos. Because koilocytes are rarely found in cervical cytology specimens of conventional SCCA, the authors emphasize the importance that a correct cytologic diagnosis is possible if one pays attention to extreme koilocytotic atypia when present. Although experience has been slow to accumulate, some authors have noted a less aggressive clinical behavior when these tumors are compared to the more common well-differentiated SCCAs of the cervix. Papillary Squamotransitional Carcinoma
Papillary squamous carcinomas were initially described by Randall et al. in 1986 who noted a histologic resemblance to transitional cell carcinoma of the bladder. The tumors are composed of papillary projections that are covered by several layers of atypical epithelial cells. Because they demonstrate a spectrum of histologic appearances, these tumors have
503
been referred to as papillary squamotransitional carcinomas (PSTC). Unlike transitional cell tumors of the urinary tract, PSTCs are typically positive for cytokeratin 7 and negative for cytokeratin 20. Koenig et al. have subdivided the tumors into three histologic groups: predominantly squamous, predominantly transitional, and mixed squamous and transitional. In cases that appear more transitional, the cells are oval with their long axis oriented perpendicular to the surface with minimal flattening.18 An inverted endophytic pattern similar to that of transitional cell carcinoma of the urothelium can also be present. In cases with more squamous differentiation, the cells are more basaloid and resemble those of a high-grade squamous intraepithelial lesion devoid of frank keratinization and koilocytic change. Typical invasive SCCA can often be identified at the base of the tumor, appearing as well-circumscribed nests of epithelium in continuity with papillae that extend deeply into the stroma. Focal invasion of the papillae themselves has also been reported. Microscopically, the tumor can be misinterpreted as a cervical intraepithelial neoplasia grade 3 with a papillary configuration or as a squamous papilloma. The tumor must also be differentiated from other rare variants of SCCA, including verrucous and condylomatous carcinoma. PSTCs are potentially aggressive, with many lesions presenting at a more advanced stage than would be expected based on a histologic appearance reminiscent of a superficial or early invasive lesion. In the report by Koenig et al. containing 32 cases, the mean age at diagnosis was 50 years (range, 22–93 years).18 Three of the 12 patients with followup information died of disease at a mean of 13 months following diagnosis. In a recent case series by Ortega-Gonzalez et al., all six of their patients presented at an advanced clinical stage, two had recurrences and one metastasized.19 Lymphoepithelioma-like Carcinoma LELC of the cervix is a distinct variant of SCCA, in which the tumor cells are well circumscribed and composed of undifferentiated cells surrounded by a marked stromal inflammatory infiltrate. The tumor may confer a more favorable prognosis than conventional SCCA. Typically, it is found in a population of younger women and is more prevalent in non-Caucasian populations, especially those of Asian descent. LELC of the cervix also lacks a clearly defined association with HPV infection.20 The Epstein-Barr virus (EBV) has been postulated to play an etiologic role in LELC in diverse anatomic locations. Tseng et al. studied 15 cervical LELC tumors and 15 SCCA controls and detected EBV genomes more frequently in LELC than in conventional SCCA (73.3 vs 26.7%, p = 0.01).21 The detection rate of HPV-16 and HPV-18 DNA was significantly lower in patients with LELC tumors than in patients with conventional SCCA (20 vs 80%, p = 0.001). After a median follow-up of 3.9 years (range, 1.8–5.3 years), the 15 patients with cervical LELC were disease free following radical hysterectomy or radiotherapy. Sarcomatoid Carcinoma Sarcomatoid carcinomas are uncommon variants of epithelial carcinoma. The diagnosis is based on histologic, immunohistochemical, and ultrastructural features. A recognizable
504
GYNECOLOGICAL CANCERS
SCCA typically merges with a spindle cell component. The spindle cell component characterizes this tumor by its prominence, and bizarre, multinucleated tumor giant cells may also be present. The lesion is distinguished from a sarcoma by immunostaining, through which both the spindled component and the multinucleated tumor cells are positive for the epithelial marker, cytokeratin. Thus, unlike true sarcomas or malignant mixed mesodermal tumors, sarcomatoid carcinomas lack a malignant stromal or mesenchymal component. Sarcomatoid carcinomas have been identified in the oral cavity, pharynx, esophagus, and larynx. There have been only 16 reported cases of sarcomatoid SCCA of the cervix. Most patients have presented with abnormal vaginal bleeding, and all have had a visible cervical lesion amenable to biopsy. Of 10 patients for whom detailed clinical information was available, six were diagnosed with locally advanced disease (International Federation of Gynecology and Obstetrics (FIGO) stages IB2 –IVA), and two presented with metastatic disease. Eight of nine patients reported by Brown et al. were treated with primary or adjuvant pelvic irradiation, and five suffered a recurrence with a median disease-free interval of 4.9 months (range, 2–9.5 months).22 Only three patients in the literature have remained disease free following primary therapy for 22, 40, and 42 months. Although the data are limited, it appears that the disease is aggressive as evidenced by an advanced stage at presentation and early recurrence following therapy. Pelvic irradiation can salvage approximately 25% of patients.
Glandular Tumors Adenocarcinoma, Usual Endocervical Type
The relative proportions and absolute incidences of SCCAs and adenocarcinomas (ACs) of the cervix have been changing in industrialized nations during the preceding 40 years since the implementation of cytologic screening. Endocervical AC and its variants now account for approximately 20–25% of cervical cancers.23 The percentage of women younger than 35 years with AC has increased from 16% in 1964 to 25% in the 1990s. Historically, cervical AC has often been referred to as mucinous AC. Although a subset of endocervical AC are overtly mucinous (see following section), the usual endocervical type of AC most often shows no mucin or no greater amount than can be seen in nonmucinous tumors. The cell of origin is likely a pluripotential subcolumnar reserve cell of the columnar endocervical epithelium, with adenocarcinoma in situ (AIS) regarded as an immediate precursor to invasive endocervical AC. Recent studies that have controlled for HPV and sexual behavior have not found oral contraceptive usage to be a significant risk factor in this disease. The differential diagnosis of the usual endocervical type of AC includes other forms of primary and secondary AC, AIS, and benign lesions such as normal endocervical glands and Nabothian cysts that extend deeply into the cervical wall. Early invasion may be appreciated when irregularly arranged glands, small clusters of cells, or single cells are clearly seen to infiltrate from a parent gland showing in situ AC, especially when there is an associated
stromal response and inflammation. Nevertheless, because of the difficulty in interpreting the irregular distribution and complex architecture of the normal endocervical crypts in the cervical stroma, microinvasive AC (i.e. stage IA) as a histologically recognizable entity has been the subject of much debate. Because of the varied morphology of these tumors, the glandular lesions account for many problematic issues in cervical pathology. The ACs of usual endocervical type account for 80% of cervical ACs, and are characterized by moderate differentiation and glands of medium size. The cells have eosinophilic cytoplasm and brisk mitotic activity with frequent apoptotic bodies. Abnormal vaginal bleeding occurs in 75% of patients. Fifty percent of cases are characterized by a fungating, polypoid, or papillary mass. In 15% of patients, the cervix is diffusely enlarged or nodular, and in 15% no gross lesion is visible. Eighty percent of patients present with stage I or stage II disease. Deep invasion is common, even with early stage tumors, because the carcinoma can arise deep within the endocervical canal. Mucinous Adenocarcinoma
These tumors are characterized by easily identifiable mucinrich cells that predominate over any other cell type that may be present. The tumors may be graded on the basis of the proportion of solid and glandular areas (akin to what is done for endometrial cancers) or on the basis of nuclear grade. The majority of glands are frankly malignant on a combined architectural and cytologic analysis. The differentiation of a primary cervical mucinous or endometrioid AC from that of an endometrioid AC of the endometrium with cervical extension can be problematic. In some cases, an endometrial lesion will be demonstrable on pelvic ultrasonography, while a primary cervical tumor may result in cervical expansion in the absence of uterine enlargement. Immunostaining can be used as a diagnostic adjunct, with carcinoembryonic antigen (CEA) and mucus antigen positivity being observed in 59–80% and 56% of endocervical tumors, respectively, as compared to 8–50% and 0% of endometrial carcinomas, respectively. In contrast, 0% of endocervical lesions will stain for vimentin, while 66% of endometrial cancers will be positive. In 1990, Konishi et al. investigated the clinical significance of mucin leakage into the cervical stroma in 35 cases of cervical AC.24 Histologic evidence of mucin leakage was identified in 14 cases (40%) as amorphorous materials dissecting the connective tissues and permeating the lymphatic channels, associated with and without tumor cells. The cases with mucin leakage showed a significantly higher incidence of lymph node involvement than those without mucin leakage (71.4 vs 23.8%, p < 0.01). In addition, when the mucin leakage was immunohistochemically positive for CEA or CA 19-9, elevated serum levels of these antigens were frequently observed. The investigators suggest that mucin leakage into the cervical stroma is indicative not only of stromal invasion but also allows for the conduct of tumor cells into the lymphatic channels.
TUMORS OF THE CERVIX
Endometrioid Adenocarcinoma
True endometrioid carcinomas of the cervix are rare, with the characteristic histology of tubular glands, sometimes with villous papillae, and ciliated cells. The cells tend to be stratified and have oval nuclei that are arranged with their long axis perpendicular to the basement membrane of the gland. The cells do not contain mucin and have less cytoplasm than do the cells of a mucinous AC. Interestingly, endometrioid ACs frequently contain small foci of squamous epithelium. There have been scattered cases of endometrioid endocervical AC arising from cervical endometriosis. In 1997, Fujiwara et al. investigated the expression and clinical significance of estrogen receptor (ER) and progesterone receptor (PR) in 84 cervical ACs.25 ER was identified in 17 (20%) and PR was identified in 23 cases (27%). ER positivity was most frequently detected in mucinous AC of the endocervical type (n = 11 of 48 cases) and endometrioid AC (n = 4 of 10 cases). PR positivity was also most frequently detected in mucinous AC of the endocervical type (n = 15 of 48 cases) and endometrioid AC (n = 6 of 10 cases). Mucinous AC of the intestinal type (n = 5), glassy cell carcinoma (GCC) (n = 2), and clear cell AC (n = 2) were uniformly negative for both ER and PR. No association was detected between FIGO stage and receptor status, but there was a somewhat lower frequency of ER positivity in poorly differentiated tumors (p = 0.07). Receptor status was not significantly associated with either overall survival or disease-free survival. Comment on the Clinical Profile of the Adenocarcinomas It is a complex task to extract clinical information from the literature specific for each of the three tumor types discussed above because in many reports concerning endocervical ACs, investigators have not made distinctions between the usual type, the mucinous AC, and the endometrioid AC, oftentimes lumping them together. The problem is compounded by the tendency of many pathologists to regard mucinous AC as the most common form of endocervical AC and relegate those tumors that demonstrate minimal-to-no mucin staining (i.e. the usual type) to either a subset of mucinous AC or to incorrectly characterize them as endometrioid ACs. This last detail has resulted in a relative increase in the proportion of endometrioid ACs that have been reported in some series, with several authorities suggesting that endometrioid ACs account for up to 30% of endocervical ACs. True endometrioid ACs are exceedingly rare. Therefore, in the passage that follows, the term “ACs” will be understood to represent the usual type, and the mucinous and the endometrioid subtypes of endocervical AC. The management of ACs follows that which has been outlined for SCCA. In 1991, Hopkins and Morley performed a survival analysis of 203 patients (21%) with adenocarcinoma and 756 (79%) with SCCA treated from 1970 to 1985.26 Survival by stage was significantly influenced by the cell type, with 90% 5-year survival observed for patients with SCCA as compared to 60% 5-year survival among patients with adenocarcinoma (p < 0.001). Patients with stage II SCCA had a 62% survival, compared with 47% for adenocarcinoma (p = 0.01); patients with stage III squamous cell disease
505
had a 36% survival, compared with 8% for adenocarcinoma (p = 0.002). Other features that influenced survival included node status (p = 0.001), poor differentiation of tumor histology (p = 0.001), diabetes (p = 0.001), and Pap smear interval (p = 0.001). Other studies, including population-based reports, have failed to confirm that prognosis is affected by histologic type. In 2002, Lea et al. identified 83 patients with stage IIB –IVB tumors, and during a median follow-up period of 33 months, 66 (80%) died of disease.27 Stage IIB disease, young patient age, and grade 1 histology were independent variables having a favorable impact on survival. Villoglandular Papillary Adenocarcinoma
Although approximately 10–15% of cervical adenocarcinomas have a papillary pattern, villoglandular papillary adenocarcinoma (VGPA) constitutes only a small proportion of these tumors. Young and Scully first established VGPA of the cervix as a distinct histologic entity and a subtype of welldifferentiated adenocarcinoma in 1989.28 It has been further subclassified into endocervical, endometrioid, and intestinal types. Approximately 85 cases have been reported in the English medical literature. Compared with the common types of cervical adenocarcinomas, VGPA is morphologically characterized by extremely villous and papillary growth. Specifically, the cardinal feature of the VGPA is a surface papillary component of variable thickness with papillae that are usually tall and thin with a fibrous, stromal core (see Figure 1). Small cellular budding may be present but to a lesser extent than what is seen in serous papillary carcinomas. The invasive portion of the tumor has a pushing border and is composed of elongated branching glands separated by a fibrous stroma; the stroma may be desmoplastic or myxoid at the advancing margin of the tumor. The papillae and glands are lined by stratified nonmucinous columnar cells, and only mild to moderate nuclear atypicality and scattered mitotic figures (MFs) are present in the tumor cells. The adjacent cervical glandular epithelium often contains an AIS.
Figure 1 Well-differentiated villoglandular adenocarcinoma of the cervix.
506
GYNECOLOGICAL CANCERS
Recently, Utsugi et al. published a series of 13 patients. The median age was 45 years (range, 36–64 years).29 Interestingly, the Pap test was positive for adenocarcinoma in all patients, with tall and thin papillae present in 10 patients allowing the investigators to predict VGPA cytologically. In contrast to earlier reports, these investigators encountered high mitotic counts, lymphovascular space invasion, and lymph node metastases in several of their patients. As of December 2002, all patients were alive and without recurrence during 3 to 19 years of follow-up. Indeed, the prognosis has been reported to be more favorable in patients with VGPA when compared with the common types of cervical adenocarcinomas, with only one of 85 patients to have appeared in the English literature to have developed recurrence. This patient originally underwent radical hysterectomy with lymphadenectomy for a stage IIB lesion, and received adjuvant pelvic radiotherapy due to lymphatic metastases; she developed a vaginal recurrence at 30 months, and died of disease at 46 months. Four other patients with lymph node metastases who received postsurgical pelvic irradiation have remained without evidence of disease for 10–220 months. Young and Clement caution that a tumor should not be placed in this group if any ominous feature is present, as that may lead to undertreatment of a potentially fatal lesion.30 It must be emphasized that adenocarcinomas of the usual type, including deeply invasive tumors, may have a papillary component, and the diagnosis of VGPA should be reserved for lesions that are exclusively grade 1 and unassociated with an underlying component of conventional adenocarcinoma. The diagnosis of VGPA should only be suggested upon small biopsy material, with a requirement of a cone biopsy or hysterectomy specimen before the term can be used definitively. Making the diagnosis with limited clinical material may result in overly conservative treatment of a lesion with a prognosis worse than that of VGPA. Adenoma Malignum (Minimal Deviation Adenocarcinoma)
Adenoma malignum of the uterine cervix was first described by Gusserow in 1987. The term is often considered to be synonymous with “minimal deviation adenocarcinoma,” which was proposed by Silverberg and Hurt in 1975.31 The tumor accounts for only 1–3% of adenocarcinomas of the cervix. It is considered a variant of mucinous adenocarcinoma. In the case series by Hirai et al., the mean age was 53.3 years (range, 38–70 years), and presenting symptoms often include both watery or mucoid discharge and atypical vaginal bleeding.32 Interestingly, several cases of Peutz-Jeghers syndrome have been complicated by adenoma malignum. It is difficult to correctly diagnose adenoma malignum because the cytological deviation from the normal glandular cells is considered to be too small, and the histological deviation is minimal (Figure 2). Hirai et al. have identified four common characteristic cytologic features in a review of Papanicoloau smears associated with this lesion. (i) Large sheets of tumor cells are apparent. The tumor cells have abundant mucin and are arranged in palisades at the circumferences of the sheets (ii) In the area with a palisading
Figure 2 Minimal deviation adenocarcinoma of the cervix (adenoma malignum).
arrangement, nuclei are constantly positioned in the cytoplasm and overlap side by side (iii) Nuclei are tensile and some are irregular in shape. The nuclear chromatin is fine granular, and reveals frequent nuclear clearing, which is suggestive of increased euchromatins (iv) The ordinary form of adenocarcinoma cells in clusters is occasionally present.32 Since most of the tumor glands are highly differentiated, Hirai et al. emphasize that it is exceedingly difficult to histologically differentiate these tumor glands from normal endocervical glands, particularly in specimens taken from cervical punch biopsies.32 Thus, since the ability to make a preoperative diagnosis of adenoma malignum is limited, the disease is often undertreated initially, and the final diagnosis assigned only after a comprehensive review of the surgical specimens. Therefore, when the preoperative punch biopsy leads to the diagnosis of typical endocervical adenocarcinoma, it is expected that the disease will be treated correctly. However, when the punch biopsy reveals no sign of malignancy, the possibility of having the disease undertreated exists. Because cytological examination is a potent aid to detect this tumor, for those cases with negative punch biopsy having cytologic features that suggest the presence of adenoma malignum, it is important to obtain a sample of the deeply positioned tumor glands preoperatively through a deep biopsy or cervical conization. The combination of clinical understaging and undertreatment has resulted in the majority of the early reports suggesting an unfavorable prognosis for this disease. Recent studies suggest that the survival of adenoma malignum is consistent with that of other forms of well-differentiated adenocarcinoma of the cervix. Intestinal-type, Signet-ring Cell, and Colloid Adenocarcinomas
The intestinal type of mucinous adenocarcinoma is composed of cells similar to those present in adenocarcinomas of the large intestine. Goblet cells are common, and are occasionally admixed with argentaffin cells and Paneth cells. There may also be pseudostratification, and only small amounts of
TUMORS OF THE CERVIX
intracellular mucin present.30 Signet-ring cells may also be found within cervical adenocarcinomas, and also in some cervical adenosquamous carcinomas (ASC). Pure signet-ring cell adenocarcinomas and colloid adenocarcinomas are both extraordinarily rare. Clear Cell Adenocarcinoma
Clear cell carcinomas (CCAs) of the cervix account for 4–9% of adenocarcinomas of the cervix. They are cytologically, histologically, and ultrastructurally identical to their counterparts at other sites of the female genital tract such as the vagina, endometrium, and ovary. It has been suggested that this tumor arises from pluripotent reserve cells of the cervix that through faulty differentiation remain at an intermediate stage of development between keratinization and mucin secretion. In addition to in utero exposure to diethylstilbestrol (DES), genetic factors, microsatellite instability, HPV infection, Bcl-2 protein overexpression, p53 mutation, and other exogenous risk factors are likely to play an etiologic role in this disease. Importantly, these tumors can develop in the absence of DES exposure. A bimodal age distribution among patients without DES exposure has been detected, with one peak at 26 years, and a second peak at 71 years. CCAs are predominantly endophytic and tend toward deep infiltration with creation of a barrel-shaped cervix. They are more likely to extend to the lower uterine segment and uterine endometrium more often than SCCAs and nonCCA adenocarcinomas. When this occurs, the tumor may be distinguished from a primary endometrial CCA with cervical extension on the basis of the pattern of invasion. Histologically, the neoplastic cells have abundant clear or vacuolated cytoplasm that can contain glycogen. The growth pattern can be classified as solid, tubulocystic, or papillary, with flat, cuboidal, and hobnail cells being prominent in the tubular and tubulocystic patterns. Several commentators have speculated that CCAs of the cervix are associated with a poorer prognosis than SCCAs and non-CCA adenocarcinoma. In 1979, Herbst et al. collected 145 cervical CCAs in women with and without in utero exposure to DES and with a median follow-up of 4 years, reported 5-year survival rates of 91% (stage I), 77% (stage IIA), and 60% (stage IIB).33 In a contemporary series of 15 patients reported by Reich et al.,34 the 5-year survival rate of 67% in patients with stages IB –IIB disease was consistent with the results reported by Herbst et al.,33 and although the CCAs tended to have a slightly worse 5-year survival rate than SCCAs and other cervical adenocarcinomas, this observation was not statistically significant. Serous Papillary Adenocarcinoma
Serous papillary carcinoma of the cervix (SPCC) is a recently described variant of cervical adenocarcinoma that is morphologically similar to serous papillary adenocarcinoma of the ovary, endometrium, fallopian tube, and peritoneum. Rose and Reale provided the first detailed description of this tumor in 1993.35 Among the first several reports,
507
angiolymphatic space invasion, lymph node metastases, or extracervical spread or both were common. In point of fact, in a review of 67 cases of cervical adenocarcinoma and ASC, Costa et al. identified the presence of serous differentiation in the glandular component to be associated with an increased risk of recurrence.36 Similar to what has been observed among women with clear cell adenocarcinoma of the cervix not associated with in utero DES exposure, some studies on SPCC of the cervix have detected a bimodal age distribution with one peak occurring before 40 years, and the second peak after the age of 65. This represents a distinguishing clinical feature of SPCC when it is compared to the more common variants of cervical adenocarcinoma. The differential diagnosis of clinical stage I SPCC includes cervical involvement of a uterine papillary serous carcinoma (UPSC) of the endometrium, as well as well-differentiated villoglandular adenocarcinoma and papillary clear cell adenocarcinoma arising in the cervix. A stage II UPSC may be excluded by a fractional dilatation and curettage or examination of a hysterectomy specimen. The distinction between SPCC and well-differentiated villoglandular adenocarcinoma (see preceding section) is critical because the latter almost always has an excellent prognosis and some cases have been successfully treated with cone biopsy. Although papillary clear cell adenocarcinoma and SPCC share the histologic features of papillae covered by round to oval cells with round nuclei and prominent nucleoli, the former typically has a hyalinized papillary core and a majority of epithelial cells with clear cytoplasm or a hobnail appearance, as well as the tubular, cystic, and solid patterns that are uncommon or absent in SPCCs. Immunohistochemistry studies have demonstrated frequent CA-125 positivity among SPCC. Interestingly, however, in noncervical sites, serous papillary carcinomas are typically CEA negative, but SPCC is frequently CEA positive (although not as consistently as mucinous adenocarcinoma of the cervix). The serum CA-125 has also been reported to be increased in some cases of SPCC, with some patients presenting with carcinomatosis, ascites, or subdiaphragmatic metastases. Zhou et al. have published a series of 17 cases in which 11 patients presented with abnormal vaginal bleeding, four with abnormal exfoliative cervical cytology, and two with a watery vaginal discharge.37 Eight of the women had a polypoid or exophytic cervical mass, two had an ulcerated or indurated cervix, and no gross abnormality was detected in seven patients. In their report, the stage distribution included stage IA (n = 1), stage IB (n = 12), stage II (n = 2), and stage III (n = 1). On low-power microscopic examination, all of the tumors had a complex papillary pattern with epithelial stratification and tuffing with the formation of cellular buds. The papillary nature of the tumors was most striking in the exophytic portions of the tumors. In nonpapillary areas, the tumors exhibited a predominantly glandular growth pattern, with elongated and slitlike spaces. In the invasive areas, nests of tumor cells infiltrated the cervical stroma in an irregular pattern, with clefts around the tumor cell nests. An intense acute and chronic inflammatory infiltrate was typically present within the core of the papillae
508
GYNECOLOGICAL CANCERS
and in areas of stromal invasion. All of the tumors had >10 mitotic figures per 10 high-power fields (MFs/10 hpfs), with most showing >30 MFs/hpfs. Only three tumors had psammoma bodies. In the study by Zhou et al., all three patients with stage II and III tumors experienced metastases.37 Three of the 12 patients with stage IB lesions received primary radiotherapy with (n = 1) and without (n = 2) chemotherapy, and all died of disease. Among the eight patients with stage IB tumors who survived without disease, six had undergone radical hysterectomy (with adjuvant radiotherapy given to three), and one was treated with simple hysterectomy and adjuvant radiation. Although the numbers are very small, the authors suggest that primary surgical therapy may be preferable to primary radiotherapy in early-stage SPCC. Although the tumors can behave aggressively and be associated with a rapidly fatal course, the outcome for patients with stage I disease is similar to that of patients with cervical adenocarcinoma of the usual type. The impact of concurrent radiosensitizing chemotherapy for locally advanced disease and the efficacy of platinum-based combination chemotherapy for metastatic disease has not been studied. Mesonephric Carcinoma
Pathologists have been fascinated by the embryology and pathology of the paired mesonephric ducts for decades. The distal portion of each duct courses through the parametrial tissues and enters the lateral wall of the cervix at the isthmus where it dilates to form an ampulla from which numerous tubules ramify. It then extends to the lateral walls of the vagina where it is referred to as Gartner’s duct. Remnants of the mesonephric duct and its tubules can be found in up to 22% of adult uterine cervices. In 1990, Ferry and Scully identified several cases of lobular mesonephric hyperplasia, diffuse mesonephric hyperplasia, mesonephric ductal hyperplasia, and mesonephric carcinoma from a batch of 49 cervixes containing mesonephric remnants.38 The mean age of patients with mesonephric carcinoma is 52–55 years (range, 34–72 years), and most present with abnormal bleeding, often with a visible cervical lesion. Mesonephric carcinomas typically exhibit morphologic diversity, with considerable variability within a tumor, including ductal, tubular, retiform, solid, and sex cordlike patterns. One pattern usually predominates, with the ductal and tubular pattern being most commonly encountered. In the tubular pattern, the tumor consists of large sheets of closely packed small round tubules, often with dense intraluminal eosinophilic secretions that resemble the malignant counterpart of mesonephric remnants (Figure 3). The ductal pattern consists of larger glands, often with intraglandular villous papillae that may simulate endometrioid carcinoma. Many tumors are found adjacent to hyperplastic mesonephric remnants. Most are pure adenocarcinomas, although half of the tumors in one series were biphasic with a sarcomatoid component (the so-called “malignant mesonephric mixed tumor”). The spindle cell component generally resembles endometrial stromal sarcoma (ESS) or a nonspecific spindle cell sarcoma.
Figure 3 Mesonephric adenocarcinoma of the cervix.
In a comprehensive review, Hart noted that while most reported cases had been confined to the cervix at diagnosis and exhibited a more indolent behavior than their mullerian counterparts, a few had been accompanied by extrauterine spread, including lymph node metastases.39 Among those tumors with a malignant spindle cell component, several had metastasized. There was an apparent tendency for late recurrences, and a few higher stage tumors had taken an aggressive course. In 2001, Silver et al. presented follow-up data on 10 cases and found six of eight patients with stage IB tumors to be alive without disease after a mean of 4.8 years.40 Whenever possible, radical surgery with lymphadenectomy appears to be the preferred treatment modality.
Other Epithelial Tumors Adenosquamous Carcinomas
ASCs account for 5–25% of all cervical cancers, occurring in both young and old women, and occasionally associated with pregnancy. The epidemiologic risk factors are more similar to those with SCCA than of adenocarcinoma. The diagnosis is confined to those tumors that contain malignant glandular and squamous elements that are recognizable without the use of special stains. The squamous component is well differentiated and can contain keratin pearls and may be glycogenated. The glandular component is commonly the usual endocervical type, but may also be mucinous, including signet-ring cell, or mixed endocervical and mucinous, endometrioid, or clear cell. As discussed earlier, several studies have suggested that cervical ACs have a poorer prognosis than SCCA, while other investigations have found no difference. Most comparison studies have not separated ACs from those tumors with ASC histology. In a prospective study by the Gynecologic Oncology Group (GOG) examining stage IB cervical cancer, Look et al. observed that patients with ASC histology had a worse prognosis than patients with SCCA or other AC.41 Farley et al. compared the survival rates between patients with endocervical AC (n = 185) and ASC (n = 88).42
TUMORS OF THE CERVIX
Although 66% of ASC tumors were poorly differentiated, there was no difference in the incidence of positive lymph nodes associated with the two histologic types. Patients with ASC had a significantly decreased 5-year survival rate compared with those with AC (65 vs 83%, p < 0.002); however, this decrease in survival was observed only in patients with stage II –IV disease (p = 0.01). The 5-year survival rate for patients with grade 1 AC was 93%, compared with 50% for patients with grade 1 ASC (p < 0.01). The investigators concluded that ASC histology is an independent predictor of poor outcome among women with cervical cancer. Using the surgicopathologic risk factors employed in GOG protocols 92 and 109, Lea et al. studied 230 patients with stage IB1 endocervical AC, for whom the overall 5-year survival rate was 89%.43 Among those patients with low-risk tumors (n = 178), ASC histology was the only independent risk factor of disease recurrence (p < 0.01), with a 5-year disease-free survival of 79% compared to that of 96% for other histologic subtypes (p < 0.01). Glassy Cell Carcinoma
GCC accounts for only 1–5% of all cervical cancers. It was first described in 1956 by Glucksman and Cherry who considered the disease to be very aggressive because there were no survivors among the 41 cases they reported. Histologically, the tumor cells have moderate ground glass cytoplasm, large nuclei, prominent nucleoli with background eosinophilia, and distinct cell membranes that stain positively with periodic acid-Schiff (PAS). In one series, the tumor was diagnosed among patients with a mean age of 44 (range, 12–69 years). Some authors have associated this tumor type with pregnancy. In 1976, Littman et al. cited a survival rate of 31% among a series of 13 patients with GCC. In 1988 Tamimi et al. reported on 29 patients with stage I disease for whom overall survival was only 55%. Gray et al. have collected approximately 103 cases from the literature and calculated the overall survival for all patients to be approximately 50%; for patients with stage I tumors, overall survival is still diminished at 64%.44 These investigators provided details on an additional 22 patients and demonstrated improved prognosis with a disease-free survival of 64% and an overall survival of 73%. Hopkins and Morley have conducted an extensive review of the literature and have added 21 of their own cases to the analysis.45 If the original report by Glucksman and Cherry is excluded, there are approximately 100 patients available for analysis. Thus, Hopkins and Morley emphasize that conclusions made for this cell type’s influence on survival must be drawn with caution. If the original report is excluded, the overall survival is 47% (48 of 107); if the original report is included, the overall survival is decreased to 33% (48 of 148). Survival for this tumor is influenced markedly by the stage of the disease. In the original report by Glucksman and Cherry, the stage of disease was not available. Thus, for those cases in which clinical stage has been reported, the overall survival for stage I disease is approximately 60%. Of particular interest was the observation that in the contemporary report by Gray et al., the overall survival for the
509
14 patients with stage I disease was 86%, comparable to what is reported in the literature for stage I SCCA. Importantly, Gray et al. found that intermediate risk histopathologic features known to predict a higher rate of relapse after radical surgery in SCCA patients (lymphovascular space involvement (LVSI), deep stromal invasion, and large tumor diameter) appear to be predictive of relapse in GCC as well.44 Because the tumor has a propensity for pelvic and vaginal failure, Gray et al. advocate adjuvant pelvic irradiation in patients with these tumor-related poor prognostic findings.44 Mucoepidermoid Carcinoma
Mucoepidermoid carcinoma (MEC) is defined as a tumor with the appearance of SCCA but lacks a recognizable glandular pattern and demonstrates intracellular mucin. The squamous component is usually large cell nonkeratinizing or focally keratinizing, and the mucin-producing cells are frequently localized in the center of nests of SCCA. These tumors can account for up to 36% of cervical carcinomas in some series. The mucinous component includes goblet or signet-ring type cells. The mucin may extrude into the intercellular spaces where it may collect in lakes. The mucin is best demonstrated by alcian blue and PAS diastase. In some series, MEC has been found more commonly among younger patients. Thelmo et al. noted that while the overall rate of lymph node metastases was 14% among 265 cases of stage IB SCCA of the cervix, those cases that were classified as MEC had a 33% incidence of nodal metastases.46 Because MEC may be more aggressive than conventional SCCA, several authorities have advised staining all cervical SCCAs if they demonstrate finely vacuolated cytoplasm and lack peripheral palisading. The detection of CEA (by immunostaining or serum measurement) may also be of clinical utility in establishing the diagnosis. Adenoid Cystic Carcinoma
ACCs may arise in diverse anatomic sites, including the salivary glands, tracheobronchial tree, and breast, and were first described by Billroth in 1859. In the female genital tract, ACC occurs in the Bartholin’s gland, the endometrium, and the uterine cervix. ACCs represent less than 1% of cervical adenocarcinomas. The first example of this tumor was reported by Paalman and Counsellor in 1949 as “cylindroma” of the cervix because of the highly characteristic cytoarchitectural features. More than 150 cases have been reported since McGee et al. introduced the currently accepted designation of “adenoid cystic” carcinoma. The only lesion in the differential is the adenoid basal carcinoma (ABC). Indeed, as has been observed with ABC, the ACC has a predilection for the postmenopausal, black patient, with a mean age of 63 (range, 31–99 years) in one literature review containing 59 cases. Grayson et al. have demonstrated that there are no significant differences in mucin staining between ACC and ABC, and, with the exception of type IV collagen and laminin staining found exclusively in ACC, the immunohistochemical profiles are similar.47 Both tumors can be S-100 positive, and both are associated with integration of high-risk HPV DNA.
510
GYNECOLOGICAL CANCERS
Young and Clement have observed that unlike ABC, patients with ACC usually have an obvious exophytic or endophytic cervical mass that can vary greatly in size.30 Microscopic examination shows nests of cells often with a focal cribriform pattern resembling that seen in ACC of the salivary glands, as well as sheets, trabeculae, and cords. The glandular lumens may contain hyaline or mucinous material, and there is usually focal palisading of cells at the periphery of tumor nests. The neoplastic cells are larger than those of ABC, and more pleomorphic nuclei are present. The mitotic rate is high, and necrosis can be extensive. Unlike ABC, there is a stromal response that may be myxoid, fibroblastic, or hyaline. Whereas ABC has an excellent prognosis, that of ACC is poor. Survival for all stages is approximately 32.5%. In a review of 43 cases, Prempree et al. noted a 56.2% 3to 5-year survival for stage I disease, regardless of treatment modality.48 The optimal primary treatment for early stage lesions is unknown. Because of the high incidence of lymphatic metastases, vascular space involvement, and distant metastases, most treatment programs include adjuvant cisplatin-based chemotherapy. King et al. reported a survivorship of 12 and 64 months for two patients with stage IB disease who were treated with primary radical hysterectomy with lymphadenectomy, and adjuvant cisplatin at 100 mg m−2 .49 In both patients, vascular space permeation was extensive and the long-term survivor had a positive obturator node and bilateral parametrial extension. Adenoid Basal Carcinoma
Baggish and Woodruff introduced the concept of the ABC of the cervix in 1966. The lesion accounts for less than 1% of cervical adenocarcinomas, with approximately 50 examples appearing in the literature. The tumor has a striking propensity to occur in postmenopausal black women, and the integration of high-risk HPV DNA, particularly HPV-16, has recently been implicated in its pathogenesis. The rarity of ABC has been attributed to an unidentified cofactor. ABC shares some characteristics with ACC with which it often has been confused. An accurate distinction between the two is best made morphologically. Numerous widely separated, small, round to oval nests of variable size composed of uniform, cytologically bland basaloid cells with peripheral palisading is often present in ABC (Figure 4). Some of the neoplastic islands can be closely packed, exhibiting a minor degree of lobulation. Stromal reaction is absent. Divergent epithelial differentiation can occur in both ABC and ACC, and the occurrence of ABC areas in some ACCs and vice versa suggest that these tumors may share a common histogenesis from pluripotent reserve cells. Circumstantial evidence even suggests that ABC may be a precursor of cervical ACC. Both ABC-like and ACC-like areas have been reported in or adjacent to some malignant mullerian mixed tumors (MMMT) of the cervix. Brainard et al. reported 12 cases of ABC in 1998.50 The mean age was 71 years (range, 30–91 years), and all were asymptomatic. Almost all presented with an abnormal Pap smear that usually showed a squamous epithelial abnormality. None of the patients had a clinically or grossly
Figure 4 Adenoid basal carcinoma of the cervix.
recognizable cervical lesion. The depth of stromal invasion ranged from 2 to 10 mm (mean, 4.3 mm), exceeding 3 mm in six tumors. Treatment was predominantly surgical, with conization alone being employed in three patients. One hundred and four lymph nodes were removed from five patients, none of which contained metastases. At the time of publication, nine patients were alive without disease at a mean follow-up period of 30 months (range, 4–82 months), and three had died without disease after 24, 63, and 87 months. The prognosis of ABC is excellent, with few clinical features of an invasive cervical carcinoma. Small Cell Carcinoma
Small cell carcinoma of the cervix (SCCC) was first described in 1957 and accounts for 3% of all cervical neoplasms. On the basis of contemporary series, the median age is 43 years at diagnosis (range, 23–75 years). Histologically, these tumors are indistinguishable from small cell carcinoma in other sites, with small cells containing scant cytoplasm, hyperchromatic nuclei, and a high nuclear-cytoplasmic ratio. Electron microscopy often reveals dense-core, membranebound, neurosecretory granules. Neuroendocrine markers are commonly used to assist in classification, with up to 80% of tumors staining for synaptophysin, chromogranin, or CD56 (neural cell adhesion molecule) or both. Similar to what has been observed among patients with small cell lung cancer, SCCC is characterized by frequent and early nodal metastases (50–60%), vascular invasion, and a relapse pattern consistent with hematogenous dissemination. For example, Viswanathan et al. observed a 66% relapse rate, with a course frequently characterized by the development of widespread distant metastases.51 Locoregional recurrence outside irradiated fields was also frequently observed. In the group studied, the overall survival rate at 5 years was only 29%, with none of the patients who had disease more extensive than FIGO stage IB1 or clinical evidence of lymph node metastases surviving their disease. In a multivariate analysis of different prognostic factors among 34 patients, Chan et al. documented that only those
TUMORS OF THE CERVIX
with early lesions amenable to extirpation were curable.52 Chang et al. have evaluated the role of adjuvant chemotherapy in patients who underwent primary radical hysterectomy for SCCC.53 From 1988 to 1996, 14 women received a combination of vincristine, doxorubicin, and cyclophoshamide alternating with cisplatin and etoposide (VAC/PE). Their outcomes were compared with those of nine patients treated from 1984 to 1988, eight of whom received a combination of cisplatin, vinblastine, and bleomycin (PVB) in the adjuvant setting. The stage distribution included 19 patients with stage IB tumors, and 4 with stage II tumors. Ten of the 14 patients (68%) who received VAC/PE had no evidence of disease during a median follow-up of 41 months, whereas only three of the nine (33%) who received PVB or another regimen survived. All 10 women who died failed at distant sites. Seventy percent of patients without nodal involvement at the time of surgery and 35% of those with lymphatic metastases survived. Cisplatin plus etoposide with the early use of irradiation is generally regarded as the standard regimen for small cell lung cancer, and by extrapolation, may have some activity in SCCC. Hoskins et al. have reported a 14-year institutional experience involving 31 patients with SCCC.54 The clinical stage distribution included stage I (n = 16), stage IIA (n = 3), stage IIB (n = 3), stage III (6), stage IVB (n = 1), and unknown (n = 2). Importantly, clinical staging significantly underestimated the true disease extent, with 38% of patients (n = 13) upstaged on the basis of further imaging studies. Seventeen patients were treated from 1988 to 1995 on protocol SMCC, which included cisplatin, etoposide, and involved-field irradiation with concurrent chemotherapy; 14 women treated from 1996 to 2002 also received carboplatin and paclitaxel, along with para-aortic irradiation (protocol SMCC2). None of the patients underwent radical surgery. The 3-year overall and failure-free survival rates for the study group was 60 and 57%, respectively. The survival results were equivalent between SMCC and SMCC2, with the latter regimen associated with increased hematologic toxicity and decreased hospital admissions for emesis and dehydration. Carcinoid Tumors
Well-differentiated carcinoid tumors of the cervix were originally described by Albores-Saavedra et al.55 Histologically, they are similar to intestinal carcinoid tumors and contain neurosecretory granules that can be observed by electron microscopy. The tumors grow in a trabecular, nodular, or cordlike pattern, and rosette-like structures are common. The classical carcinoid has neoplastic cells with finely granular cytoplasm and oval spindle-shaped nuclei. Mitoses are infrequently observed. Greater than 70% are argyrophilic and stain with synaptophysin, chromogranin A, and neuronspecific enolase. Syndrome X and elevated serum serotonin levels have been associated with cervical carcinoid tumors. Although there has been one report of carcinoid syndrome occurring in a patient with an atypical carcinoid tumor the cervix, there have been no cases of carcinoid syndrome associated with the classic cervical carcinoid. While the growth pattern has been
511
previously described as indolent, some reports have been notable for a malignant course, with both local and distant metastases. For example, Seidel and Steinfeld reported brain metastases manifesting 4 years after the diagnosis and primary treatment of a stage IB tumor.56 In their review of reported cases, they found that even patients with early stage lesions may die of disseminated disease.
Sarcomas A variety of sarcomas can arise from the cervix, including carcinosarcoma, leiomyosarcoma, epithelioid leiomyosarcoma, Mullerian adenosarcoma (MA), and extrauterine ESS. Granulocytic sarcoma (also known as chloroma), and embryonal rhabdomyosarcoma along with its variant, sarcoma botryoides (SB), will be discussed separately. Sarcomatoid carcinomas, which lack a malignant stromal or mesenchymal component, were described in an earlier section. Wright et al. conducted a single institution review spanning nearly two decades, which included 1583 cervical neoplasms and found only eight sarcomas (0.005%), five of which were carcinosarcomas.57 Five of their patients (62.5%) were alive without evidence of disease and one patient (12.5%) was alive with disease at a median followup of 2.2 years. Among the disease-free survivors, two had been diagnosed with carcinosarcoma, and one each with a leiomyosarcoma, unspecified sarcoma, and extrauterine ESS. Approximately 50 cases of cervical carcinosarcoma have been reported. The largest review was conducted by Clement et al. who noted that most patients present with vaginal bleeding and a cervical mass.58 Disease is confined to the cervix in the majority of patients, although in the series by Wright et al.,57 four of five patients with cervical carcinosarcoma had bulky stage IB2 tumors and one had a large stage IIIA tumor. Histologically, the tumors exhibit a malignant epithelial component in conjunction with a malignant stromal component. Both homologous and heterologous sarcomatous elements have been reported in cervical carcinosarcomas. Interestingly, integrated HPV DNA has been detected in not only the epithelial components, but also in the sarcomatous elements of some tumors. Radical abdominal hysterectomy with bilateral salpingoophorectomy and lymphadenectomy is a reasonable approach for tumors confined to the cervix. Less than 25 cases of primary cervical leiomyosarcoma have appeared in the literature. Most afflicted patients are perimenopausal and present with vaginal bleeding and a cervical mass. Diagnostic criteria have been outlined by Bell et al. and include a combination of high mitotic rate, coagulative tumor cell necrosis, or high-grade cytologic atypia.59 Treatment recommendations have been extrapolated from the soft tissue sarcoma and uterine sarcoma literatures. When disease is confined to the cervix, radical abdominal hysterectomy with bilateral salpingoophorectomy may be considered. Because of the low incidences of nodal spread in uterine leiomyosarcoma (3.5%) and in soft tissue sarcoma (5.8%), the value of routine lymphadenectomy has been questioned for primary cervical leiomyosarcomas. MA is a variant of the mixed mesodermal tumor of the uterus. It is composed of benign epithelial (glandular) and
512
GYNECOLOGICAL CANCERS
malignant stromal components. The tumor most commonly occurs within the uterine body in postmenopausal women, sometimes in association with tamoxifen therapy. Roth et al. first described its location in the cervix and the presence of heterologous elements (i.e. cartilage, bone, skeletal muscle, etc.) in 1976. Only 14 reports have been published subsequently. Ramos et al. have noted that an interesting feature of cervical MA with heterologous elements is the age of clinical presentation with five postmenarchal patients and eight young women (range, 18–45 years) diagnosed with this unusual neoplasm.60 Many had a history of recurrent polyps. Cartilage and striated muscle were the most common heterologous elements encountered in 13 of the reported cases, and crossed cytoplasmic striations were observed in half of the study population. Many authors have recommended radical or extrafascial hysterectomy and bilateral salpingoophorectomy, with adjuvant radiation therapy reserved for deeply infiltrating tumors. Ten of the 15 patients (66%) reviewed by Ramos et al. were known to be alive without disease at 2.5–7 years of follow-up.60 Because the lesion seems to appear in the earliest stages of the reproductive lifespan in women, local excision may be curative in rare cases when a pedunculated cervical tumor and uninvolved stalk is encountered, thereby allowing conservation of reproductive function. Epithelioid leiomyosarcoma is a rare variant characterized by a proliferation of predominantly round and polygonal epithelioid cells with eosinophilic cytoplasm in a malignant tumor component of smooth muscle cells. Most tumors have occurred in the uterine body, with only four reported cases having arisen from the uterine cervix. The age distribution ranged from 47 to 72 years, and initial treatment included total abdominal hysterectomy and bilateral salpingoophorectomy in all cases.61 Three patients received adjuvant therapy, and three patients were alive without disease at 4, 10, and 20 months of follow-up. Finally, extrauterine ESS histologically resembles ESS and commonly arises within foci of endometriosis. The tumors have been found in the peritoneum, omentum, ovary, and in the cervix. In two of the three cases of primary cervical extrauterine ESS, the tumor appeared to arise from an endocervical polyp. This rare entity is included in this chapter only for the sake of completeness. Embryonal Rhabdomyosarcoma
E-RMS is a highly malignant tumor that accounts for 4–6% of all malignancies in childhood and young adults. It is the most common soft tissue sarcoma in this age-group and occurs primarily in the head and neck region, the genitourinary tract, and the extremities. The Intergroup Rhabdomyosarcoma Study (IRS) Group recognizes three major histologic subtypes: embryonal, alveolar, and undifferentiated. Histologically, SB contains malignant rhabdomyoblasts within a loose, myxoid stroma. Unlike alveolar E-RMS, SB is not associated with a unique genetic translocation. The SB is a variant of E-RMS and manifests with a “grapelike” appearance due to a layer of spindle cells pushing up beneath the mucosa in polypoid masses. Although SB is usually found in the infantile vagina, it can also arise
from the vulva, the uterus, and the cervix. Interestingly, while vaginal RMS manifests normally before the age of 4 (mean age, 23.5 months), cervical RMS usually presents during the second decade of life (mean age, 14 years). Brand et al. published the first comprehensive review of cervical SB, adding four of their own cases to the 17 that had appeared in the literature up to 1987.62 After 68 months of followup, 80% of the patients were alive. Among five patients (24%) with recurrent disease, three had died of disease. The treatment of SB has undergone rigorous scrutiny and dramatic changes over the decades. Traditionally, treatment included ultraradical or radical, fertility-compromising surgery. For example, during the 1960s pelvic exenteration was the treatment of choice; however, a comprehensive review showed that the results were unsatisfactory. Multiagent chemotherapy or pelvic irradiation with limited surgery or both resulted in a considerable improvement in survival during the 1970s. The 1980s were dominated by combined treatment options including radical hysterectomy; however, during the 1990s there was a gradual shift toward less invasive and more organ-sparing procedures such as local excision, polypectomy, trachelectomy, or conization with or without adjuvant chemotherapy. Importantly, in recent years it has been recognized that cervical SB behaves less aggressively than vaginal or uterine SB. Given the predilection of this tumor to afflict young children and adolescents, functional preservation of the bladder and reproductive organs is implicit when considering longterm quality of life. Fertility-sparing surgery appears to be appropriate in IRS clinical classification group I patients (i.e. those with localized disease, completely resected, with no regional nodes involved). The most widely used chemotherapy regimen includes vincristine, dactinomycin, and cyclophosphamide (VAC), and 6 to 12 cycles should permit a resumption of menstrual function and preservation of reproductive capacity. In 2003, Behtash et al. reviewed the literature and identified 17 additional cases since 1987, and added two of their own.63 All 19 patients have survived beyond 24 months (range, 24–96 months). Eleven patients underwent conservative surgery, including polypectomy, dilatation and curettage, trachelectomy, partial excision and local excisions, three of whom did not receive adjuvant chemotherapy. Two additional reports of cervical SB in patients treated with fertility-sparing surgery and adjuvant chemotherapy appeared in 2004.64,65 Granulocytic Sarcoma
GS is an extramedullary tumor of malignant granulocytic progenitor cells that heralds, accompanies, or signals relapse of acute myelogenous leukemia (AML). GS has been reported in 3–5% of AML patients and because it can precede other manifestations of leukemia, misdiagnosis often occurs. A conclusive diagnosis of GS requires demonstration of granulocytic differentiation of the tumor cells, for which the Leder stain and the antilysozyme immunoperoxidase stain are particularly useful.66 GS of the cervix is very rare and is also called chloroma because of its greenish appearance. The majority of patients
TUMORS OF THE CERVIX
with cervical GS present with vaginal bleeding and abdominal pain and may also have other systemic symptoms. Chemotherapy has been the mainstay of treatment for GS. Unfortunately, the overall 2-year survival rates for all patients with GS in the literature is 6%, and none of the reported patients lived 5 years.
Miscellaneous Melanoma
Only 5% of melanoma in women are located in the genitalia, with the vast majority occurring in the vulva. Rarely, the ovary, uterus, or uterine cervix may be the primary site of origin. Historically, the lack of melanocytes in the vaginal and cervical mucous membranes has deterred the acceptance of primary melanomas in these areas; however, in 1959, Cid reported the presence of melanin-containing cells in 3.5% of cervices. The principal requirement is the evidence of junctional activity of the cervical epithelium along with the absence of any demonstrable skin, vaginal, or retinal lesion. In a review of the published literature, Cantuaria et al. have identified 27 cases of primary malignant melanoma of the cervix.67 The mean age at diagnosis was 55 years (range, 26–78 years), and 83% presented with abnormal vaginal bleeding. In the majority of cases, an exophytic cervical lesion was encountered, varying in color, which included red, brown, gray, black, and blue. The tumor was typically fragile and easily bled to touch. Because of the extreme rarity of this lesion, the FIGO clinical staging system is recommended rather than one of the microstaging systems of Clark et al., Breslow et al., or Chung et al. Importantly, 88% of the patients identified by Cantuaria et al. presented with stage I (n = 12) and stage II (n = 9) disease.67 Treatment of melanoma is primarily surgical, and a variety of procedures have been applied to this disease when it originates in the cervix. When there is only clinical involvement of the cervix, we recommend a radical abdominal hysterectomy with the performance of a concomitant upper vaginectomy as needed to obtain satisfactory margins of at least 2 cm. Because the presence of lymphatic metastases is negligible in patients with melanomas <1-mm thick, we do not advocate elective lymphadenectomy for such lesions. The management of clinically negative regional lymph nodes for thicker tumors, and especially for those >4-mm thick, is controversial. Randomized trials by the Mayo Clinic have not demonstrated enhanced survival among patients with melanoma of the extremities who underwent elective lymphadenectomy when compared with similar patients treated with wide excision alone. For these reasons, unless a patient is interested in investigational adjuvant therapy that may require lymph node biopsy as part of the protocol eligibility criteria, we do not recommend elective lymphadenectomy for this disease. Prophylactic pelvic and para-aortic lymphadenectomy for grossly enlarged lymph nodes may have a palliative role. Radiation therapy has not been studied in the adjuvant setting for this disease, and may be employed for palliation of unresectable, advanced tumors and possibly in the setting when satisfactory surgical resection has not been accomplished. The role of adjuvant chemotherapy and biologic
513
modifiers such as interferon are also topics of debate, with disappointing results when chemotherapy has been used for metastatic cutaneous melanomas. Response rates for dacarbazine have ranged between 15% and 20%. Many of the 27 patients reported by Cantuaria et al. lack follow-up data.67 For five patients with stage I disease and for six with stage II disease, the mean survival was 49 and 41 months, respectively. Despite the majority of patients having been diagnosed with early stage tumors, only two have survived beyond 5 years. It appears that the prognosis is unfavorable when compared with SCCAs and typical endocervical adenocarcinomas of the cervix. Extrapolating from the cutaneous melanoma literature, even long diseasefree intervals are no guarantee against recurrent disease with this unpredictable tumor. Fortunately, because most patients present with early stage disease, many are candidates for surgical therapy, at least initially. Lymphoma
Lymphomas account for 3.5% of all cancers in women. Twenty-five percent arise in extragonadal tissues, with the gastrointestinal tract and skin being most common. Secondary involvement of the genital tract may occur in up to 40% of disseminated lymphomas. Primary genital tract lymphomas, however, account for only 1.5% of extranodal lymphomas, of which 0.6% are of cervical origin. Primary cervical lymphomas are defined as lymphomas that originate and localize from the uterine cervix without any myometrial involvement and without any evidence of leukemia at the time of diagnosis. They were first reported by Freeman et al. in 1972. The etiology and pathogenesis of cervical lymphomas are perplexing, with some authorities attributing the increase in incidence of extranodal lymphomas during the preceding two decades to be secondary to the increase in immunosuppressive therapies, human immunodeficiency virus (HIV) infections, environmental toxins, and even chronic inflammatory conditions. Approximately 80% of patients with cervical lymphoma are premenopausal (range, 20–80 years), and when compared to other non-Hodgkin’s lymphomas (NHLs), there is a tendency to occur in younger age-groups. Although some reports claim that patients with systemic lymphoma have more abnormal Pap smears than controls, primary cervical lymphomas are rarely diagnosed by screening cytology because they originate from the cervical stroma, and the overlying squamous epithelium is generally preserved. Up to 67% of these tumors may present with a subepithelial mass without obvious ulceration. In a review from 1983 to 2003, Dursun et al. identified 31 patients (including two of their own) for whom abnormal lymphoid cells were present in the cervical smears of only 6.5% of cases.68 Patients typically present with abnormal vaginal bleeding or discharge, although pelvic pain and dyspareunia may also occur. Interestingly, fever, night sweats, and weight loss, which are common among patients with systemic lymphomas, are rarely reported in women with primary cervical lymphoma. On examination, there is cervical-uterine enlargement with fixation, and vaginal or parametrial infiltration or
514
GYNECOLOGICAL CANCERS
both may be palpable. The differential diagnosis of cervical lymphomas include benign chronic inflammation, poorly differentiated and small cell cervical carcinomas, sarcomas, and lymphoma-like lesions. Because therapy for cervical lymphomas differs from that of other cervical cancers, it is imperative that the correct diagnosis is made. Histologic analysis of a deep cervical biopsy is diagnostic, with the appearance being similar to lymphomas from other sites. Occasionally, immunophenotypings may be required as diagnostic adjuncts. Although the World Health Organization classification scheme has replaced the Ann Arbor staging system (which includes stages IE, IIE, IIIE, and IVE), it must be recognized that most reports from the literature have used the latter system. The histologic classification is according to the International Working Formulation for NHLs and includes low grade (A –C), intermediate grade (D –G), and high grade (H –J). Treatment modalities employed in the management of cervical lymphoma have included chemotherapy alone, neoadjuvant chemotherapy followed by surgery, radiotherapy alone, or radiotherapy combined with either chemotherapy or surgery. The most commonly used chemotherapy regimen is CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone). Some authorities have advocated surgical resection in localized lymphoma (e.g. large loop excision of the transformation zone (LLETZ), trachelectomy, or hysterectomy), and in well-selected cases this may be a reasonable approach. There is no clear evidence, however, that hysterectomy improves survival for this group of patients. Chan et al. have conducted an extensive review of the literature and identified 64 patients (including six of their own).69 The disease-free survival rate among the 39 patients with stage IE disease (i.e. involvement of a single extralymphatic organ or site) was 87%. While patients with localized stage IE disease with nonbulky lesions will often respond to primary surgery, chemotherapy, or radiotherapy alone, combined therapy seems to be preferred in most centers. Since the survival rates in patients with early stage cervical lymphomas are excellent, young women who desire future childbearing will benefit from combination chemotherapy. For patients with more advanced disease, combination chemotherapy with tailored radiotherapy appears to be favored. Curiosities
There have been at least 10 reports of a peripheral primitive neuroectodermal tumor (pPNET) of the cervix.70 These tumors should be treated in accordance with the protocol for bony Ewing’s sarcoma with induction chemotherapy, surgery, and consolidation chemotherapy. Gestational choriocarcinoma has also been reported, most possibly arising from a preexisting cervical pregnancy or displaced intrauterine molar tissue.71 Finally, primary tumors of germ cell origin have arisen from the cervix. These include cases of mature teratomas with lymphoid hyperplasia72 or pulmonary differentiation,73 as well as malignant germ cell tumors, including an immature teratoma,74 yolk sac tumor,75 and choriocarcinoma.76 Dysgerminoma has not yet been reported to originate in the cervix.
SPECIALIZED CLINICAL SCENARIOS Fertility-sparing Surgery in Early Stage Disease It is estimated that 10 to 15% of cases of cervical cancer are diagnosed during the childbearing years. In western countries, women often delay childbearing until the midto late-30s, and for many the idea of losing their uterus to treat their cancer is devastating. Although patients can be offered alternative reproductive options, such as in vitro fertilization with embryo cryopreservation and gestational surrogacy, those approaches remain fairly complex and are not ethically acceptable or affordable to many. The question whether some lesions can be addressed in a conservative surgical manner has gained considerable attention during the preceding decade. Fertility-sparing surgery is appealing to this select group of patients. Cervical Conization
For women with stage IA1 squamous cell cervical cancer, because the rates of parametrial involvement and nodal metastases are negligible, cold-knife conization alone may be suitable under certain conditions (e.g. lack of LVSI, nulliparity, desirous of childbearing, etc.). Alternatively, laser CO2 cervical conization under local anesthesia may be contemplated for microinvasive carcinoma, with a study of 62 patients by Diakomanolis et al. noting a 6.6% recurrence rate (CIN I only) over a mean follow-up period of 54 months.77 The use of cervical conization in the management of “microinvasive” endocervical adenocarcinoma may also be an option for select women who wish to preserve fertility. One must always keep in mind that the primary objective is to clear the cancer with a satisfactory margin; all concerns regarding future pregnancy are secondary. Vaginal Radical Trachelectomy
Patients with stage IA2 disease have a 6.3% risk of nodal metastases and therefore treatment must include a formal pelvic lymphadenectomy along with bilateral parametrectomy. Thus, conization is not sufficient in these cases. In 1987, Dargent designed a fertility-preserving operation for stage IA2 and some IB1 lesions.78 A variant of the classical Shauta operation of vaginal radical hysterectomy, the vaginal radical trachelectomy (VRT) is always preceded by bilateral laparoscopic lymphadenectomies. The VRT is performed with division of the uterus underneath the isthmus, and at the completion of the procedure, the uterus is sutured to the vagina. Oncologically, the technique is satisfying as a wide margin around the lesion is obtained containing the parametria and the upper vagina, but leaving the body of the uterus in situ. Intraoperative mandatory frozen section analysis should be performed on both the nodal tissue and the upper endocervical margins of the trachelectomy specimen. Upon a review of 61 VRT specimens, Tanguay et al. recommend a complementary radical hysterectomy when the tumor extends to within 5 mm of the margin. These investigators also prefer a longitudinal rather than transverse frozen section when a macroscopic lesion is present, as it permits measurement of the distance between the tumor and the endocervical margin.
TUMORS OF THE CERVIX
Aggregate data from four centers (Dargent in France, n = 82; Covens et al. from Toronto, n = 58; Roy and Plante from Quebec n = 44; Shepherd et al. in the United Kingdom, n = 40)78 – 81 have documented a 3.1% (n = 7) recurrence rate among 224 patients, three of which were at distant sites. The data reflecting obstetrical outcomes are quite encouraging and have been noteworthy for 96 pregnancies, of which 51 live births have resulted. Covens et al. reported that all women in their series became pregnant within 12 months of attempting to conceive, giving a conception rate of 37% at 1 year.79 Importantly, most women were able to become pregnant without assisted reproductive technology. There have been 12 second trimester losses because of cervical weakness. Bernardini et al. presented the obstetric outcomes of 80 patients from the Toronto group.82 Thirty-nine patients attempted to conceive during a median follow-up period of 11 months, resulting in 22 pregnancies in 18 patients. Of the 22 pregnancies, 18 were viable, with 12 progressing to term and delivering by cesarean section. Preterm premature rupture of the membranes was the primary cause of preterm delivery. We currently recommend placement of a transabdominal cerclage over the mouth of the lower uterine segment with subsequent delivery of the neonate by cesarean section. In an updated series of 72 cases and review of the literature, the group from Quebec still maintain that the VRT is an oncologically safe procedure in well-selected patients with early stage disease.83 Excluding a patient with a small cell neuroendocrine tumor who had a rapid recurrence and died, there were two recurrences (2.8%) and one death (1.4%) at a median follow-up of 60 months. The authors suggest that lesion size beyond 2 cm appears to be associated with a higher risk of recurrence. Additionally, upon discovering a central pelvic recurrence of an endocervical adenocarcinoma 7 years after VRT, Bali et al. have raised the question whether patients treated by VRT (in particular, those with adenocarcinoma) should be offered hysterectomy once childbearing has been accomplished.84 By allowing for the preservation of the body of the uterus and thereby the potential for reproductive function, the VRT emerges as a true breakthrough in the management of young women with early-stage cervical cancer. VRT is currently the fertility-sparing procedure with the most available data supporting its use. Although these results are encouraging, there is lack of level I evidence (i.e. randomized controlled trials) comparing safety and survival rates between conservative and radical methods. Therefore, these techniques should be used by fully trained operators, with the understanding that it is not the standard treatment for this disease at present. In our opinion, the technique can be considered in conjunction with laparoscopic transperitoneal lymphadenectomy in the patient who strongly desires future fertility and harbors a stage IA1 lesion with LVSI, a IA2 lesion or a IB1 tumor <2 cm in diameter. Additional requirements include squamous cell histology when dealing with a clinical lesion, and limited endocervical involvement as determined by colposcopy and magnetic resonance imaging (MRI).
515
Abdominal Radical Trachelectomy
Potential benefits of the abdominal approach for radical parametrectomy include wider parametrial resection, possible lower intraoperative complication rates, and techniques familiar to most gynecologic oncologists. Ungar et al. performed this procedure in 30 patients, 10 with FIGO stage IA2 tumors, five with stage IB1 lesions, and five with stage IB2 .85 During a median follow-up of 47 months, no recurrences have been detected. Among five women who attempted to conceive, three women have fallen pregnant, resulting in one first trimester pregnancy loss and two cesarean section deliveries at term. Although this technique has not gained wide application, the authors contend that it appears to provide equivalent oncological safety to a standard Wertheim radical hysterectomy. The performance of a satisfactory VRT can be technically complex when dealing with the proximal endocervical margins of an adenocarcinoma. We have moved away from the vaginal approach for glandular lesions, and exclusively perform the radical trachelectomy abdominally when confronted with an early-stage endocervical adenocarcinoma in patients who desire fertility preservation.
Cervical Cancer in the Setting of Suboptimal Surgery Cervical Stump Cancer
From 1959 to 1987, 145 patients were treated for an infiltrating carcinoma of the cervical stump at Radiumhemmet.86 This group represented 2.2% of all cervical cancers at that institution. Three control cases to each case were selected from the cohort of cervical carcinoma cases and matched to year of treatment, stage, histology, and age. The treatment program included two brachytherapy applications (separated by three weeks) followed by external pelvic irradiation. The dose of irradiation from the intracavitary application given to the stump cancers was lower than that given to comparable cases of the common cervical cases. No evidence was found of poorer long-term prognosis for radiologically treated SCCA of the uterine stump compared to that of the ordinary cervical carcinomas. Stump cancers of the adenocarcinoma type had a worse prognosis than adenocarcinomas in an intact uterus (p < 0.07) and also compared with stump cancers of the squamous epithelial type (p = 0.05). The complication rate was higher for the stump cancer cases compared with that for cervical cancers arising in the intact uterus. The mean time interval from subtotal hysterectomy to the stump cancer diagnosis was 17.6 years, with a range from 1 to 46 years. It is necessary to weigh the possible gains associated with a subtotal hysterectomy against the relatively low risk to fall victim of a stump cancer. We do not recommend a subtotal hysterectomy in any patient with a history of cervical dysplasia or abnormal Papanicoloau testing. Furthermore, subtotal hysterectomy should be avoided in populations with restricted access to screening programs for cervical cancer. The ‘‘Cut-Thru’’ Hysterectomy
Two management strategies exist for the patient harboring an occult cervical tumor who undergoes a total abdominal
516
GYNECOLOGICAL CANCERS
hysterectomy, and for whom subsequent clinical staging confirms the presence of an early-stage carcinoma (FIGO stage IA1 with LVSI, IA2 , or occult IB1 ). Radical parametrectomy with upper vaginectomy and bilateral pelvic lymphadenectomies is an acceptable option for the nonobese, young patient in whom it is highly desirable to avoid radiotherapy. Leath et al. identified 23 patients who had undergone total abdominal hysterectomy and were found to have occult cervical carcinoma on final pathology; two with FIGO stage IA2 lesions and 21 with occult FIGO stage IB1 cancers.87 The most common indication for performing the extrafascial hysterectomy (48%) was for carcinoma in situ of the cervix. All patients underwent radical parametrectomy, lymphadenectomy, and upper vaginectomy, with four patients (17%) having metastases to the pelvic nodes or evidence of tumor at the surgical margin. There were seven (30%) operative complications, including four patients who received blood transfusions. With a median follow-up of 61 months, the overall 5-year survival reached 96%. We do not advocate a radical parametrectomy with upper vaginectomy and lymphadenectomy for those cut-through cases in which there is invasive tumor at the surgical margin. Munstedt et al. examined the consequences of inadvertent suboptimal primary surgery in carcinoma of the cervix, comparing the outcomes of 80 women treated with simple hysterectomy and adjuvant radiotherapy to 89 women who underwent radical hysterectomy alone to 119 patients who underwent radical hysterectomy and received adjuvant radiotherapy.88 There was a trend toward better overall survival in the radical hysterectomy group. The disease-free survival was significantly better in the radical hysterectomy group followed by the simple hysterectomy plus adjuvant radiotherapy group (P < 0.002). Aborted Radical Hysterectomy
A small percentage of patients with early-stage cervical carcinoma who are deemed operable will undergo laparotomy and have the radical operation aborted secondary to the intraoperative assessment of disease extension or the presence of extensive adhesive disease, endometriosis, or inflammatory processes or both. Fortunately, many of these patients can be salvaged with radiation therapy. We recommend full lymphadenectomies, and when possible, an extrafascial hysterectomy should be performed to get the cervix out. An omental carpet can then be mobilized to provide a fresh blood supply to the pelvis in anticipation of parametrial radiation therapy postoperatively. If there is obvious parametrial involvement that had not been appreciated preoperatively, then the uterus and cervix should be left in situ so as to avoid cutting into the tumor and distributing it throughout the pelvis. Leath et al. identified 23 patients who had an aborted radical hysterectomy at the University of Birmingham, including, 17 with FIGO stage IB1 , four with FIGO stage IB2 , and two with FIGO stage IIA lesions.89 The reasons for aborting the radical hysterectomy included pelvic extension in 11 patients, positive pelvic nodes in seven, and positive paraortic nodes in five. All 23 received postoperative radiation therapy, with 12 receiving concurrent chemotherapy.
Four patients (17%) had radiation-associated complications, and six women (26%) experienced a recurrence. The 5year overall survival was 83%, with a median follow-up of 59 months.
The Immunocompromised Host Women infected with the human immunodeficiency virus (HIV) have an increased risk of HPV infection and preinvasive stages of cervical neoplasia. Lomalisa et al. performed a retrospective review of 60 HIV-seropositive and 776 HIV-seronegative new cases of invasive cervical carcinoma seen in Johannesburg.90 The HIV seroprevalence was 7.2%, with SCCA being the histologic subtype in more than 90% of both cohorts of patients. Although the HIVpositive patients presented with invasive cervical cancer almost 10 years earlier than the HIV-negative patients, there was no difference in the distribution of advanced lesions in the two groups. However, HIV-seropositive patients with CD4 cell counts less than 200 mm−3 had significantly more advanced-stage disease than HIV-seronegative patients (77% vs 55.8%, respectively (P = 0.041)). Ozsaran et al. investigated the risk of cervical intraepithelial neoplasia and the coexistence of HPV infection in 48 renal transplant patients receiving immunosuppressive therapy who underwent cervical Papanicoloau testing and colposcopic examinations.91 Genital neoplasia was encountered in 20 of the 48 renal transplant patients. Jonas et al. studied 500 liver transplantations performed in 458 patients.92 The median follow-up was 50 months for all patients with de novo neoplasias detected in 33 patients (7.2%), including seven with intraepithelial neoplasias of the cervix. Only a positive T-crossmatch and a low CD4+ /CD8+ ratio in patients receiving cyclosporine-A-based immunosuppression demonstrated a significant correlation with the development of a de novo tumor. In otherwise uncomplicated patients infected with HIV or receiving immunosuppressive therapy to prevent organ transplant rejection, our protocol is to obtain cervical cytology samples every 6 months and perform colposcopic evaluation of the cervix annually. HIV-seropositive women with CD4 counts less than 200 mm−3 as well as organ transplant recipients with a low CD4+ /CD8+ ratio may benefit from even more frequent screening.
Cervical Cancer in Pregnancy Cervical cancer complicates approximately one in 1200 pregnancies. Hacker et al. noted that the median age was 33.8 years (range, 17–47 years), and the average parity is 4.5.93 Presenting symptoms include abnormal vaginal bleeding (63%), vaginal discharge (13%), postcoital bleeding (4%), and pelvic pain (2%). Importantly, 20% of patients are asymptomatic. The diagnosis of microinvasive carcinoma in pregnancy is typically established with colposcopic-directed biopsy, and in a minority of cases by a shallow “coin” biopsy or wedge excision of the area under suspicion. It is imperative that an experienced colposcopist examines the pregnant cervix, as many of the physiologic changes of pregnancy can result in an abnormal appearance to untrained eyes.
TUMORS OF THE CERVIX
517
PAPANICOLOAOU TEST SCCA
HSIL
LSIL
ASC-H
ASCUS
Normal
AGC-NOS
AGC-H
AIS
HC II (HPV Testing)
Colposcopy Impression: ≥ CIN II
Positive or unknown
Negative
Impression: normal or CIN I
Colposcopy Squamous or glandular dysplasia
Biopsy
Repeat Pap/colposcopy postpartum if Pap < HSIL; Repeat colposcopy every trimester if Pap ≥ HSIL
Repeat colposcopy every trimester if Pap AGC-NOS; consider sonography, MRI and/or excisional biopsy for AGC-H or AIS
in ic ro M
Figure 6
III IN C ≤ Suspicious for microinvasion
va si o
n
Biopsy
Normal
≤ CIN III/CIS Repeat colposcopy or AIS every trimester CIN III/CIS or AIS
Repeat colposcopy every trimester
Suspicious for microinvasion
Excisional biopsy
Excisional biopsy Microinvasion
Microinvasion Figure 6 Repeat Pap postpartum Microinvasion
CIN III/CIS
Figure 5 Potential management algorithm for cervical intraepithelial neoplasia in pregnancy. SCCA, squamous cell carcinoma; HSIL, high-grade squamous intraepithelial lesions; LSIL, low-grade squamous intraepithelial lesions; ASC-H, atypical squamous cells; ASCUS. atypical squamous cells of undetermined significance; AGC-Nos, atypical glandular cells, not otherwise specified; AGC-H, atypical glandular cells; AIS, adenocarcinoma in situ; CIN, cervical intraepithelial neoplasia; CIS, carcinoma in situ; MRI, magnetic resonance imaging; HPV, human papillomavirus.
The FIGO staging system applies in pregnancy. To avoid the risks of ionizing radiation exposure to a developing fetus, we recommend an ultrasound of the kidneys to evaluate for hydronephrosis, and an MRI of the pelvis to evaluate the parametria and lymph nodes. A chest radiograph may be performed with appropriate abdominal shielding to exclude pulmonary metastases. Figures 5 and 6 depict potential management algorithms for cervical intraepithelial neoplasia and invasive cervical carcinoma in pregnancy. For patients with carcinoma in situ or microinvasive carcinoma of any gestational age, the pregnancy may advance to term with regular colposcopic surveillance every trimester. Patients with carcinoma in situ may be offered a trial of labor, but in patients harboring invasive disease, because of the risk of episiotomy-site metastases and the theoretical risks of dissemination into the vascular system, we recommend a cesarean section when fetal maturation is demonstrable followed by intrapartum total abdominal hysterectomy (for FIGO IA1 tumors) or modified radical abdominal hysterectomy with bilateral pelvic lymphadenectomies (for FIGO IA2 or occult IB1 lesions). We advocate the deployment of a vertical uterine incision so as to leave the lower uterine segment undisturbed for subsequent detailed pathologic examination. Historically, patients with clinical stage IB lesions remote from term have been recommended to undergo pregnancy termination followed by immediate treatment. This dogma has been challenged, with multiple
reports of a safe outcome for mother and child with a deliberate delay in therapy to permit gestational advancement. For example, Duggan et al. reported a mean diagnosis-totreatment interval of 144 days (range, 53–212 days) in eight pregnant women with FIGO stage I tumors, all of whom have remained disease free at a median follow-up of 23 months after therapy.94 Monk and Montz examined the surgical management of cervical cancer complicating pregnancy.95 They identified 13 patients treated with radical hysterectomy and lymphadenectomy with the fetus in situ (mean blood loss, 777 mL), and eight others treated with cesarean delivery followed by radical hysterectomy and lymphadenectomy (mean blood less, 1750 mL). The mean operative time was 281 minutes. Seven healthy infants were delivered. Twenty patients (95%) were alive and free of disease at 40 months follow-up. Patients with locally advanced disease in pregnancy should be offered immediate therapy. Sood et al. evaluated 26 pregnant women treated with radiation therapy, three of whom were treated with the fetus in situ during the first trimester.96 Spontaneous abortions occurred 20–24 days after the start of radiation at a mean dose of 34 Gy. Patients with locally advanced disease who refuse to interrupt the pregnancy present a clinical conundrum. A novel approach was reported by Tewari et al. who administered neoadjuvant cisplatin and vincristine during the early second
518
GYNECOLOGICAL CANCERS
IB1
Microinvasion
< 20 weeks
Pregnancy desired
Termination
Strongly desires pregnancy
(−)
IB2–IIA
Pregnancy not desired
(+)
_ Surgical candidate +
Cone
EBRT
IA1 TVH IA1 with LVSI, IA2, occult IB1, IA adenocarcinoma
RH-BPLND, + Abortion − LOT
Modified RH-BPLND
BT
IIB–IVA
IVB
Pregnancy termination
Termination of pregnancy strongly recommended Palliative chemotherapy
Neoadjuvant chemotherapy
Modified RH-BPLND, LOT
Colposcopy every trimester
Chemoirradiation
> 20 weeks
Antepartum fetal surveillance 23 weeks + Corticosteroids
NOTE: IB2 – IVB diagnosed after 20 weeks should have very limited delays; corticosteroids and delivery as soon as fetus is viable/mature.
Last dose neoadjuvant chemotherapy at 30 weeks
C/S-TAH (FIGO IA1 based on pregnant cone; does not desire fertility)
− Surgical candidate +
Delivery at term Route determined by obstetric indications; C/S may be advisable with LVSI
C/S, LOT
C/S-RH BPLND, LOT
EBRT + BT
Chemoirradiation
Postpartum
IA1 Surveillance Pap, ECC, colposcopy, and episiotomy scar (if applicable)
a
IA1with LVSI, IA2, occult IB1, IA adenocarcinoma
etri
IA1 desires fertility
LV dia SI, la me rge ter , de tumo ep r inv asi o
Cone (if not already done)
ram , pa ins s arg ode + M /or n and
n
Intrapartum
Amniocentesis 32–34 weeks
EBRT Modified RH-BPLND
Surveillance
TVH
Surveillance Pap, episiotomy scar (if applicable)
Figure 6 Potential management algorithm for invasive cervical carcinoma in pregnancy. TVH, total vaginal hysterectomy; EDRT, external beam radiation therapy; LVSI, lymphovascular space invasion; LOT, lateral ovarian transposition; BT, brachytherapy; TAH, total abdominal hysterectomy; FIGO, International Federation of Gynecology and Obstretics; ECC, endocervical curettage; PAP, papanicoloau test; c/s, cesarean section.
TUMORS OF THE CERVIX
and third trimesters.97 Tumor regression occurred, allowing definitive surgical treatment at the time of delivery of the healthy infants. While one patient has remained without evidence of recurrence for over 2 years, unfortunately, the second patient relapsed within 5 months of therapy. The prognosis for all stages is similar to that among nonpregnant women, with a favorable overall prognosis in pregnancy related to a greater proportion of patients presenting with early stage disease. For more advanced disease, pregnancy may have an unfavorable effect on prognosis because of problems with radiation dosimetry in pregnancy and the need to interrupt radiation therapy more frequently because of genital tract sepsis.
ACKNOWLEDGMENT Figures 1–4 courtesy of Fritz Lin, MD, Professor, Department of Pathology, University of California, Irvine – Medical Center, 101 The City Drive, Orange, CA 92868.
REFERENCES 1. Freund AW. Zu meiner methode der total uterus extirpation. Zbl Gynaek 1878; 2: 265. 2. Clark JG. A more radical method for performing hysterectomy for cancer of the uterus. Bull Johns Hopkins Hosp 1895; 6: 120. 3. Schauta F. Die Erweiterte Vaginale Totalexstirpation des Uterus bei Kollumkarzinom. Wien, Leipzig, Germany: Josef Safar, 1908. 4. Wertheim E. The extended abdominal operation for carcinoma uteri (based on 500 operative cases). Am J Obstet Gynecol 1912; 66: 169. 5. Roentgen WC. On a new kind of rays. Nature 1896; 53: 274 – 6. 6. Curie P, Curie MS, B´emont G. On a new, strongly radioactive substance, contained in pitchblende. C R Acad Sci Paris 1898; 127: 1215 – 7. 7. Pasteau O, Degrais P. The radium treatment of cancer of the prostate. Rev Mal Nutr 1911; 10: 363 – 7. 8. Taussig FJ. Iliac lymphadenectomy with irradiation in the treatment of cancer of the cervix. Am J Obstet Gynecol 1934; 28: 650 – 2. 9. Meigs JV. Radical hysterectomy with bilateral node dissection. A report of 100 patients operated five or more years ago. Am J Obstet Gynecol 1951; 62: 854. 10. Okabayashi H. Radical abdominal hysterectomy for cancer of the cervix uteri. Surg Gynecol Obstet 1921; 33: 335 – 41. 11. Brunschwig A. Complete excision of pelvic viscera for advanced cancer. Cancer 1948; 1: 177. 12. Papanicoloau GN, Trout HF. Diagnosis of Uterine Cancer by Vaginal Smears. New York: The Commonwealth Fund, 1943. 13. Gey GO, Coffman WD, Kubicek MT. Tissue culture studies of the proliferative capacity of cervical carcinoma and normal epithelium. Cancer Res 1952; 12: 264 – 5. 14. Durst M, et al. A papillomavirus DNA from a cervical carcinoma and its prevalence in cancer biopsy samples from different geographic regions. Proc Natl Acad Sci USA 1983; 80: 3812 – 5. 15. Boshart M, et al. A new type of papillomavirus DNA, its presence in genital cancer biopsies and in cell lines derived from cervical cancer. EMBO J 1984; 3: 1151 – 7. 16. Degefu S, et al. Verrucous carcinoma of the cervix: a report of two cases and literature review. Gynecol Oncol 1986; 25: 37 – 47. 17. Ng WK, Cheung LK, Li AS. Warty (condylomatous) carcinoma of the cervix. A review of 3 cases with emphasis on thin-layer cytology and molecular analysis for HPV. Acta Cytol 2003; 47: 159 – 66. 18. Koenig C, et al. Papillary squamotransitional cell carcinoma of the cervix: a report of 32 cases. Am J Surg Pathol 1997; 21: 915 – 21. 19. Ortega-Gonzalez P, Chanona-Vilchis J, Dominguez-Malagon H. Transitional cell carcinoma of the uterine cervix. A report of six cases with clinical, histologic and cytologic findings. Acta Cytol 2002; 46: 585 – 90.
519
20. Bais AG, et al. Lymphoepithelioma-like carcinoma of the uterine cervix: absence of Epstein-Barr virus, but presence of multiple human papillomavirus infection. Gynecol Oncol 2005; 97: 716 – 8. 21. Tseng CJ, et al. Lymphoepithelioma-like carcinoma of the uterine cervix: association with Epstein-Barr virus and human papillomavirus. Cancer 1997; 80: 91 – 7. 22. Brown J, et al. Sarcomatoid carcinoma of the cervix. Gynecol Oncol 2003; 90: 23 – 8. 23. Wright TC, Ferenczy A, Kurman RJ. Carcinoma and other tumors of the cervix. In Kurman RJ (ed) Blaustein’s Pathology of the Female Genital Tract, 5th ed. New York: Springer-Verlag, 2002. 24. Konishi I, et al. Mucin leakage into the cervical stroma may increase lymph node metastasis in mucin-producing cervical adenocarcinomas. Cancer 1990; 65: 229 – 37. 25. Fujiwara H, et al. Adenocarcinoma of the cervix. Expression and clinical significance of estrogen and progesterone receptors. Cancer 1997; 79: 505 – 12. 26. Hopkins MP, Morley GW. A comparison of adenocarcinoma and squamous cell carcinoma of the cervix. Obstet Gynecol 1991; 77: 912 – 7. 27. Lea JS, et al. Stage IIB-IVB cervical adenocarcinoma: prognostic factors and survival. Gynecol Oncol 2002; 84: 115 – 9. 28. Young RH, Scully RE. Villoglandular papillary adenocarcinoma of the uterine cervix. A clinicopathologic analysis of 13 cases. Cancer 1989; 63: 1773 – 9. 29. Utsugi K, et al. Clinicopathologic features of villoglandular papillary adenocarcinoma of the uterine cervix. Gynecol Oncol 2004; 92: 64 – 70. 30. Young RH, Clement PB. Endocervical adenocarcinoma and its variants: their morphology and differential diagnosis. Histopathology 2002; 41: 185 – 207. 31. Silverberg SG, Hurt WG. Minimal deviation adenocarcinoma (“adenoma malignum”) of the cervix: a reappraisal. Am J Obstet Gynecol 1975; 121: 971 – 5. 32. Hirai Y, et al. A clinicocytopathologic study of adenoma malignum of the uterine cervix. Gynecol Oncol 1998; 70: 219 – 23. 33. Herbst AL, Anderson D. Clear cell adenocarcinoma of the vagina and cervix secondary to intrauterine exposure to diethylstilbestrol. Semin Surg Oncol 1990; 6: 343 – 6. 34. Reich O, et al. Clear cell carcinoma of the uterine cervix: pathology and prognosis in surgically treated stage IB-IIB disease in women not exposed in utero to diethylstilbestrol. Gynecol Oncol 2000; 76: 331 – 5. 35. Rose PG, Reale FR. Serous papillary carcinoma of the cervix. Gynecol Oncol 1993; 50: 361 – 4. 36. Costa MJ, McIlnay KR, Trelford J. Cervical carcinoma with glandular differentiation: histological evaluation predicts disease recurrence in clinical stage I or II patients. Hum Pathol 1995; 26: 829 – 37. 37. Zhou C, et al. Papillary serous carcinoma of the uterine cervix: a clinicopathologic study of 17 cases. Am J Surg Pathol 1998; 22: 113 – 20. 38. Ferry JA, Scully RE. Mesonephric remnants, hyperplasia, and neoplasia in the uterine cervix. A study of 49 cases. Am J Surg Pathol 1990; 14: 1100 – 11. 39. Hart WR. Symposium part II: special types of adenocarcinoma of the uterine cervix. Int J Gynecol Pathol 2002; 21: 327 – 46. 40. Silver SA, et al. Mesonephric adenocarcinomas of the uterine cervix: a study of 11 cases with immunohistochemical findings. Am J Surg Pathol 2001; 25: 379 – 87. 41. Look KY, et al. An analysis of cell type in patients with surgically staged stage IB carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 1996; 63: 304 – 11. 42. Farley JH, et al. Adenosquamous histology predicts a poor outcome for patients with advanced-stage, but not early-stage, cervical carcinoma. Cancer 2003; 97: 2196 – 202. 43. Lea JS, et al. Adenosquamous histology predicts poor outcome in low-risk stage IB1 cervical adenocarcinoma. Gynecol Oncol 2003; 91: 558 – 62. 44. Gray HJ, et al. Glassy cell carcinoma of the cervix revisited. Gynecol Oncol 2002; 85: 274 – 7. 45. Hopkins MP, Morely GW. Glass cell adenocarcinoma of the uterine cervix. Am J Obstet Gynecol 2004; 190: 67 – 70.
520
GYNECOLOGICAL CANCERS
46. Thelmo WL, et al. Mucoepidermoid carcinoma of uterine cervix stage IB. Long-term follow-up, histochemical and immunohistochemical study. Int J Gynecol Pathol 1990; 9: 316 – 24. 47. Grayson W, Taylor LF, Cooper K. Adenoid cystic and adenoid basal carcinoma of the uterine cervix: comparative morphologic, mucin, and immunohistochemical profile of two rare neoplasms of putative ‘reserve cell’ origin. Am J Surg Pathol 1999; 23: 448 – 58. 48. Prempree T, Villasanta U, Tang CK. Management of adenoid cystic carcinoma of the uterine cervix (cylindroma): report of six cases and reappraisal of all cases reported in the medical literature. Cancer 1980; 46: 1631 – 5. 49. King LA, et al. Adenoid cystic carcinoma of the cervix in women under age 40. Gynecol Oncol 1989; 32: 26 – 30. 50. Brainard JA, Hart WR. Adenoid basal epitheliomas of the uterine cervix: a reevaluation of distinctive cervical basaloid lesions currently classified as adenoid basal carcinoma and adenoid basal hyperplasia. Am J Surg Pathol 1998; 22: 965 – 75. 51. Viswanathan AN, et al. Small cell neuroendocrine carcinoma of the cervix: outcome and patterns of recurrence. Gynecol Oncol 2004; 93: 27 – 33. 52. Chan JK, et al. Prognostic factors in neuroendocrine small cell cervical carcinoma: a multivariate analysis. Cancer 2003; 97: 568 – 74. 53. Chang TC, et al. Prognostic factors in surgically treated small cell cervical carcinoma followed by adjuvant chemotherapy. Cancer 1998; 83: 712 – 8. 54. Hoskins PJ, et al. Small-cell carcinoma of the cervix: fourteen years of experience at a single institution using a combined-modality regimen of involved-field irradiation and platinum-based combination chemotherapy. J Clin Oncol 2003; 21: 3495 – 501. 55. Albores-Saavedra J, et al. Carcinoid tumors of the uterine cervix. Cancer 1976; 38: 2328 – 42. 56. Seidel RJ, Steinfeld A. Carcinoid of the cervix: natural history and implications for therapy. Gynecol Oncol 1988; 30: 114 – 9. 57. Wright JD, et al. Cervical sarcomas: an analysis of incidence and outcome. Gynecol Oncol 2005; 99: 348 – 51. 58. Clement PB, et al. Malignant mullerian mixed tumors of the uterine cervix: a report of nine cases of a neoplasm with morphology often different from its counterpart in the corpus. Int J Gynecol Pathol 1998; 17: 211 – 22. 59. Bell SW, Kempson RL, Hendrickson MR. Problematic uterine smooth muscle neoplasms. A clinicopathologic study of 213 cases. Am J Surg Pathol 1994; 18: 535 – 8. 60. Ramos P, et al. Mullerian adenosarcoma of the cervix with heterologous elements: report of a case and review of the literature. Gynecol Oncol 2002; 84: 161 – 6. 61. Toyoshima M, et al. Epithelioid leiomyosarcoma of the uterine cervix: a case report and review of the literature. Gynecol Oncol 2005; 97: 957 – 60. 62. Brand E, et al. Rhabdomyosarcoma of the uterine cervix. Sarcoma botryoides. Cancer 1987; 60: 1552 – 60. 63. Behtash N, et al. Embryonal rhabdomyosarcoma of the uterine cervix: case report and review of the literature. Gynecol Oncol 2003; 91: 452 – 5. 64. Gruessner SE, et al. Management of stage I cervical sarcoma botryoides in childhood and adolescence. Eur J Pediatr 2004; 163: 452 – 6. 65. Bernak KL, et al. Embryonal rhabdomyosarcoma (sarcoma botryoides) of the cervix presenting as a cervical polyp treated with fertility-sparing surgery and adjuvant chemotherapy. Gynecol Oncol 2004; 95: 243 – 6. 66. Pathak B, et al. Granulocytic sarcoma presenting as tumors of the cervix. Gynecol Oncol 2005; 98: 493 – 7. 67. Cantuaria G, et al. Primary malignant melanoma of the uterine cervix: case report and review of the literature. Gynecol Oncol 1999; 75: 170 – 4. 68. Dursun P, et al. Primary cervical lymphoma: report of two cases and review of the literature. Gynecol Oncol 2005; 98: 484 – 9. 69. Chan JK, et al. Clinicopathologic features of six cases primary cervical lymphoma. Am J Obstet Gynecol 2005; 193: 866 – 72. 70. Snijders-Keilholz A, et al. Primitive neuroectodermal tumor of the cervix uteri: a case report – changing concepts of therapy. Gynecol Oncol 2005; 98: 516 – 9. 71. Roopnarinesingh R, Igoe S, Gillan JE. Choriocarcinoma presenting as a primary lesion of the cervix. Ir Med J 2004; 97: 147 – 8.
72. Lim SC, et al. Mature teratoma of the uterine cervix with lymphoid hyperplasia. Pathol Int 2003; 53: 327 – 31. 73. Khoor A, et al. Mature teratoma of the uterine cervix with pulmonary differentiation. Arch Pathol Lab Med 1995; 119: 848 – 50. 74. Cortes J, et al. Immature teratoma primary of the uterine cervix. First case report. Eur J Gynaecol Oncol 1990; 11: 37 – 42. 75. Yadav K, et al. Endodermal sinus tumor of cervix – case report. Indian J Cancer 1996; 33: 43 – 5. 76. Maesta I, et al. Primary non-gestational choriocarcinoma of the uterine cervix: a case report. Gynecol Oncol 2005; 98: 146 – 50. 77. Diakomanolis E, et al. Laser CO2 conization: a safe mode of treating conservatively microinvasive carcinoma of the uterine cervix. Eur J Obstet Gynecol Reprod Biol 2004; 113: 229 – 33. 78. Dargent D, et al. Laparoscopic vaginal radical trachelectomy. A treatment to preserve the fertility of cervical carcinoma patients. Cancer 2000; 88: 1877 – 82. 79. Covens A, et al. Is radical trachelectomy a safe alternative to radical hysterectomy for patients with stage IA-B carcinoma of the cervix? Cancer 1999; 86: 2273 – 9. 80. Roy M, Plante M. Radical vaginal trachelectomy for invasive cervical cancer. J Gynecol Obstet Biol Reprod (Paris) 2000; 29: 279 – 81. 81. Shepherd JH, Mould T, Oram DH. Radical trachelectomy in early stage carcinoma of the cervix: outcome as judged by recurrence and fertility rates. Br J Obstet Gynaecol 2001; 108: 882 – 5. 82. Bernardini M, et al. Pregnancy outcomes in patients after radical trachelectomy. Am J Obstet Gynecol 2003; 189: 1378 – 82. 83. Plante M, et al. Vaginal radical trachelectomy: an oncologically safe fertility-preserving surgery. An updated series of 72 cases and review of the literature. Gynecol Oncol 2004; 94: 614 – 23. 84. Bali A, et al. Central pelvic recurrence 7 years after radical vaginal trachelectomy. Gynecol Oncol 2005; 96: 854 – 6. 85. Ungar L, et al. Abdominal radical trachelectomy: A fertility-preserving option for women with early cervical cancer. Br J Obstet Gynaecol 2005; 112: 366 – 9. 86. Hellstrom AC, Sigurjonson T, Pettersson F. Carcinoma of the cervical stump. The radiumhemmet series 1959 – 1987. Treatment and prognosis. Acta Obstet Gynecol Scand 2001; 80: 152 – 7. 87. Leath CA III, et al. The role of radical parametrectomy in the treatment of occult cervical carcinoma after extrafascial hysterectomy. Gynecol Oncol 2004; 92: 215 – 9. 88. Munstedt K, et al. Consequences of inadvertent suboptimal primary surgery in carcinoma of the uterine cervix. Gynecol Oncol 2004; 94: 515 – 20. 89. Leath CA III, et al. The impact of aborted radical hysterectomy in patients with cervical carcinoma. Gynecol Oncol 2004; 95: 204 – 7. 90. Lomalisa P, Smith T, Guidozzi F. Human immunodeficiency virus infection and invasive cervical cancer in South Africa. Gynecol Oncol 2000; 77: 460 – 3. 91. Ozsaran AA, et al. Evaluation of the risk of cervical intraepithelial neoplasia and huyman papillomavirus infection in renal transplant patients receiving immunosuppressive therapy. Eur J Gynaecol Oncol 1999; 20: 127 – 30. 92. Jonas S, et al. De novo malignancies after liver transplantation using tacrolimus-based protocols or cyclosporine-based quadruple immunosuppression with an interleukin-2 receptor antibody or antithymocyte globulin. Cancer 1997; 80: 1141 – 50. 93. Hacker NF, et al. Carcinoma of the cervix associated with pregnancy. Obstet Gynecol 1982; 59: 735 – 46. 94. Duggan B, et al. Cervical cancer in pregnancy: reporting on planned delay in therapy. Obstet Gynecol 1993; 82: 598 – 602. 95. Monk BJ, Montz FJ. Invasive cervical cancer complicating intrauterine pregnancy: treatment with radical hysterectomy. Obstet Gynecol 1992; 80: 199 – 203. 96. Sood AK, et al. Radiotherapeutic management of cervical carcinoma that complicates pregnancy. Cancer 1997; 80: 1073 – 8. 97. Tewari K, et al. Neoadjuvant chemotherapy in the treatment of locally advanced cervical carcinoma in pregnancy: a report of two cases and review of issues specific to the management of cervical carcinoma in pregnancy including planned delay of therapy. Cancer 1998; 82: 1529 – 34.
Section 7 : Gynecological Cancers
46
Tumors of the Vulva and Vagina
Jonathan E. Tammela, Wainwright Jaggernauth, Paulette Mhawech-Fauceglia and Shashikant B. Lele
INTRODUCTION Malignant neoplasms of the vulva and vagina are uncommon; almost 600 cases are projected for 2005 in the United States.1 Together they represent approximately only 7% of all gynecologic oncology cases. For both sites, squamous cell carcinoma represents the most common histology.2 In this chapter, we discuss current information regarding these uncommon tumors, with an emphasis on diagnosis and management, but with the caveat that we have incorporated approaches based on opinion and clinical experience where clinical trial data are not available. Rare histological types of these tumors will also be discussed.
TUMORS OF THE VULVA More than 3800 new cases of malignant vulvar neoplasms are projected for diagnosis in 2005 in the United States. The most common histology is represented by squamous cell carcinomas. Approximately 90% of cases are squamous cell tumors and the second most common histology is melanoma.
Squamous Cell Carcinoma These neoplasms occur most often in the elderly; the peak incidence occurs during the seventh and eight decades of life, although cases can be diagnosed in adolescents.3 Early observational studies suggest an association between hypertension, diabetes mellitus, and obesity with vulvar cancer.3 However, more recent studies have failed to confirm this association.4 Therefore these factors may simply reflect coexisting medical conditions associated with age, rather than bearing specific histogenetic implications. The etiology of these neoplasms is still unclear. Several infectious agents have been proposed as possible etiologic factors. Among these, human papillomavirus (HPV) has been studied extensively. Epidemiological studies have suggested an association between vulvar condyloma and subsequent development of vulvar cancer.4 Several seroepidemiological studies have shown an increased risk for the development of vulvar cancer in HPV-seropositive patients.5 – 8 In addition,
HPV DNA has been identified in both intraepithelial and invasive lesions.7,9,10 Although HPV DNA has been identified in the majority (70–80%) of intraepithelial lesions, it has not been detected in as many invasive cancers (10–50%). HPV type 16 is the most common, although types 6, 18, and 33 have also been identified.11 It has been proposed that vulvar carcinoma has two variants: one associated with HPV and one not associated with this infection.12,13 The more common type of vulvar cancer is seen in older women, unrelated to smoking or HPV, often associated with adjacent dystrophic lesions and more aggressive in nature compared with the other variant of vulvar cancer more common in younger women, associated with HPV infection and smoking and usually of a less aggressive nature. Additional studies suggest other etiological factors, such as infection with herpes simplex virus 2 (HSV2), condyloma, and smoking.5,7,14 It appears that the relative risk for developing vulvar cancer increases when more than one factor is present. Other possible risk factors include chronic granulomas, syphilis, vulvar dystrophy, and lichen sclerosus.4 However, most studies that reported these associations are older and retrospective. Hart et al. were unable to identify a transition from lichen sclerosis to vulvar cancer in their large pathologic review.15 It is now generally accepted that there is low risk (1–2%) of progression to vulvar cancer from lichen sclerosis.16 There may also be a role for nutritional and dietary factors.17 Clinical Presentation
Data collected from four large series show that pruritus and the presence of a mass are the most common presenting symptoms.4,18 – 20 Other frequently described symptoms are bleeding and pain (see Table 1). Less commonly, patients experience dysuria, discharge, or ulceration. Most patients have more than one symptom. Unfortunately, for diverse reasons, frequently there is a delay between the onset of symptoms and the diagnosis, often because of a significant interval between the appearance of symptoms and the seeking of medical attention or a delay secondary to medical management for an unbiopsied presumed benign vulvar lesion.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
522
GYNECOLOGICAL CANCERS
Table 1 Symptoms at presentation in patients with vulvar squamous cell carcinoma.
Symptom
Frequency (%)
Pruritus Pain Mass Bleeding Dysuria
36 23 36 26 7
In one series 18% of patients had experienced symptoms for over 6 months.19 More significant is the delay in medical diagnosis. In this same series the mean time for diagnosis was 3.9 months. Vulvar carcinomas are usually unifocal. The lesion can be ulcerative, infiltrative, or exophytic. Therefore it can resemble several benign lesions such as condyloma accuminata, vulvar dystrophy, and Bartholin’s duct cyst. For these reasons it is recommended that patients who have a vulvar lesion should undergo colposcopy and/or directed biopsy. As other anogenital neoplasms can occur simultaneously, it is important to thoroughly examine the vagina, cervix, and anus, in addition to the vulva. Staging and Prognosis
Vulvar cancer is staged surgically. The International Federation of Gynecology and Obstetrics (FIGO) and the American Joint Committee on Cancer (AJCC) staging systems are commonly used.21 – 23 (see Table 2). These staging systems incorporate the major prognostic factors in vulvar cancer: tumor size, depth of invasion, nodal involvement, and distant metastasis. Nodal involvement is probably the most important prognostic factor in patients with operable vulvar carcinoma.24 – 28 However, the inadequacy of clinical assessment in detecting inguinal nodes has been demonstrated. The incidence Table 2 FIGO (1995) Staging of vulvar carcinoma.
Stage Clinical findings Stage 0 Carcinoma in situ: intraepithelial carcinoma Stage 1 Tumor confined to the vulva or perineum: 2 cm or less in greatest dimension; no nodal metastasis Stage IA: stromal invasion ≤ 1.0 mm Stage IB: stromal invasion >1.0 mm Stage II Tumor confined to the vulva or perineum; more than 2 cm in greatest dimension; no nodal metastasis Stage III Tumor of any size with adjacent spread to the urethra, vagina, or the anus, or with unilateral regional lymph node metastasis Stage IVA Tumor invades upper urethra, bladder mucosa, rectal mucosa, pelvic bone, or bilateral regional node metastases Stage IVB Any distant metastasis, including pelvic lymph nodes
of inguinal metastasis is correlated with the clinical stage; approximately 10% of patients with stage I disease have involved nodes, 26% of stage II, 56% of stage III, and 75% of stage IV.24,29 The importance of node status is made apparent by correlating it with survival: stages I and II patients (negative groin nodes) have greater than 90% survival compared with 50% in stage III (positive unilateral groin nodes) and lower still in stage IV (positive bilateral groin nodes) patients. These findings provided the rationale for the surgical based staging system now incorporated. Several other factors, such as age, tumor size, histologic grade, lymphovascular invasion, depth of invasion, and pattern of invasion are correlated with prognosis. However, many of these factors do not have independent value in a multivariate analysis and they predict only risk of nodal involvement. On the basis of the factors that have independent prognostic value (lymph node status: lateralism and number, and tumor size) the Gynecologic Oncology Group (GOG) has developed a surgicopathological staging system.25 In addition to the prognostic factors described above, innovative molecular factors have been recently described. Their precise role still remains to be clarified. However, it is important to identify additional prognostic factors that can aid in individualizing treatment. It has been reported that high S-phase fraction and presence of Her-2/neu overexpression are more common in node-positive patients.30 However, it is still unclear if these markers provide additional prognostic information. A number of studies have reported the prognostic role of Ki67 expression, another marker of cell proliferation. The results have been contradictory, with some studies suggesting a worse prognosis, or failed to find any independent additional prognostic information, or found that distribution of expression, not percentage of expression, was of prognostic significance.31 – 34 Similarly, it appears that DNA ploidy does not provide additional prognostic information.35 The p53 tumor suppressor gene plays a key role in cellular apoptosis. Mutation of this gene has been associated with poor prognosis in several solid tumors. Overexpression of p53 has been seen in approximately 50% of vulvar cancers. Some reports suggest that p53 overexpression is a poor prognostic factor.31,36 – 38 However, other studies have failed to confirm this finding.39 Recent reports have described an inverse relationship between increased expression of epidermal growth factor, microvessel density, or vascular endothelial growth factor and survival.40 – 43 The expression of CD44v3, an adhesion molecule, was associated with reduced overall survival.44 Similarly, the expression of vimentin and some cytokeratins may preclude a worse prognosis.45,46 Treatment
The traditional management of vulvar carcinoma involves aggressive surgical resection (radical vulvectomy with bilateral inguinofemoral lymphadenectomy). This treatment provides excellent local control and overall survival.47,48 However, the management of this neoplasm has evolved and strategies to individualize treatment have been proposed. On
TUMORS OF THE VULVA AND VAGINA
the basis of the FIGO staging system patients can be categorized into groups to plan their management: microinvasive tumors, stages I –II cancers, stages III –IV cancers, and nodepositive patients.
Microinvasive Tumors Patients with stage IA tumors have a negligible risk of nodal metastasis. Therefore this situation can be treated with local excision only. A 1-cm surgical margin appears to be adequate.49 Microinvasive tumors are more common in young women with multifocal preinvasive disease, and are frequently associated with HPV infections. Therefore the entire vulva and lower genital tract should be evaluated with colposcopy prior to surgical resection. Postoperative surveillance is also necessary.
Stages I and II Tumors The traditional management for these patients is a radical vulvectomy with bilateral inguinofemoral lymphadenopathy. The results of this approach are excellent, with local control and overall survival exceeding 90%.50,51 It appears that the margin of clearance of tumor is an important prognostic factor. A margin of at least 1.0 cm should be obtained.52,53 However, the morbidity is substantial: alterations in cosmetic and sexual function, local inguinal complications, and leg edema.19,54 In 1979, DiSaia et al. was the first to suggest that a superficial groin node dissection would be an adequate assessment of the deeper lymph nodes.55 Since then, many have advocated the same, stressing also a potential reduction in morbidity. Groin relapse rate with a full inguinofemoral dissection, as described by GOG protocol 36 and 37, and in a study by Burger et al., has been shown to be less than 1%.56 – 58 The groin relapse rate has now been shown to be consistently higher at 7–17% when only a superficial dissection is employed.59 – 61 Most impressive is the failure rate of almost 17% presented in the retrospective review of 227 patients treated at MD Anderson Cancer Center.61 Of these 227 patients, 107 patients had a superficial groin node dissection with 96% of patients with negative pathology: 18 of these patients had a recurrence. Worse yet, groin recurrence are most often fatal.59 This is in line with the finding of Gaarenstroom et al. where 5 of 93 groin dissections contained deep lymph node metastases without the presence of superficial node metastases.62 Despite the potential increase in lymphocyst formation and other morbidity, we advocate the more effective complete inguinofemoral lymphadenectomy. In the same manner, new surgical methods are being evaluated to minimize morbidity. For example, intraoperative lymphatic mapping is a method that has been developed mostly in cases of melanoma and breast cancer. Early data suggest that this may be a useful technique to identify a sentinel node in patients with vulvar cancer.63,64 Although these techniques are promising, their role has not yet been determined. Both the GOG and the European Organisation for Research and Treatment of Cancer (EORTC) are currently evaluating this technique to determine its ultimate feasibility.
523
Stages III and IV Tumors Radical vulvectomy with inguinofemoral lymph node dissection remains the standard of therapy for these patients. Close resection margins constitute a poor prognostic factor.52 Since resection margins have a major prognostic role and since this surgery often results in substantial disfigurement with high morbidity, preoperative therapy has been evaluated to improve operability. Initial studies evaluated the role of radiation therapy as neoadjuvant therapy.65 More recent studies have explored the role of chemoradiation, but it is not yet clear whether this technique offers a true survival benefit, although the morbidity of combined modality treatment is often greater than for radiotherapy alone.66 – 72 Nonetheless, this remains a viable option for those women with large vulvar lesions that otherwise may have required extensive surgery up front to be treated effectively. Another possibility, although not yet well explored, is the idea of neoadjuvant chemotherapy with the same goal in mind as radiation: decrease extent and morbidity of the surgery. Geisler et al. presented data using cisplatin and 5-fluorouracil (5-FU) in this setting.73 They showed a 100% clinical response rate and a pathologic complete response rate of 44% in women with large bulky vulvar lesions. This may become another treatment alternative once further study confirms their findings. Adjuvant Radiation
Adjuvant radiation is recommended for tumors with lymphovascular invasion, margins less than 8 mm, thickness greater than 5 mm, or involved lymph nodes.28,52 The decision to proceed with radiation in the presence of unfavorable nodal factors is more complex. High-risk factors include: residual disease, two or more positive lymph nodes, capsular extension, or residual macroscopic disease. The data supporting omitting radiation when only one node is positive is largely based on a study from 1986 by Homesley et al.58 That study was originally designed to compare radiation therapy with pelvic lymphadenectomy in patients who underwent an inguinofemoral lymphadenectomy. Survival rate was not different in the group that had one positive groin node whether they went on to have radiation or further pelvic lymphadenectomy. In contrast, survival was significantly improved in those patients with two or more positive groin lymph nodes that underwent radiation compared with those patients who went on to further pelvic lymphadenectomy. The implication was that radiation did not add any benefit and therefore inguinofemoral lymphadenectomy was sufficient therapy in the subset of patients with only one involved node (assuming that the current exclusion of pelvic lymphadenectomy would not change this conclusion). Two other studies went on to show that survival rate was similar and favorable in patients with one positive lymph node compared with negative nodes after a full inguinofemoral lymphadenectomy, giving further validation for omitting radiation in this group.74,75 While this remains the accepted standard, one might exercise caution extrapolating this data to include patients who receive only a superficial lymphadenectomy. As discussed above, these patients who have negative nodes have a high groin failure rate likely because of unresected, deeper positive groin
524
GYNECOLOGICAL CANCERS
nodes. Presumably, this failure rate might even be higher if the same group of patients with a superficial groin dissection and one positive node were also not treated with radiation therapy. Denying adjuvant radiation to these patients with one positive pelvic lymph node after a superficial dissection based on data obtained from a full inguinal pelvic lymphadenectomy may lead to undertreatment. However, there is no direct evidence to support or refute this at the time. Origoni et al. presented their findings evaluating prognostic factors for pathological patterns in vulvar cancer in 1992.76 They were able to show in this elegant study that even one node with extracapsular invasion or larger than 5 mm was significantly associated with worse survival rate and therefore would probably benefit from adjuvant radiation.
Recurrent Vulvar Cancer The interval between initial therapy and local relapse has a strong prognostic effect on survival. Treatment with radiation and chemotherapy can offer long-term control to 50% of patients when this interval exceeds 2 years.69,72,77 Surgical resection of local recurrences can also provide long-term control to more than 50% of patients.52,53,78,79 One benefit of treating bulky localized vulvar cancer with neoadjuvant chemotherapy is that if these patients do recur they are still left with two good options for retreatment: surgery and radiation. The reverse situation of neoadjuvant radiation before surgery presents two less desirable options: surgery to a radiated field and/or chemotherapy, which inherently has a low response in this setting. Both, recurrence in the groin and distant recurrence may be treated with chemotherapy, but are almost always fatal. Chemotherapy The role of chemotherapy in vulvar cancer remains unclear. Historically, these tumors have been considered chemoresistant. However, more recently, large EORTC trials using bleomycin, methotrexate, and CCNU have revealed encouraging response rates of 64 and 56% in patients with advanced or recurrent disease.80,81 Several drugs have been evaluated as single agents, including anthracyclines, antimetabolites, platinum complexes, and alkylating agents. An encouraging response rate was initially reported with doxorubicin, but the numbers of treated patients was too small to permit definitive assessment. Bleomycin also appears to have activity in this disease.77 Other agents that have shown disappointing activity when used as single agents are cisplatin, mitoxantrone, and etoposide. Although no clinical trials have been published, preclinical data suggest that paclitaxel may exhibit activity in this disease, as does the combination of cisplatin and paclitaxel in vitro.82,83 Additionally, topotecan has shown superiority in in vitro studies over other often-used single agents.84 Multiagent chemotherapy regimens appear to produce a higher response rate,85,86 although it is not clear whether this translates into survival benefit. Chemotherapy in vulvar cancer remains an active area of investigation.
Melanoma of the Vulva Melanoma constitutes the second most common malignant tumor of the vulva. This tumor frequently affects perimenopausal women, with most cases occurring between the
sixth and seventh decade of life. Vulvar melanomas are classified into three categories: superficial spreading melanoma, nodular melanoma, and acral lentiginous melanoma.87 – 89 Acral lentiginous and nodular melanoma are the most common types. Clinical Presentation
Most melanomas originate in the labia minora and clitoris and are elevated, pigmented, and frequently ulcerated growths with an irregular border. Presenting symptoms include bleeding, discharge, pain, dysuria, while signs include mass or lump. Differential diagnosis includes Paget’s disease, benign nevus, dysplastic nevus, seborrheic keratosis, and basal cell carcinoma. The so-called amelanotic melanoma does not express melanin, and may be confused with squamous cell carcinomas. Diagnosis should be made with Keyes punch biopsy or simple excision with care to excise full thickness of the lesion. Prognostic Factors
Overall, the prognosis of vulvar melanoma is poor;90 5year survival as low as 10% has been described,91 although most series report a 5-year survival of approximately 50%. The level of invasion and tumor thickness are the most important prognostic factors.92 Modifications to the classical definitions of Clark and Breslow have been suggested for vulvar melanomas.93 In addition, involvement of lymph nodes is an independent poor prognostic feature.94,95 Both the Clark and Breslow staging systems are generally believed to be superior to clinical FIGO staging (see Table 3). Other prognostic risk factors are ulceration, inflammatory reaction, high mitotic rate, and age.95 – 97 More recently it has been described that DNA ploidy also is a prognostic factor.98,99 Table 3 Clark, Breslow, and Chung staging systems.
Clark’s levels of invasion of cutaneous melanoma of nongenital origin Level I In situ lesions; neoplastic cells confined to the epithelium Level II Lesion penetrates the basement membrane and extends into the loose papillary dermis Level III Melanoma invading and usually filling the papillary dermis and involving the reticular dermis, accumulating at the interface Level IV Invasion of the deep reticular dermis Level V Melanoma invading the subcutaneous adipose tissue Breslow depth of invasion of cutaneous melanoma 1. 0.75 mm or less 2. 0.76 to 1.50 mm 3. 1.51 to 2.26 mm 4. 2.26 to 3.0 mm 5. Greater than 3.0 mm Chung’s levels of involvement in melanoma of the vulva Level I Tumor confined to the epithelium Level II Lesion penetrates the basement membrane and extends into the dermis or lamina propria to 1 mm or less from the granular layer or its estimated position in the epidermis or from the outermost epithelial layer Level III Melanoma penetrating between 1 and 2 mm into subepithelial tissue Level IV Invasion beyond 2 mm, but not into underlying fat Level V Melanoma invading the subcutaneous adipose tissue
TUMORS OF THE VULVA AND VAGINA
Treatment
Vulvar melanoma is an uncommon tumor, and thus no randomized clinical trials have been conducted regarding optimal clinical care. However, the GOG conducted a prospective clinicopathologic study of vulvar melanoma and concluded that the biologic natural history of vulvar melanoma is similar to that of nonvulvar cutaneous melanoma.100 Therefore, it is reasonable to assume that some of the therapeutic principles in the management of cutaneous melanoma can be applied to the treatment of vulvar melanoma. Surgery represents the cornerstone of treatment of melanoma. The extension of the surgical procedure is determined by the tumor thickness. Randomized clinical trials have been conducted in cutaneous melanomas that establish the recommended surgical margins. A large randomized study of patients with tumors of less than 2 mm was conducted. For a tumor of less than 1 mm, a margin of 1 cm is appropriate.101 Similar findings have been reported for tumors of less than 0.76 mm.102 A second study evaluated the surgical management of melanomas of intermediate thickness, defined as 1–4 mm. This study concluded that a 2 cm margin is appropriate.103 A more recent study shows similar results for melanomas of 1–2 mm.104 No study has addressed the appropriate margins for thicker melanomas. As described above, tumor thickness represents a major prognostic factor. Tumor thickness is a useful prognostic factor in predicting the risk of regional lymph node involvement and distant metastasis.105 Thin tumors (less than 1 mm) rarely involve regional lymph nodes and have an excellent prognosis. Intermediate thickness melanomas (1–4 mm) have up to a 60% risk of involving regional lymph nodes and have a relatively low risk of harboring distant metastases (less than 20%). On the other hand, thick melanomas (more than 4 mm) have a high risk (60% or more) of hiding regional or distant metastasis. The role of lymph node dissection in patients with clinically negative nodes is controversial. Several randomized clinical studies have been published. These studies have shown no improvement in the survival rates in those patients treated with a lymph node dissection.106 – 109 However, these studies have identified subgroups that might benefit from lymph node dissection. It appears that young patients (less than 60 years) with tumors of 1.1–2.0 mm might benefit from an elective lymph node dissection. The primary benefit to performing a lymphadenectomy appears to be prognostic in nature. The surgical management of vulvar melanoma is complicated by the lymphatic drainage of the vulva, which would require a bilateral lymph node dissection. Therefore, the role of intraoperative lymphatic mapping and sentinel node biopsy is being addressed. However, at this time, this approach cannot be considered the standard of care for patients with vulvar melanoma. A randomized study reported an improvement in disease-free and overall survival for patients with melanoma treated with adjuvant α-2binterferon.110 The most significant benefit was seen in those patients with involved lymph nodes. Two additional studies showed similar results.111,112 The three studies used relatively high doses of interferon and the treatment was administered
525
for 1 year or more. On the other hand, two other additional studies showed no improvement in survival.113,114 These studies used either lower doses of interferon or a shorter duration of treatment. In conclusion, it appears that patients with intermediate thickness melanomas and those with positive involved lymph nodes benefit from adjuvant treatment with high-dose interferon for 1 year. Other forms of adjuvant therapy (vaccine, chemotherapy) have shown no improvement in survival. Treatment of Metastatic Disease
Systemic therapy in recurrent or metastatic melanoma is associated with low response rates. Surgical resection can be considered for isolated recurrences. Single-agent chemotherapies with documented activity (response rate approximately 20%) are dacarbazine, carmustine, cisplatin, and fotemustine. Combination chemotherapy may produce a higher response rate but with no definite improvement in survival. Interleukin-2 (IL-2) and high-dose α-interferon produce a response rate of approximately 16%. Current clinical trials are evaluating the combination of IL-2 and interferon with chemotherapy.
Basal Cell Carcinoma Basal cell carcinoma is a relatively uncommon tumor that represents approximately 2–4% of vulvar malignancies.115 – 117 Most tumors are diagnosed in postmenopausal women. Frequent symptoms are pruritus, a palpable nodule, ulcer, or area of pigmentation or hypopigmentation. The lesion may resemble a papilloma or a benign nevus; the differential diagnosis includes squamous cell carcinoma, Merkel cell carcinoma, and melanoma. Most tumors arise from the labia majora and typically measure less than 2 cm. Usually the symptoms have been present for several months, even years. These tumors rarely metastasize. The treatment of choice is wide excision. Elective lymph node dissection is not routinely indicated, except maybe in lesions larger than 4 cm.
Sarcomas of the Vulva Vulvar sarcomas are uncommon and represent only 1–2% of all vulvar malignancies. Multiple histological types have been described.118,119 Most patients present with a painful mass in the labium majora or Bartholin’s gland area.
Leiomyosarcoma Leiomyosarcoma is the most common sarcoma of the vulva.120,121 These tumors are frequently larger than 5 cm in diameter. Histologically they are composed of interlacing spindle-shaped cells; epithelioid cells can also be present. Diagnostic criteria are a mitotic count of 10 or more per 10 high-power fields, infiltrating border, or nuclear atypia with pleomorphism and size above 5 cm. Tumors that exhibit three or four of these features can be classified as leiomyosarcomas, while those with only one feature should be diagnosed as leiomyoma. Those tumors that exhibit two features can be classified as atypical leiomyomas or tumors of uncertain malignant potential. Some authors accept that
526
GYNECOLOGICAL CANCERS
the diagnosis of leiomyosarcoma can be made when the only criteria present is a high mitotic count (10 or more mitoses). Leiomyosarcomas require surgical treatment with a wide excision; less aggressive tumors can be treated with more conservative surgery. There is no indication for lymph node dissection. The role of adjuvant therapy is unclear. In advanced or recurrent disease the response rate to systemic chemotherapy is low.
Malignant Fibrous Histiocytoma This tumor originates from histiocytes with fibroblastic differentiation. It is the second most common sarcoma of the vulva. Histologically the tumor has a complex interlacing cellular growth pattern, there is marked pleomorphism, and multinucleated cells are common. This tumor usually infiltrates adjacent tissues. The treatment of choice is radical local excision. Ipsilateral lymph node dissection can be considered for large, deeply invading tumors. These tumors tend to have local and distant recurrences.122 It is not clear whether adjuvant radiation might improve local control.
Epithelioid Sarcoma This tumor may resemble squamous cell carcinoma. The cell of origin is undetermined. On histological examination the tumor is nodular and with areas of necrosis.123 The tumor cells have an epithelioid appearance, but metaplastic components (bone, cartilage) can be seen. On immunohistochemistry these tumors contain cytokeratin, which is useful in distinguishing them from other sarcomas. As with other pelvic sarcomas, the sheet anchor of therapy is surgical resection, where possible. Although local cytoreduction by radiotherapy is possible, many of these tumors are relatively radioresistant. The role of cytotoxic chemotherapy is unproven in this context.
Rare Types of Vulva Carcinoma Other histologic vulvar types include verrucous carcinoma, Kaposi’s sarcoma, metastatic disease (endometrium, ovary, cervix, breast, kidney, urethra, and lymphoma), lymphoma, and Merkel cell carcinoma. The evidence base in these tumors is weak and the basis for treatment decisions is often anecdotal. In general, these lesions are excised with wide margins. The role of lymphadenectomy is minimal.
TUMORS OF THE VAGINA Approximately 2000 new cases of primary malignant vaginal neoplasms are diagnosed annually in the United States. Overall more than 80% of vaginal neoplasms are metastatic tumors that involve the vagina either by direct extension or by lymphatic or hematogenous dissemination.124 Of the primary vaginal tumors, the most common histology is represented by squamous cell carcinomas. The second most common histology is represented by melanoma. Although uncommon, primary adenocarcinoma of the vagina is an important entity because of its presentation in young women and its etiological relationship to in utero exposure to diethylstilbestrol (DES).
Squamous Cell Carcinoma This neoplasm occurs mostly in postmenopausal women, with a mean age at presentation of 60–65 years.125 Some risk factors have been described: lower socioeconomic status, previous genital warts, chronic vaginal discharge or irritation, a previous abnormal Pap smear, and early hysterectomy.126 Although some authors have suggested that prior pelvic radiation may increase the risk of developing vaginal cancer127 others have failed to confirm this finding.128 Data suggests that HPV infection increases the risk of developing vaginal cancer.8,129 Clinical Presentation
The most common symptom is abnormal vaginal bleeding. Less frequent symptoms are vaginal discharge, a palpable mass, and pain. Most of the symptomatic patients have locally advanced disease. Approximately 20% of patients are asymptomatic and the diagnosis is made during investigation of an abnormal Pap smear. Most cancers appear in the upper third region of the vagina. These tumors are usually exophytic, although ulcerative and superficial lesions are also observed. Staging and Prognosis
Vaginal cancers are staged using the FIGO or American Joint Committee (AJC) staging systems. The clinical stage of disease represents the most important prognostic factor.130 The prognosis of patients in stage I is very good, with over 75% of patients being alive at 5 years. On the contrary, the survival in cases with stage IV is poor, with only 18% of patients alive at 5 years.131 Other prognostic factors that have been evaluated are age, grade of differentiation, keratinization, and extent of mucosal involvement.132,133 However, none of these factors provides additional prognostic information. Recent efforts have been directed to identifying the role of biological prognostic markers. Preliminary results suggested that overexpression of Her-2/neu was associated with worse prognosis.132 However, other authors have not confirmed this finding.134 In addition, detection of p53, retinoblastoma protein, and epidermal growth factor does not appear to provide prognostic information.135,136 Treatment
The treatment of vaginal cancer is established on the basis of clinical stage, tumor size, location, and antecedent of pelvic radiation (usually for prior cervical cancer). Patients with superficial stage I disease (tumor thickness less than 0.5 cm) can be treated with intracavitary radiation.137 For tumors located in the lower third of the vagina it is recommended that patients also receive external beam radiation to the pelvic ± inguinal lymph nodes.138 Surgical treatment, with wide local excision or total vaginectomy is also effective. Adjuvant radiation therapy is recommended for tumors with close or positive margins.135 Patients with stage I lesions greater than 0.5 cm and stage IIA are best treated with combined external beam and intracavitary radiation.138,139 Surgical treatment by means of a radical vaginectomy and pelvic lymphadenectomy provides a similar cure rate.139,140
TUMORS OF THE VULVA AND VAGINA
Therapy for patients with more advanced disease is frequently disappointing. The use of radiation therapy alone (external beam radiation and brachytherapy) can be associated with local recurrence in 30–50% of patients and 5-year survival of 18–31%. For selected patients, preoperative radiation followed by radical surgery can be attempted, and often achieves significant tumor cytoreduction.140,141 Chemotherapy appears to play only a minor role in the management of squamous cell vaginal cancer. Etoposide, mitoxantrone, and cisplatin are among the agents that have been studied in small phase II trials and have not shown meaningful activity. Adriamycin has reported activity in a small subset of patients.142 Some authors have reported their results with cisplatin-based combination chemotherapy. However, these are very small studies and therefore no definite conclusions can be made regarding the benefit of this approach. Finally, other authors have used combined treatment with chemotherapy and radiation. The most encouraging results were reported with the combination of 5-FU and mitomycin in combination with external beam radiation, where a response rate of 57% was observed.143 However, the design of this study does not allow the critical analysis of the specific effect of chemotherapy in this scenario.
Adenocarcinoma Primary vaginal adenocarcinoma is an uncommon neoplasm. It represents less than 5% of vaginal cancers and more frequently is a metastatic lesion. However, clear cell adenocarcinomas (CCA) represent an important subtype. These tumors occur more frequently in young women between 17 and 21 years of age.144 This tumor frequently is associated with maternal ingestion of DES during pregnancy. DES can produce a variety of effects in the exposed female offspring: vaginal adenosis, vaginal, cervical, or uterine malformations, infertility, and CCA. The risk of developing CCA is relatively small; the highest risk estimate is approximately 1 : 1000. Vaginal adenosis is present in one-third of women exposed to in utero DES.145 Only a small number of women with vaginal adenosis will develop CCA. The peak incidence of CCA secondary to DES exposure occurred in the late 1970s.146 Since then there has been a decrease in the incidence, but 20–30 cases are still diagnosed annually.147 In approximately one-third of cases of CCA there is no prior history of exposure to DES. Microsatellite instability has been identified in all cases of DES-associated tumors, while this event is observed in only 50% of tumors not associated with DES.148 CCA is diagnosed mostly in young women. However, recent reports suggest that non-CCA can also occur in older DESexposed patients.149 Some authors have identified a second peak age during the eighth decade of life.150 In addition, late recurrences of CCA have been reported.151 For this reason long-term surveillance of DES-exposed women may be warranted. Clinical Presentation
These lesions are frequently exophytic tumors, located primarily in the anterior or posterior vaginal wall. On some
527
occasions these tumors can be submucosal and are identified only by palpation. Rarely are these tumors identified on a screening cytology. Treatment
Patients with stage I –IIA disease can be treated with surgery or radiation.139,140 Surgical management involves a total radical vaginectomy and hysterectomy with lymph node dissection. A pelvic lymphadenectomy is indicated for tumors located in the upper vagina while an inguinal dissection is done for lower tumors. More recently it has been recommended that patients who desire future fertility can be treated with a more conservative approach. If the vaginal lesion is amenable to local resection the patient should undergo a laparoscopic or retroperitoneal pelvic lymph node dissection. If there is no evidence of lymph node involvement the vaginal lesion is excised. Patients receive postoperative local radiation. The 5- and 10-year survivals (92 and 88%, respectively) for patients with stage I disease are comparable with those of patients treated with more radical surgery.152 At least 20% of patients can achieve normal pregnancies following this conservative approach. Patients with more advanced disease are treated following the same principles as for squamous cell carcinoma.
Melanoma Vaginal melanoma is the second most common primary neoplasm of the vagina. However, it accounts for only less than 5% of all vaginal neoplasms.153,154 These tumors have been diagnosed from the third to the ninth decades, with the mean age of presentation occurring during the sixth decade. The tumor arises from vaginal melanocytes, which are present in 3% of adult females.155 Frequently, the patient presents with vaginal bleeding or discharge. Vaginal melanomas are more frequently found in the anterior wall and the distal third. The tumor can be polypoid, pedunculated, papillary, or fungating, and is usually brownish or black. Prognosis
Vaginal melanomas have a poor prognosis, the 5-year survival is less than 20%.156 – 158 Tumor depth represents the most important prognostic factor.157,159 However, some authors report that tumor size is a stronger factor.160 Treatment
Therapeutic modalities include local excision, radiation, and/or radical surgery. It appears that overall survival is similar with any modality. Patients treated with conservative resection experience a very high local recurrence. Radical surgery may decrease this risk. Radical surgery is frequently used, although some authors question this because of the low survival.160 Some authors have demonstrated a possible improvement in local control with combined radiotherapy and radical resection.161 Most patients will eventually develop distant metastases. It is reasonable to recommend adjuvant interferon to patients with thick tumors or lymph node involvement. The treatment of patients with metastatic
528
GYNECOLOGICAL CANCERS
disease is disappointing. Adjuvant treatment and management of metastatic disease can follow the same recommendations described above for vulvar melanomas.
Sarcoma Embryonal rhabdomyosarcoma is the most common malignant tumor of the vagina in the pediatric population. Most tumors are diagnosed before the age of 5 years160 – 162; however, adult cases have been reported.163 Patients usually present with a grape-like vaginal mass. Other common symptoms are vaginal bleeding and discharge. The tumor frequently extends to the bladder or rectum, and pelvic or inguinal lymph node involvement is common. These tumors may be confused with yolk sac tumors and therefore αfetoprotein (AFP) should be obtained to rule out this diagnosis. Prognosis
Patients are staged using the Intergroup Rhabdomyosarcoma Study (IRS) Grouping System.164 This staging system takes into account the tumor size, lymph node involvement, extension of resection, and presence of distant metastases. Current treatment recommendations include preoperative chemotherapy followed by surgical resection. Most patients experience a response to chemotherapy and therefore radical surgery is seldom necessary. Postoperative radiation is indicated in the event of involved margins or positive lymph nodes.
Other Sarcomas Leiomyosarcoma is the most common vaginal sarcoma in the adult patient. More rarely other sarcomas, such as fibrosarcoma and angiosarcoma, have been reported.165 – 167 The prognosis and treatment are similar to those of vulvar sarcomas, described above.
Hematologic Neoplasms Granulocytic sarcoma represents an infiltrative process seen in acute leukemias that has been reported to occur in the female genital tract.168 Most cases are identified during the initial presentation of the leukemia. The vagina can also be involved by lymphoma, usually of the diffuse large cell type.169,170 On occasions it is difficult to distinguish between these tumors. In those cases, immunohistochemical stains are useful: chloroacetate esterase and myeloperoxidase are commonly used in this scenario. These tumors should be managed with the systemic therapy appropriate for the histological subtype (see Chapter 49, Rare Lymphomas).
Endodermal Sinus Tumor Malignant germ cell tumors account for approximately 3% of pediatric tumors. Endodermal sinus tumor is the most common histology.171 Several cases of vaginal tumors have been reported.172 – 174 These tumors can resemble rhabdomyosarcomas histologically and in their presentation. However, a clear distinction can usually be achieved by the measurement of blood levels of AFP. Local therapy alone is associated with poor prognosis because of the risk of local relapse and systemic spread. Current treatment regimens are predicated on
systemic chemotherapy, usually comprising regimens more commonly applied to the management of testicular cancer, followed by surgical resection. With this therapy prognosis is satisfactory, with the majority of patients ultimately achieving cure.
SUMMARY Tumors of the vulva and vagina, although uncommon, are a protean collection of vastly different malignancies, sharing only similarities in the pattern of presentation and local spread. In general, careful clinical and pathological staging allows the determination of optimal therapeutic approaches, which have largely been dominated by surgical excision. In view of the rarity of many of the tumors discussed above, it is not possible to make definitive statements regarding the role of chemotherapy or of combined modality treatments. However, it is frequently helpful to apply principles of treatment garnered from presentations at other sites (e.g. of squamous cell carcinoma, germ cell tumors, soft tissue sarcomas, malignant melanoma) to the management of these tumors when they arise in the pelvis.
REFERENCES 1. Ries LAG, Kosary EM, Hankey CL (eds). SEER Cancer Statistics Review, 1975 – 2002. National Cancer Institute, 2005. 2. Sheperd J, et al. Carcinoma of the vulva. J Epidemiol Biostat 1998; 3: 103 – 10. 3. Franklin EW, Rutledge FD. Epidemiology of epidermoid carcinoma of the vulva. Obstet Gynecol 1972; 39: 165. 4. Brinton LA, et al. Case-control study of cancer of the vulva. Obstet Gynecol 1990; 75: 859. 5. Hildesheim A, et al. Human papillomavirus type 16 and risk of preinvasive and invasive vulvar cancer: results from a seroepidemiological case-control study. Obstet Gynecol 1997; 90: 748. 6. Madeleine MM, et al. Cofactors with human papillomavirus in a population-based study of vulvar cancer. J Natl Cancer Inst 1997; 89: 1516. 7. Bjorge T, et al. Prospective seroepidemiological study of role of human papillomavirus in non-cervical anogenital cancers. Br Med J 1997; 315: 646. 8. Downey GO, et al. Condylomatous carcinoma of the vulva with special reference to human papillomavirus DNA. Obstet Gynecol 1988; 72: 68. 9. Monk BJ, et al. Prognostic significance of human papillomavirus DNA in vulvar carcinoma. Obstet Gynecol 1995; 85: 709. 10. Iwasawa A, et al. Human papillomavirus in squamous cell carcinoma of the vulva by polymerase chain reaction. Obstet Gynecol 1997; 89: 81. 11. Kurman RJ, Trimble CL, Shah KV. Human papillomavirus and the pathogenesis of vulvar carcinoma. Curr Opin Obstet Gynecol 1992; 4: 582. 12. Hording U, et al. Vulvar squamous cell carcinoma and papillomaviruses: indications for two different etiologies. Gynecol Oncol 1994; 52: 241. 13. Trimble CL, Hildesheim A, Brinton LA. Heterogeneous etiology of squamous carcinoma of the vulva. Obstet Gynecol 1996; 87: 59. 14. Kaufman RH, et al. Herpes-virus induced antigens in squamous-cell carcinoma in situ of the vulva. N Engl J Med 1981; 305: 483. 15. Hart WR, Norris HJ, Helwig EB. Relation of lichen sclerosus et atrophicus of the vulva to development of carcinoma. Obstet Gynecol 1975; 45(4): 369 – 77. 16. Carlson JA, et al. Vulvar lichen sclerosus and squamous cell carcinoma: a cohort, case control, and investigational study with historical perspective; implications for chronic inflammation and
TUMORS OF THE VULVA AND VAGINA
17. 18. 19.
20. 21.
22. 23. 24.
25.
26.
27. 28.
29. 30.
31. 32.
33. 34.
35.
36.
37. 38.
39.
40.
41.
sclerosis in the development of neoplasia. Hum Pathol 1998; 29(9): 932 – 48. Parazzini F, et al. Selected food intake and risk of vulvar cancer. Cancer 1995; 76: 2291. Figge CD, Gaudenz R. Invasive carcinoma of the vulva. Am J Obstet Gynecol 1974; 39: 165. Rhodes CA, Cummins C, Shafi MI. The management of squamous cell vulvar cancer: a population based retrospective study of 411 cases. Br J Obstet Gynaecol 1988; 105: 200. Rosen C, Malmstrom H. Invasive cancer of the vulva. Gynecol Oncol 1997; 65: 213. Homesley HD, et al. 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 1991; 164: 997. Boyce J, et al. Prognostic factors in carcinoma of the vulva. Gynecol Oncol 1985; 20: 364. Creasman WT. New gynecologic cancer staging. Gynecol Oncol 1995; 58(2): 157 – 8. Binder SW, et al. Risk factors for the development of lymph node metastasis in vulvar squamous cell carcinoma. Gynecol Oncol 1990; 37: 9. Sedlis A, et al. Positive groin lymph nodes in superficial squamous cell vulvar cancer: a Gynecologic Oncology Group study. Am J Obstet Gynecol 1987; 156: 1159. Creasman WT, Phillips JL, Menck HR, The American College of Surgeons Commission on Cancer and the American Cancer Society. The national cancer data base report on early stage invasive vulvar carcinoma. Cancer 1997; 80: 505. Rutledge FN, et al. Prognostic indicators for invasive carcinoma of the vulva. Gynecol Oncol 1991; 42: 239. Homesley HD, et al. Prognostic factors for groin node metastasis in squamous cell carcinoma of the vulva: a Gynecologic Oncology Group study. Gynecol Oncol 1993; 49: 279. Hacker NF, et al. Management of regional lymph nodes and their prognostic influence in vulvar cancer. Obstet Gynecol 1983; 61: 408. Gordinier ME, et al. S-phase fraction, p53, and HER-2/neu status as predictors of nodal metastasis in early vulvar cancer. Gynecol Oncol 1997; 67: 200. Marchetti M, et al. Ki-67 expression in vulvar carcinoma: preliminary results. Eur J Gynecol Oncol 1996; 17: 361. Emanules AG, et al. Quantitation of proliferation-associated markers Ag-NOR and Ki-67 does not contribute to the prediction of lymph node metastases in squamous cell carcinoma of the vulva. Hum Pathol 1996; 27: 807. Hantschmann P, et al. Tumor proliferation in squamous cell carcinoma of the vulva. Int J Gynecol Pathol 2000; 19(4): 361 – 8. Modesitt SC, et al. Expression of Ki-67 in vulvar carcinoma and vulvar intraepithelial neoplasia III: correlation with clinical prognostic factors. Gynecol Oncol 2000; 76(1): 51 – 5. Drew PA, et al. Prognostic factors in carcinoma of the vulva: a clinicopathologic and DNA flow cytometric study. Int J Gynecol Pathol 1996; 15: 235. Sliutz G, et al. Detection of p53 point mutations in primary human vulvar cancer by PCR and temperature gradient gel electrophoresis. Gynecol Oncol 1997; 64: 93. Kohlberger P, et al. Prognostic value of immunohistochemically detected p53 expression in vulvar carcinoma. Cancer 1995; 76: 1786. Salmaso R, et al. Prognostic value of protein p53 and ki-67 in invasive vulvar squamous cell carcinoma. Eur J Gynaecol Oncol 2000; 21(5): 479 – 83. Kagie MJ, et al. P53 protein overexpression, a frequent observation in squamous cell carcinoma of the vulva and in various synchronous vulvar epithelia, has no value as a prognostic parameter. Int J Gynecol Pathol 1997; 16: 124. Johnson GA, et al. Epidermal growth factor receptor in vulvar malignancies and its relationship to metastasis and patient survival. Gynecol Oncol 1997; 65: 425. Obermair A, et al. Influence of microvessel density and vascular permeability factor/vascular endothelial growth factor expression on prognosis in vulvar cancer. Gynecol Oncol 1996; 63: 204.
529
42. Hefler LA, et al. The prognostic value of immunohistochemically detected CD44v3 and CD44v6 expression in patients with surgically staged vulvar carcinoma: a multicenter study. Cancer 2002; 94(1): 125 – 30. 43. Tempfer C, et al. CD44v3 and v6 variant isoform expression correlates with poor prognosis in early-stage vulvar cancer. Br J Cancer 1998; 78(8): 1091 – 4. 44. Tempfer C, et al. Prognostic value of immunohistochemically detected CD44 expression in patients with carcinoma of the vulva. Cancer 1996; 78: 273. 45. Ansink A, et al. Cytokeratin subtypes and involucrin in squamous cell carcinoma of the vulva: an immunohistochemical study of 41 cases. Cancer 1995; 76: 638. 46. Weikel W, et al. Cytokeratin and vimentin expression in primary and recurrent carcinoma of the vulva: correlations with prognostic factors and the course of disease. Int J Gynecol Pathol 1996; 15: 326. 47. Way S. Carcinoma of the vulva. Am J Obstet Gynecol 1960; 79: 692. 48. Taussig FJ. Cancer of the vulva: an analysis of 155 cases. Am J Obstet Gynecol 1940; 40: 764. 49. Heaps JM, et al. Surgical-pathologic variables predictive of local recurrence in squamous cell carcinoma of the vulva. Gynecol Oncol 1990; 38: 309. 50. Morley GW. Infiltrative carcinoma of the vulva: results of surgical treatment. Am J Obstet Gynecol 1976; 124: 874. 51. Podratz KC, et al. Carcinoma of the vulva: analysis of treatment and survival. Obstet Gynecol 1983; 61: 63. 52. Thomas GM, et al. Changing concepts in the management of vulvar cancer. Gynecol Oncol 1991; 42: 9. 53. Hacker NF, Van der Velden J. Conservative management of early vulvar cancer. Cancer 1993; 71: S1673. 54. Rutledge F, Smith JP, Franklin EW. Carcinoma of the vulva. Am J Obstet Gynecol 1970; 106: 1117. 55. DiSaia PJ, Creasman WT, Rich WM. An alternate approach to early cancer of the vulva. Am J Obstet Gynecol 1979; 133(7): 825 – 32. 56. Sedlis A et al., A Gynecologic Oncology Group Study. Positive groin lymph nodes in superficial squamous cell vulvar cancer. Am J Obstet Gynecol 1987; 156(5): 1159 – 64. 57. Homesley HD, et al. Radiation therapy versus pelvic node resection for carcinoma of the vulva with positive groin nodes. Obstet Gynecol 1986; 68(6): 733 – 40. 58. Homesley HD, et al. 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 1991; 164(4): 997 – 1003; discussion 1003 – 4. 59. Gordinier ME, et al. Groin recurrence in patients with vulvar cancer with negative nodes on superficial inguinal lymphadenectomy. Gynecol Oncol 2003; 90(3): 625 – 8. 60. Stehman FB, et al. 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 1992; 79(4): 490 – 7. 61. Katz A, et al. The role of radiation therapy in preventing regional recurrences of invasive squamous cell carcinoma of the vulva. Int J Radiat Oncol Biol Phys 2003; 57(2): 409 – 18. 62. Gaarenstroom KN, et al. Postoperative complications after vulvectomy and inguinofemoral lymphadenectomy using separate groin incisions. Int J Gynecol Cancer 2003; 13(4): 522 – 7. 63. Decesare SL, et al. A pilot study utilizing intraoperative lymphoscintigraphy for identification of the sentinel lymph nodes in vulvar cancer. Gynecol Oncol 1997; 66: 425. 64. Levenback C, et al. Potential applications of intraoperative lymphatic mapping in vulvar cancer. Gynecol Oncol 1995; 59: 216. 65. Boronow RC, et al. Combined therapy as an alternative to exenteration for locally advanced vulvovaginal cancer: II. Results, complication and dosimetric and surgical considerations. Am J Clin Oncol 1987; 10: 171. 66. Russell AH, et al. Synchronous radiation and cytotoxic chemotherapy for locally advanced or recurrent squamous cancer of the vulva. Gynecol Oncol 1992; 47: 14. 67. Berek JS, et al. Concurrent cisplatin and 5-fluorouracil chemotherapy and radiation therapy for advanced stage squamous carcinoma of the vulva. Gynecol Oncol 1991; 42: 197.
530
GYNECOLOGICAL CANCERS
68. Koh WJ, et al. Combined radiotherapy and chemotherapy in the management of local-regional advanced vulvar cancer. Int J Radiat Oncol Biol Phys 1993; 26: 809. 69. Thomas G, et al. Concurrent radiation and chemotherapy in vulvar carcinoma. Gynecol Oncol 1989; 34: 263. 70. Eifel PJ, et al. Prolonged continuous infusion cisplatin and 5fluorouracil with radiation for locally advanced carcinoma of the vulva. Gynecol Oncol 1995; 59: 51. 71. Lupi G, et al. Combined preoperative chemoradiotherapy followed by radical surgery in locally advanced vulvar carcinoma: a pilot study. Cancer 1996; 77: 1472. 72. Landoni F, et al. Concurrent preoperative chemotherapy with 5fluorouracil and mitomycin C and radiotherapy (FUMIR) followed by limited surgery in locally advanced and recurrent vulvar carcinoma. Gynecol Oncol 1996; 61: 321. 73.. Geisler J, et al. Neoadjuvant Chemotherapy in Vulvar Cancer: Avoiding Primary Exenteration. 2004. 74. Hacker NF, et al. Management of regional lymph nodes and their prognostic influence in vulvar cancer. Obstet Gynecol 1983; 61(4): 408 – 12. 75. Podratz KC, Symmonds RE, Taylor WF. Carcinoma of the vulva: analysis of treatment failures. Am J Obstet Gynecol 1982; 143(3): 340 – 51. 76. Origoni M, et al. Prognostic value of pathological patterns of lymph node positivity in squamous cell carcinoma of the vulva stage III and IVA FIGO. Gynecol Oncol 1992; 45(3): 313 – 6. 77. Piver MS, et al. Adriamycin alone or in combination in 100 patients with carcinoma of the cervix or vagina. Am J Obstet Gynecol 1978; 131: 311. 78. Hopkins MP, Reid GC, Morley GW. The surgical management of recurrent squamous cell carcinoma of the vulva. Obstet Gynecol 1990; 75: 1001. 79. Piura B, et al. Recurrent squamous cell carcinoma of the vulva: a study of 73 cases. Gynecol Oncol 1993; 48: 189. 80. Durrant KR, et al. Bleomycin, methotrexate, and CCNU in advanced inoperable squamous cell carcinoma of the vulva: a phase II study of the EORTC Gynaecological Cancer Cooperative Group (GCCG). Gynecol Oncol 1990; 37(3): 359 – 62. 81. Wagenaar HC et al., European Organization for Research and Treatment of Cancer. Bleomycin, methotrexate, and CCNU in locally advanced or recurrent, inoperable, squamous-cell carcinoma of the vulva: an EORTC Gynaecological Cancer Cooperative Group Study. Gynecol Oncol 2001; 81(3): 348 – 54. 82. Sirisabya N. Clinical trials of bleomycin on female genital cancer: a preliminary report. J Med Assoc Thai 1974; 56: 101. 83. Raitanen M, et al. Supra-additive effect with concurrent paclitaxel and cisplatin in vulvar squamous cell carcinoma in vitro. Int J Cancer 2002; 100(2): 238 – 43. 84. Boabang P, et al. Anti-neoplastic activity of topotecan versus cisplatin, etoposide and paclitaxel in four squamous cell cancer cell lines of the female genital tract using an ATP-Tumor Chemosensitivity Assay. Anticancer Drugs 2000; 11(10): 843 – 8. 85. Jaakkola M, et al. Vulvar squamous cell carcinoma cell lines are sensitive to paclitaxel in vitro. Anticancer Res 1997; 17: 939. 86. Deppe G, Cohen C, Bruckner H. Chemotherapy of squamous cell carcinoma of the vulva: a review. Gynecol Oncol 1979; 7: 345. 87. Thigpen JT, et al. Phase II trials of cisplatin and piperazinedione in advanced or recurrent squamous cell carcinoma of the vulva: a Gynecologic Oncology Group study. Gynecol Oncol 1986; 23: 358. 88. Benda JA, Platz CE, Anderson B. Malignant melanoma of the vulva: a clinical pathologic review of 16 cases. Int J Gynecol Pathol 1986; 5: 202. 89. Podratz KC, et al. Melanoma of the vulva: an update. Gynecol Oncol 1983; 16: 153. 90. Raber G, et al. Malignant melanoma of the vulva: report of 89 patients. Cancer 1996; 78: 2353. 91. Morrow CP, DiSaia PJ. Malignant melanoma of the female genitalia: a clinical analysis. Obstet Gynecol Surv 1976; 31: 233. 92. Johnson TL, Kumar N, White CD. Prognostic features of vulvar melanoma: a clinicopathologic analysis. Int J Gynecol Pathol 1986; 5: 110.
93. Chung AF, Woodruf JM, Lewis JL Jr. Malignant melanoma of the vulva: a report of 44 cases. Obstet Gynecol 1975; 45: 638. 94. Morrow CP, Rutledge FN. Melanoma of the vulva. Obstet Gynecol 1972; 39: 745. 95. Ariel IM. Malignant melanoma of the female genital system: a report of 48 patients and review of the literature. J Surg Oncol 1981; 16: 371. 96. MacKie RM, et al. Prognostic models for subgroups of melanoma patients from the Scottish Melanoma Group Database 1979 – 1986, and their subsequent validation. Br J Cancer 1995; 71: 173. 97. Balch CM. Cutaneous melanoma: prognosis and treatment results. Semin Surg Oncol 1992; 8: 400. 98. Scheistroen M, et al. Malignant melanoma of the vulva FIGO stage I: evaluation of prognostic factors in 43 patients with emphasis on DNA ploidy and surgical treatment. Gynecol Oncol 1996; 61: 253. 99. Scheistroen M, et al. Malignant melanoma of the vulva. Evaluation of prognostic factors with emphasis on DNA ploidy in 75 patients. Cancer 1995; 75: 72. 100. Phillips GL, et al. Malignant melanoma of the vulva treated by radical hemivulvectomy: a prospective study of the Gynecologic Oncology Group. Cancer 1994; 73: 2626. 101. Veronesi U, et al. Primary cutaneous melanoma 2 mm or less in thickness: results of a randomized study comparing wide with narrow surgical excision: a preliminary report. N Engl J Med 1988; 318: 1159. 102. Sim FH, Taylor WF, Pritchard DJ. Lymphadenectomy in the management of stage I malignant melanoma: a prospective randomized study. Mayo Clin Proc 1986; 61: 697. 103. Balch CH, et al. Efficacy of 2 cm surgical margins for intermediate thickness melanoma (1 to 4 mm): results of a multi institutional randomized surgical trial. Ann Surg 1993; 218: 262. 104. Ringborg U, et al. Resection margins of 2 versus 5 cms for cutaneous malignant melanoma with a tumor thickness of 0.8 to 2 mm: randomized study by the Swedish Melanoma Study Group. Cancer 1996; 77: 1809. 105. Balch CM. Surgical management of regional lymph nodes in cutaneous melanoma. J Am Acad Dermatol 1980; 3: 511. 106. Veronesi U, et al. Inefficacy of immediate node dissection in stage I melanoma of the limbs. N Engl J Med 1977; 297: 627. 107. Veronesi U, et al. Delayed regional lymph node dissection in stage I melanoma of the skin of the lower extremities. Cancer 1982; 49: 2420. 108. Cascinelli N, et al. Immediate or delayed dissection of regional nodes in patients with melanoma of the trunk: a randomized trial WHO Melanoma Program. Lancet 1998; 351: 793. 109. Balch CM, et al. 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 1996; 224: 255. 110. Kirkwood JM, et al. Interferon alfa-2b adjuvant therapy for highrisk resected cutaneous melanoma: the Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 1996; 14: 7. 111. Pehamberger H et al., Austrian Malignant Melanoma Cooperative Group. Adjuvant interferon alfa-2a treatment in resected primary stage II cutaneous melanoma. J Clin Oncol 1998; 16: 1425. 112. Rusciani L, et al. Postsurgical adjuvant therapy for melanoma: evaluation of a 3-year randomized trial with recombinant interferon alpha after 3 and 5 years of follow-up. Cancer 1997; 79: 2354. 113. Creagan ET, et al. Randomized surgical adjuvant clinical trial of recombinant interferon alfa 2a in selected patients with malignant melanoma. J Clin Oncol 1995; 13: 2776. 114. Cascinelli N. Evaluation of efficacy of adjuvant rIFNa2a 2A in melanoma patients with regional node metastases. Proc Am Soc Clin Oncol 1995; 14: 410. 115. Hoffman MS, Roberts WS, Ruffolo EH. Basal cell carcinoma of the vulva with inguinal lymph node metastases. Gynecol Oncol 1988; 29: 113. 116. Benedet JL, et al. Basal cell carcinoma of the vulva: clinical features and treatment results in 28 patients. Obstet Gynecol 1997; 90: 765. 117. Feakins RM, Lowe DG. Basal cell carcinoma of the vulva: a clinicopathologic study of 45 cases. Int J Gynecol Pathol 1997; 16: 319. 118. DiSaia PJ, Rutledge R, Smith JP. Sarcoma of the vulva. Obstet Gynecol 1971; 38: 180.
TUMORS OF THE VULVA AND VAGINA 119. Davos I, Abell MR. Soft tissue sarcoma of the vulva. Gynecol Oncol 1976; 4: 70. 120. Tavassoli FA, Norris HJ. Smooth muscle tumors of the vulva. Obstet Gynecol 1979; 53: 213. 121. Nielsen GP, et al. Smooth-muscle tumors of the vulva: a clinicopathological study of 25 cases and review of the literature. Am J Surg Pathol 1996; 20: 779. 122. Weiss SW, Enzinger FM. Malignant fibrous histiocytoma: an analysis of 200 cases. Cancer 1978; 41: 2250. 123. Perrone T, et al. Malignant rhabdoid tumor of the vulva: is distinction from epithelioid sarcoma possible? Am J Surg Pathol 1989; 13: 848. 124. Hilborne LH, Fu YS. Intraepithelial, invasive and metastatic neoplasms of the vagina. In Wilkinson EJ (ed) Pathology of the Vulva and Vagina. New York: Churchill Livingston, 1987: 184. 125. Herbst AL, Green TH Jr, Ulfelder H. Primary carcinoma of the vagina. Am J Obstet Gynecol 1970; 106: 210. 126. Brinton LA, et al. Case-control study of in situ and invasive carcinoma of the vagina. Gynecol Oncol 1990; 38: 49. 127. Boice JD, et al. Radiation dose and second cancer risk in patients treated for cancer of the cervix. Radiat Res 1988; 116: 3. 128. Lee JY, et al. The risk of second primaries subsequent to irradiation for cervix cancer. Int J Radiat Oncol Biol Phys 1982; 8: 207. 129. Ikenberg H, et al. Human papillomavirus DNA in invasive carcinoma of the vagina. Obstet Gynecol 1990; 76: 432. 130. Perez CA, et al. Malignant tumors of the vagina. Cancer 1973; 31: 36. 131. Kucera H, Vavra N. Primary carcinoma of the vagina: clinical and histopathological variables associated with survival. Gynecol Oncol 1991; 40: 12. 132. Berchuck A, et al. Expression of epidermal growth factor receptor and HER-2/neu in normal and neoplastic cervix, vulva and vagina. Obstet Gynecol 1990; 76: 381. 133. Perez CA, et al. Radiation therapy in carcinoma of the vagina. Obstet Gynecol 1974; 44: 862. 134. Skomedal H, Kristensen G, Holm R. Expression of retinoblastoma tumor suppressor gene protein, epidermal growth factor receptor, and c-erbB-2 oncoprotein in primary vaginal carcinoma. Gynecol Oncol 1995; 59: 379. 135. Stock RG, Chen AS, Seski J. A 30-year experience in the management of primary carcinoma of the vagina: analysis of prognostic factors and treatment modalities. Gynecol Oncol 1995; 56: 45. 136. Skomedal H, et al. TP53 gene mutations and protein accumulation in primary vaginal carcinomas. Br J Cancer 1995; 72: 129. 137. Perez CA, et al. Definite irradiation in carcinoma of the vagina: longterm evaluation of results. Int J Radiat Oncol Biol Phys 1988; 15: 1283. 138. Chyle V, et al. Definite radiotherapy for carcinoma of the vagina: outcome and prognostic factors. Int J Radiat Oncol Biol Phys 1996; 35: 891. 139. Davis KP, et al. Invasive vaginal carcinoma: analysis of early-stage disease. Gynecol Oncol 1991; 42: 131. 140. Rubin SC, Young J, Mikuta JJ. Squamous carcinoma of the vagina: treatment complications and long-term follow-up. Gynecol Oncol 1985; 20: 346. 141. Perez CA, Camel HM. Long-term follow-up in radiation therapy of carcinoma of the vagina. Cancer 1982; 49: 1308. 142. Piver MS, Barlow JJ, Xynos FP. Adriamycin alone or in combination in 100 patients with carcinoma of the cervix or vagina. Am J Obstet Gynecol 1978; 131: 311. 143. Evans LS, et al. Concomitant 5-fluorouracil, mitomycin-C and radiotherapy for advanced gynecologic cancer. Int J Radiat Biol Phys 1988; 15: 901. 144. Herbst AL. Clear cell adenocarcinoma and the current status of DESexposed females. Cancer 1981; 48: 484. 145. Robboy SJ, et al. Increased incidence of cervical and vaginal dysplasia in 3,980 diethylstilbestrol-exposed young women. JAMA 1984; 252: 2979. 146. Melnick S, et al. Rates and risks of diethylstilbestrol-related clear-cell adenocarcinoma of the vagina and cervix: an update. N Engl J Med 1987; 316: 514.
531
147. Trimble EL, et al. Vaginal clear cell adenocarcinoma in the United States. Gynecol Oncol 1996; 61: 113. 148. Boyd J, et al. Molecular genetic analysis of clear cell adenocarcinomas of the vagina and cervix associated and unassociated with diethylstilbestrol exposure in utero. Cancer 1996; 77: 507. 149. DeMars LR, et al. Primary non-clear cell adenocarcinomas of the vagina in older DES-exposed women. Gynecol Oncol 1995; 58: 389. 150. Hanselaar A, et al. Clear cell adenocarcinoma of the vagina and cervix: an update of the central Netherlands registry showing twin age incidence peaks. Cancer 1997; 79: 2229. 151. Fishman DA, et al. Late recurrences of vaginal clear cell adenocarcinoma. Gynecol Oncol 1996; 62: 128. 152. Senekjian ER, et al. Local therapy in stage I clear cell adenocarcinoma of the vagina. Cancer 1987; 60: 1319. 153. Chung AF, et al. Malignant melanoma of the vagina: report of 19 cases. Obstet Gynecol 1980; 55: 720. 154. Iverson K, Robins RE. Mucosal malignant melanomas. Am J Surg 1980; 139: 660. 155. Nigogosyan G, de la Pava S, Picken JW. Melanoblasts in vaginal mucosa: origin for primary malignant melanoma. Cancer 1964; 17: 912. 156. Ragnarsson-Olding B, et al. Malignant melanoma of the vulva and vagina: trends in incidence, age distribution, and long-term survival among 245 consecutive cases in Sweden 1960 – 1984. Cancer 1993; 71: 1893. 157. Weinstock MA. Malignant melanoma of the vulva and vagina in the United States: patterns of incidence and population-based estimates of survival. Am J Obstet Gynecol 1994; 171: 1225. 158. Buchanan DJ, Schlaerth J, Kurosaki T. Primary vaginal melanoma: thirteen-year disease-free survival after wide local excision and review of recent literature. Am J Obstet Gynecol 1998; 178: 1177. 159. Bonner JA, et al. The management of vaginal melanoma. Cancer 1988; 62: 2066. 160. Davos I, Abell MR. Sarcoma of the vagina. Obstet Gynecol 1976; 4: 342. 161. Hilgers RD, Malkasian GD, Soule EH. Embryonal rhabdomyosarcoma (botryoid type) of the vagina: a clinicopathologic review. Am J Obstet Gynecol 1970; 107: 484. 162. Copeland LJ, et al. Sarcoma botryoides of the female genital tract. Obstet Gynecol 1985; 66: 262. 163. Shy SW, et al. Rhabdomyosarcoma of the vagina in a postmenopausal woman: report of a case and review of the literature. Gynecol Oncol 1995; 58: 395. 164. Rodary C, Flamant F, Donaldson SS. An attempt to use a common staging system in rhabdomyosarcoma: a report from an international workshop initiated by the International Society of Pediatric Oncology (SIOP). Med Pediatr Oncol 1989; 17: 210. 165. Tavassoli FA, Norris JH. Smooth muscle tumors of the vagina. Obstet Gynecol 1979; 53: 689. 166. Peters WA III, et al. Primary sarcoma of the adult vagina: a clinicopathologic study. Obstet Gynecol 1985; 65: 699. 167. Curtin JP, Saigo P, Slucher B. et al. Soft-tissue sarcoma of the vagina and vulva: a clinicopathologic study. Obstet Gynecol 1995; 86: 269. 168. Oliva E, et al. Granulocytic sarcoma of the female genital tract: a clinicopathologic study of 11 cases. Am J Surg Pathol 1997; 21: 1156. 169. Harris NL, Scully RE. Malignant lymphoma and granulocytic sarcoma of the uterus and vagina. Cancer 1984; 53: 2530. 170. Chorlton I, et al. Primary malignant reticuloendothelial disease involving the vagina, cervix, and corpus uteri. Obstet Gynecol 1974; 44: 735. 171. Davidoff AM, et al. Endodermal sinus tumor in children. J Pediatr Surg 1996; 31: 1075. 172. Andersen WA, et al. Endodermal sinus tumor of the vagina. Cancer 1985; 56: 1025. 173. Copeland LJ, et al. Endodermal sinus tumor of the vagina and cervix. Cancer 1985; 55: 2558. 174. Young RH, Scully RE. Endodermal sinus tumor of the vagina: a report of nine cases and review of the literature. Gynecol Oncol 1984; 18: 380.
Section 7 : Gynecological Cancers
47
Gestational Trophoblastic Diseases Emily Berry and John R. Lurain
HISTORICAL BACKGROUND Gestational trophoblastic disease encompasses four clincopathologic forms of growth disturbances of the human placenta: (i) hydatidiform mole (complete and partial), (ii) invasive mole, (iii) choriocarcinoma, and (iv) placentalsite trophoblastic tumor (PSTT). The term gestational trophoblastic neoplasia (GTN) has been applied collectively to the latter three conditions, because the diagnosis and decision to institute treatment are often made without knowledge of the precise histology. Prior to the introduction of chemotherapy to the management of GTN approximately 50 years ago, outcomes were poor. The mortality rate for invasive mole approached 15%, most often due to sepsis, hemorrhage, embolic phenomena, or complications from surgery. Choriocarcinoma, when thought to be confined to the uterus, was cured by hysterectomy in only 40% of patients and almost all patients with metastatic disease died. The overall cure rate in the treatment of GTN now exceeds 90% even in the presence of widespread metastatic disease. This achievement is attributable to a combination of factors, including the inherent sensitivity of trophoblastic tumors to chemotherapy, the effective use of sensitive assays for the tumor marker human chorionic gonadotropin (hCG), the development of specialized treatment centers, the identification of prognostic factors that predicts treatment response and enhances individualization of therapy, and the use of aggressive, combined modality treatment with chemotherapy, surgery, and radiation in the highest risk patients.1
EPIDEMIOLOGY The incidence and etiologic factors contributing to the development of gestational trophoblastic disease are not well defined. The difficulty in accumulating reliable epidemiologic data can be attributed to a number of factors. Among these are (i) inconsistencies in case definitions, (ii) inability to adequately characterize the population at risk, (iii) lack of centralized databases, (iv) lack of well-chosen control groups against which to compare possible risk factors, and (v) rarity of gestational trophoblastic tumors, especially some variants, like choriocarcinoma.
Earlier epidemiologic studies reported wide regional variation in the incidence of hydatidiform moles, with the highest rates found in parts of Asia. Further scrutiny of such reports called into question the possibility of falsely elevated numbers, as the rates were often calculated from the number of cases at the reporting hospital divided by the total number of pregnancies or deliveries at the same hospital.2 This denominator was recognized to exclude large portions of the population who did not receive care and/or were not delivered at a hospital, or those who had unrecognized pregnancies or pregnancies not resulting in delivery. As such, one recognizes the difficulty in comparing studies from various regions of the world, as data collection and reporting were in no way standardized. In an analysis of epidemiologic data on the incidence of molar pregnancies between 1980 and 1994, Palmer was able to draw several conclusions.3 Estimates from studies conducted in North America, Asia, Australia, New Zealand, and Europe showed the incidence of hydatidiform mole to range from 0.57 to 1.1 per 1000 pregnancies. Studies from Japan suggested an incidence as high as 2.0 per 1000 pregnancies.4 Investigations into possible ethnic and racial differences in the incidence of trophoblastic disease suggest that rates of molar pregnancy may be higher in American Indian, Eskimo, non-White Hispanic, and various Asian (Filipino, Indonesian, and Japanese) populations, with continued highest rates in women from Southeast Asia and Japan. No evidence has yet emerged to suggest whether such ethnic differences are attributable to genetic traits or cultural factors. Around the world, incidence rates of molar pregnancies and choriocarcinoma have declined over the last 25–30 years.5 – 7 Several potential etiologic risk factors have been examined in relation to the development of complete hydatidiform mole. Thus far, two established risk factors have emerged, advanced or very young maternal age and prior molar pregnancy. Parazzini et al. showed that compared to women aged 21–35, the risk of complete mole was 1.9 times higher for women over age 35, 7.5 times higher for women over 40 years, and 1.9 times higher for women less than 21 years of age.8 The risk of repeat molar pregnancy after 1 mol is about 1%, 10 times the risk for women who never had a
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
GESTATIONAL TROPHOBLASTIC DISEASES
molar pregnancy.9,10 In addition, a study from Milan, Italy, demonstrated a two- to threefold increased risk of hydatidiform mole in women with prior spontaneous abortions as compared to those without a history of prior miscarriage.11 Although many other factors have been studied, none have consistently been shown to effect the risk of hydatidiform mole. These include parental age, ABO blood group, cigarette smoking, herbicide, and radiation exposure.3 Two studies from the United States and Italy found an inverse relationship of β-carotene dietary intake and the incidence of molar pregnancy. Berkowitz et al. showed that dietary carotene above the median level of consumption in a control group appeared to lower the risk of complete moles Relative risk (RR 0.6), while Parazzini et al. found a significant inverse relationship between molar pregnancy rates and β carotene as well as animal fat consumption.12,13 The relationship between oral contraceptives and gestational trophoblastic disease has also been examined. Studies have shown a possible increased risk of molar pregnancy and even choriocarcinoma with oral contraceptive use.3,7 Data with respect to choriocarcinoma incidence rates are even more limited. Collection of data on the incidence of choriocarcinoma has been made difficult not only for reasons similar to those encountered with hydatidiform moles but also because of the rarity of choriocarcinoma and the difficulty in clinically distinguishing postmolar choriocarcinoma from invasive mole. In Europe and North America, approximately 1 in 40 000 pregnancies and 1 in 40 molar pregnancies are affected, whereas in Southeast Asia and Japan, rates are higher at 0.23 per 1000 births and 0.83 cases per 10 000 births respectively.14,15 In terms of epidemiologic risk factors, choriocarcinoma is approximately 1000 times more likely to develop after a complete molar pregnancy rather than a normal gestation. Similar to molar pregnancies, the median age for women with choriocarcinoma is somewhat higher than that for normal pregnancies, and the risk is increased in women of Asian and American Indian descent and among US blacks.15 Other potential risk factors deserving further investigation include ABO blood group (higher prevalence of patients with group A), low endogenous estrogen states, and prolonged oral contraceptive use.3,15
PATHOLOGY Molar pregnancies and gestational trophoblastic neoplasms all take their origins from the most basic cells to human makeup, the trophoblasts. Shortly after fertilization, the zygote evolves to a morula and then to a blastocyst comprised of an outer cell layer called the trophoblast and an inner cell layer named the embryoblast. Once this blastocyst has attached to the uterine lining, the trophoblast begins to proliferate rapidly and differentiate into two layers: the cytotrophoblast and the syncytiotrophoblast. The syncytiotrophoblast is instrumental in blastocyst implantation as it extends into the endometrial stroma, resembling a large multinucleated mass in which no cell boundaries are discernible. It is the syncytiotrophoblast that produces hCG, the biochemical marker of pregnancy. The cytotrophoblast not only continues to supply the growing syncytiotrophoblast
533
with cells, but it also goes on to form the chorionic sac, a main component of the placental membranes containing the embryo and amniotic sac. The chorionic sac becomes covered with chorionic villi, outpouchings of the cytotrophoblast cells. The villi branch until they contain blood vessels through which fetal blood flows. In some areas, the villi degenerate and the chorion becomes smooth and avascular. In the area adjacent to the endometrium, the chorion remains villous, and it is this villous chorion together with the basalis layer of the endometrium that forms the placenta, the site of maternal-fetal nutrient and waste exchange.16 It is these placental cellular elements, the cytotrophoblast, syncytiotrophoblast, and a third component called the intermediate trophoblast, that form the constituents for molar pregnancies and gestational trophoblastic tumors. The classic description of the pathologic features of hydatidiform mole was published in two parts by Szulman and Surti in 1978. Asserting that the term mole is reserved “for conceptuses with macroscopic or microscopic villous hydrops accompanied by trophoblastic hyperplasia”, they went on to classify moles as either complete or partial, on the basis of the categories of gross morphology, histopathology, and karyotype.17,18 Complete moles are characterized by the absence of embryonic or fetal tissue. The chorionic villi are grossly and uniformly edematous with central cisterns of fluid, and diffuse trophoblastic hyperplasia is present. Initial villous capillaries disappear as cisterns form so that complete moles lack intrinsic vascularity. Approximately 90% of complete moles have a 46, XX karyotype, with the chromosomes entirely paternal in origin. Typically, a haploid (23X) sperm fertilizes an ovum, then duplicates its own chromosomes; the maternal chromosomes are either inactive or absent. Six to ten percent of complete moles have a 46, XY karyotype, again with all chromosomes of paternal origin. In these cases, the mole results from a dispermic event where an empty ovum is fertilized by two separate sperm.17 – 19 More recent flow cytometry studies have demonstrated that a small percentage of complete moles may also possess tetraploid karyotypes.20 Traditional descriptions of complete hydatidiform moles originated in the 1960s and 1970s, when moles were typically diagnosed in the second trimester. More recently, with improved sensitivity of hCG detection and the widespread use of first-trimester ultrasound, molar pregnancies are more likely to be diagnosed in the first trimester. This has allowed for pathologic study of younger complete molar specimens and has led to the discovery that several histologic and morphologic differences exist at earlier gestational ages. In 1998, Mosher et al.21 compared 23 contemporary complete moles (1994–1997) with 20 historical cases (1969–1975) in regard to clinical and pathologic findings. Numerous differences were demonstrated between contemporary and historic complete moles. Contemporary moles had less villous cavitation (39 vs 80%), smaller villous size (5.7 vs 8.2 mm), less circumferential trophoblastic hyperplasia (39 vs 75%), less frank villous necrosis (22 vs 54%), and finally an increase in the incidence of immature villous stromal characteristics (61–74 vs 5–10%) compared with historic moles. Other studies have confirmed the progression
534
GYNECOLOGICAL CANCERS
of histologic changes that occur as a complete mole matures, suggesting a risk for misdiagnosis if one is not familiar with the evolving microscopic appearance of a complete mole.22 Partial hydatidiform moles demonstrate identifiable fetal or embryonic tissue and histologically consist of chorionic villi with focal villous edema and some persistently immature unaffected villi. Trophoblastic hyperplasia is focal and described as mild to moderate. Villous scalloping, stromal trophoblastic inclusions, and a functioning villous circulation are additional microscopic features of partial moles. Karyotypes tend to be triploid (69, XXX or XXY), resulting from the fertilization of an apparently normal ovum by two sperm. Lage et al.20 reported a flow cytometric study in which 90% of partial moles had a triploid karyotype. Fetuses identified from partial molar specimens generally have the stigmata of triploidy, including growth restriction and multiple congenital anomalies.17,18 In general, a relationship seems to exist between excessive paternal chromosomes and trophoblastic hyperplasia.19 Choriocarcinoma can occur after any pregnancy event, from complete and partial molar pregnancies, to term gestations, stillbirths, abortions, and ectopic pregnancies. Macroscopically, the tumor appears as a hemorrhagic mass within the uterine cavity, sometimes demonstrating invasion into the myometrial wall. Microscopically, choriocarcinoma consists of sheets of anaplastic syncytiotrophoblast and cytotrophoblast, with an absence of chorionic villi. The tumor tends to elicit little to no surrounding host tissue inflammatory reaction. Viable cells are typically seen only at the periphery of an otherwise necrotic and hemorrhagic tumor, and as a rule no intrinsic vasculature can be demonstrated. Invasion into host blood vessels allows the tumor to grow, and also explains the propensity for hematogenous metastasis seen in choriocarcinoma. The presence of syncytiotrophoblast cells confers the ability to secrete hCG, and it has been shown that some predominantly cytotrophoblastic tumors secrete smaller amounts of hCG.23 PSTT was first described in 1976, when Kurman et al. introduced the term trophoblastic pseudotumor to refer to a series of 12 cases believed to exhibit an exaggerated but benign placental-site reaction.24 Scully and Young renamed the lesion in 1981, when further clinical evidence revealed that these presumed benign lesions of the uterus could exhibit locally aggressive behavior and the ability to metastasize.25 PSTT, the rarest of the trophoblastic diseases, is comprised of intermediate trophoblastic cells of the placental bed. Histologically, the intermediate trophoblasts are mononuclear with prominent nucleoli, lack chorionic villi, and are often found in sheets or cords infiltrating between myometrial fibers. Mitotic counts range from 1–2 mitotic figures to 50 mitotic figures per 10 high-power fields (hpf), with most tumors having <5 mitoses per 10 hpf’s. Immunohistochemistry staining reveals the diffuse presence of cytokeratin and human placental lactogen (hPL), while hCG is only focal. Grossly, the tumor is usually confined to the uterus and appears as either a polypoid mass within the uterine cavity or a lesion invading the myometrium. As opposed to choriocarcinoma, PSTTs have a propensity for metastasis through lymphatic rather than vascular channels, and necrosis is seen more commonly
than hemorrhage.26 – 28 Cytogenetic studies have revealed that PSTTs are more often diploid than aneuploid.29
CLINICAL PRESENTATION, DIAGNOSTIC CONSIDERATIONS, AND MANAGEMENT OF COMPLETE AND PARTIAL HYDATIDIFORM MOLES Complete Mole – Presenting Signs and Symptoms Vaginal bleeding remains the most common presenting symptom in patients with complete molar pregnancies, occurring in 89–97% of reported cases at trophoblastic disease centers.30 – 32 Associated anemia has historically been noted at presentation, especially in the second trimester, with 10–54% of patients having a hemoglobin <10 g dL−1 .30,32 Excessive uterine size has historically been seen in approximately 50% of patients at presentation, and is defined as larger than dates by four weeks. Theca lutein ovarian cysts are frequently seen in association with molar pregnancies. Their reported frequency can vary depending on the mode of diagnosis, with studies citing a 20–26% incidence on clinical exam as opposed to a 46% detection rate using ultrasound.33,34 Cysts are typically 6–12 cm in size, resulting from ovarian hyperstimulation by high serum levels of hCG. It has been shown that the mean time to resolution of these cysts after uterine evacuation of a molar pregnancy is 8 weeks, but theca lutein cysts have been known to enlarge even after termination of a molar pregnancy.33 Complications such as an ovarian torsion are infrequent. Presenting signs of preeclampsia and hyperemesis gravidarum have a strong association with excessive uterine size and increased hCG levels and have been reported to occur in as many as 12–25% of cases.30 – 32 Hyperthyroidism, reported in 2–7% of cases of hydatidiform moles, is associated with high levels of circulating hCG, as hCG is known to have thyroid-stimulating activity that influences thyroid function.35 One risk of poorly controlled or unrecognized hyperthyroidism is the development of thyroid storm at the time of anesthesia induction for molar evacuation. Signs of thyroid storm include hyperthermia, delirium, coma, atrial fibrillation, and ultimately cardiovascular collapse; β-adrenergic blocking agents are instrumental in preventing and reversing this condition. Finally, earlier reports cited increased incidence of acute onset respiratory insufficiency after evacuation of complete molar pregnancies at or greater than 16 weeks gestation.36 Such respiratory decline (tachycardia, tachypnea, hypoxia, and respiratory alkalosis) has been attributed to a host of factors, including embolization of molar tissue, complications of preeclampsia, thyroid storm, or large volume fluid replacement. It should be remembered that only some of these signs and symptoms might occur in any one patient. The classic signs and symptoms of complete hydatidiform moles were first described in the 1960s and 1970s as centralized trophoblastic disease centers organized and began to publish large series reports on the characteristics of complete and later partial hydatidiform moles. The common signs mentioned in the previous paragraph (vaginal bleeding, excessive uterine size, theca lutein cysts, hyperemesis
GESTATIONAL TROPHOBLASTIC DISEASES
gravidarum, preeclampsia, and hyperthyroidism) were seen in many patients who presented for diagnosis and treatment in the second trimester of pregnancy, around 16 weeks gestation. Today, the majority of patients with complete molar pregnancies are being diagnosed from 8 to 11 weeks gestation as a result of the increased sensitivity and availability of first-trimester ultrasound and radiolabeled monoclonal antibody hCG assays. Soto-Wright et al. from the New England Trophoblastic Disease Center compared the clinical presentation of complete moles in 306 patients treated from 1965 to 1975 to 74 patients treated from 1988 to 1993. They found a significant reduction in the occurrence of the most common symptoms: vaginal bleeding (97 vs 84%), size > dates (51 vs 28%), anemia (54 vs 5%), preeclampsia (27 vs 1.3%), hyperemesis (26 vs 8%), hyperthyroidism (7 vs 0%), and respiratory distress (2 vs 0%). Interestingly, even though the authors were able to show a change in the clinical presentation over time, they observed no difference in the incidence of postmolar GTN.37
Partial Mole – Presenting Signs and Symptoms Patients with partial moles do not have the same presenting features as those with complete moles. A review of 81 patients with partial molar pregnancies diagnosed on curettage specimens was undertaken to better delineate the clinical presentation and natural history. Vaginal bleeding (72.8%) and absent fetal heartbeat (14.8%) were the most common presenting symptoms. Excessive uterine size was seen at a rate of 3.7%, and toxemia at 2.5%. Another 2.5% of patients were asymptomatic. Only two patients (6.6%) had a preevacuation hCG of >100 000 mIU mL−1 ; one presented with preeclampsia and the other with enlarged uterine size on exam. No patients had theca lutein cysts, hyperemesis, or hyperthyroidism.38 Similar studies of partial moles by Czernobilsky et al.39 and Szulman and Surti40 confirmed that partial moles lack uniform presenting signs and symptoms, and are often presumed to be incomplete or missed abortions prior to evacuation. The diagnosis of a partial hydatidiform mole is, therefore, most reliably made after pathologic review of a curettage specimen.
Diagnostic Investigations Ultrasonography plays a role in the diagnosis of both the complete and partial mole. Historically, with complete moles, ultrasound has proven to be a very sensitive and reliable exam, as patients presenting at later gestational ages were found to have the diffuse and marked swelling of the chorionic villi producing a characteristic vesicular pattern. More recently, however, the sensitivity of ultrasound in the diagnosis of complete moles has been reevaluated as more patients undergo ultrasound studies earlier in pregnancy. Benson et al.41 evaluated the sonographic appearance of first-trimester moles (mean gestational age 8 weeks). Of the 22 patients with complete molar pregnancies, interpretation on review of the images was a complete mole in 18 cases (82%). The typical sonographic appearance of a first-trimester complete mole was a complex, echogenic, intrauterine mass containing many small cystic spaces. An
535
earlier study by Soto-Wright et al. reported similar findings when reviewing 69 first-trimester ultrasounds in patients with complete moles, where ultrasound suggested the diagnosis in 71% of cases.37 The pathologic evolution of a complete mole has a direct effect on the ultrasound appearance and will make diagnosis less sensitive at earlier gestational ages. Ultrasonography may also contribute to the diagnosis of partial moles. In a study from the New England Trophoblastic Disease Center, which examined the ultrasound findings in 22 partial molar pregnancies, two sonographic characteristics were found to be significantly associated with partial moles: ratio of transverse to anteroposterior dimension of the gestational sac greater than 1.5 and irregular appearance with cystic changes in the tissues surrounding the gestational sac. Taken separately, each feature had a positive predictive value for partial mole of 56 and 67%, respectively. Taken together, the positive predictive value for partial mole was 87%. In more than half the cases, the diagnosis of partial mole was not made by ultrasound.42 Molar pregnancies, and trophoblastic diseases in general, are unique in that they produce the specific disease marker hCG, which can be measured easily in urine and blood and has been shown to correlate with the amount of disease present. hCG is a placental glycoprotein composed of two dissimilar subunits, α and β; the α subunit resembles that of the pituitary glycoprotein hormones and the β subunit is unique. Markedly elevated levels of hCG are commonly seen in patients with complete molar pregnancy, with preevacuation levels >100 000 mIU mL−1 in approximately 50% of patients.43,44 Partial moles, on the other hand, are often not distinguished by such elevated levels. Several forms of hCG exist, including at least six major variants of hCG detected in serum: hyperglycosylated, nicked, hCG missing the β-subunit C-terminal segment, free β subunit, nicked free β subunit, and free α subunit. In urine, these forms plus urine β-core fragments exist. In the serum and urine samples of patients with trophoblastic disease, the hCG molecules are more heterogeneous or degraded than those in normal pregnancy samples. Knowing this, it is important that when following patients an assay is used that will detect all main forms of hCG and its multiple fragments. Developed in the 1950s, the radioimmunoassay was the mainstay of hCG detection, using an hCG β-subunit polyclonal antibody and measuring all forms of the β subunit of hCG. Today, most facilities use one of many fast and automated radiolabeled monoclonal antibody “sandwich” assays that measure different mixtures of the hCG-related molecules.45 Studies using monoclonal antibodies with high sensitivity and specificity for measuring hCG and its free subunits suggest that trophoblastic cells in normal pregnancy, complete and partial moles, and choriocarcinoma differ substantially in the manner in which they secrete the free subunits of hCG. One study looking at the ratio of β-hCG to hCG showed that such ratios could distinguish normal pregnancy from complete mole, hydatidiform mole from choriocarcinoma, and choriocarcinoma from normal pregnancy with high probability.46 Another study suggested that the trophoblastic cells in complete and partial moles differed significantly in
536
GYNECOLOGICAL CANCERS
the manner in which they secreted free hCG subunits, in that complete moles were found to have a higher percentage of free β-hCG, whereas partial moles made more α-hCG.47
Treatment Once the diagnosis of molar pregnancy is suspected by history, hCG levels, and ultrasound findings, the patient should be evaluated for the presence of medical complications (anemia, preeclampsia, hyperthyroidism) by way of physical exam and laboratory tests, including complete blood counts, basic chemistry, hepatic and thyroid panels, and chest X ray. A decision must then be made concerning the most appropriate method of evacuation. If a patient no longer desires to retain fertility, hysterectomy may be performed with the mole in situ. This allows for aspiration of prominent theca lutein cysts at the same time along with conservation of ovaries. While hysterectomy eliminates the risk of local invasion, it does not protect the patient from potential metastatic disease. For patients who wish to maintain fertility, suction curettage is a safe, rapid, and effective method of evacuation, regardless of uterine size.48 After anesthesia is achieved, the cervix is dilated and a suction curette placed inside the uterus and rotated as the intrauterine contents are removed. It is recommended that an oxytocin infusion be started at the onset of suction curettage to increase myometrial tone and facilitate uterine contractions, thus minimizing blood loss. Suction curette is generally followed by gentle sharp curettage of the endometrium. Small series have shown an increased operative complication rate in a uterus enlarged beyond 16 weeks.49 Finally, it should be emphasized that adequate equipment, access to blood products, and early anticipation of potential complications, such as thyroid storm, pulmonary embolism, and blood loss, can improve the patients’ ultimate outcome. Patients who are Rh negative should receive Rh immune globulin at the time of evacuation, as Rh D factor is expressed on trophoblastic cells. Prophylactic administration of either methotrexate or actinomycin D chemotherapy at the time of evacuation of a hydatidiform mole is associated with a reduction in the incidence of postmolar GTN from approximately 15–20% down to 3–8%.50,51 The use of prophylactic chemotherapy should be limited, however, to special situations in which the risk of developing postmolar trophoblastic neoplasia is much greater than normal or where adequate hCG follow-up is not possible, since essentially all patients who are followed with serial hCG testing after evacuation of a hydatidiform mole and are found to have plateauing or rising hCG levels with or without metastases can be cured with appropriate chemotherapy.
Follow-up of Molar Pregnancies After evacuation of a molar pregnancy, it is recommended that all patients be followed with serial serum hCG measurements, to assure that remission is obtained, as approximately 20% of patients will develop persistent disease requiring chemotherapy. Ideally, serum hCG levels should be obtained within 48 h of evacuation, every 1–2 weeks while elevated, and then monthly for 6 months once a normal hCG level has been reached. If a chest X ray was not obtained preoperatively, it should be performed postevacuation to serve as a
baseline study. Regular follow-up visits with pelvic exams should be performed to assist in the early detection of vaginal metastases and to follow uterine involution and ovarian cyst regression. Most patients who develop postmolar GTN will be diagnosed within 6 months of evacuation and have nonmetastatic invasive mole, but gestational choriocarcinoma and metastatic disease can develop in this setting as well. Lurain et al. reviewed the clinical course of 738 patients with hydatidiform mole followed at the Brewer Trophoblastic Disease Center after evacuation. The authors reported that 80.8% of patients had spontaneous regression of hCG levels to normal, while 19.2% of patients required treatment with chemotherapy for persistent GTN (16.9% with invasive mole and 2.3% with choriocarcinoma). Interestingly, the study also examined the rate of hCG regression postevacuation. Regression of hCG to normal occurred in 1.8% by postoperative day 10, in 20.8% between days 11 and 30, in 42.8% between days 31 and 60, and in 34.6% between days 61 and 170.52 Complete and partial hydatidiform moles differ in their invasive potential and propensity for malignant transformation. Approximately 8–20% of patients with complete moles will develop persistent trophoblastic disease following uterine evacuation. Partial moles give rise to persistent trophoblastic disease in less than 3% of cases.53 After evacuation, patients are encouraged to use effective contraception throughout the follow-up interval (6–12month remission), preferably hormonal or barrier methods, as intrauterine devices carry risks of uterine perforation and uterine infection if residual tumor remains. In the past, concerns have been raised about the effects of exogenous hormones on trophoblastic tissue, with reports of increased incidence of persistent gestational trophoblastic disease developing in patients taking oral contraceptives. On the contrary, studies from the Brewer Center54 and the Gynecologic Oncology Group,55 as well as others, have shown that oral contraceptives do not increase the risk of postmolar trophoblastic neoplasia, making oral contraceptives the preferred method of contraception after evacuation of a hydatidiform mole.
GESTATIONAL TROPHOBLASTIC NEOPLASIA GTN includes invasive moles, choriocarcinomas, and PSTTs. These are diagnosed when there is clinical, radiologic, pathologic, and/or hormonal evidence of persistent trophoblastic tissue. Although GTN most commonly follows a molar pregnancy, it may occur after any type of gestation, from term and ectopic pregnancies to induced or spontaneous abortions. On rare occasions, the type of antecedent pregnancy cannot be documented. If left untreated, these tumors have the capacity to invade, metastasize, and kill. Unique to GTN is the elaboration of an easily determined tumor marker (hCG) and an inherent sensitivity to chemotherapy.56 Invasive mole is a histologically benign condition resulting from invasion of abnormal trophoblasts into myometrium and/or embolization of molar tissue through pelvic venous plexi. Approximately 15% of patients with invasive mole have metastases, most commonly in the lungs and vagina. The most common symptom of invasive mole is irregular
GESTATIONAL TROPHOBLASTIC DISEASES
vaginal bleeding following evacuation of a molar pregnancy. If metastases have occurred, bleeding from these sites may be present as well. The diagnosis is usually made clinically by observation of persistently elevated or rising hCG levels after molar evacuation.56,57 Choriocarcinoma is a malignancy that arises from trophoblastic tissue of term pregnancies, ectopic gestations, spontaneous/induced abortions, or molar elements. Some tumors are diagnosed long after the antecedent pregnancy. Choriocarcinoma is known to invade and metastasize early and is often widespread at the time of diagnosis. Lung metastases are present in almost all patients with extrauterine disease, with other common sites of metastases being the brain, liver, vagina, kidney, spleen, and the intestines. Two to three percent of hydatidiform moles progress to choriocarcinoma, comprising 50% of all cases, with another 25% from aborted pregnancies and the remaining 25% following term deliveries. The incidence of choriocarcinoma is approximately 1 in 40 000 pregnancies. Presenting signs and symptoms of choriocarcinoma are highly variable, and are often related to invasion at metastatic sites, causing a variety of pulmonary, abdominal, or neurologic symptoms.56 PSTT is a rare manifestation of GTN that develops at the placental implantation site and may complicate any type of pregnancy, most commonly a term gestation. PSTT differs from choriocarcinoma in that there is only one cell type, intermediate trophoblasts, instead of alternating sheets of syncytio- and cytotrophoblasts. hCG levels are typically elevated, but not to the degree seen with choriocarcinoma. hPL is present in tumor cells on immunohistochemical staining, but cannot be used as a tumor marker. These tumors are relatively resistant to chemotherapy, but surgery is usually curative if the disease is localized. Almost all patients with PSTT present with irregular vaginal bleeding and rarely virilization or nephrotic syndrome. Time from antecedent gestation has been reported from 1 week to 14 years.58,59 GTN is most often diagnosed by rising or plateauing hCG levels following evacuation of a hydatidiform mole, as well as a histopathologic diagnosis of invasive mole, choriocarcinoma or PSTT, or persistent elevation of hCG, frequently in conjunction with the discovery of metastases, following any pregnancy event. Once the diagnosis of GTN is suspected or established, immediate evaluation for metastases and risk factors is crucial. Along with a complete history and physical exam, the following laboratory tests should be obtained: complete blood count including platelets, coagulation studies, serum chemistries including renal and liver function panels, blood type and antibody screen, and determination of quantitative serum hCG levels. Researchers at the Charing Cross Hospital have also advocated measurement of hCG in cerebrospinal fluid as a diagnostic aid for determining brain metastases.60 Recommended radiographic studies include chest X ray with computed tomography (CT) scan of the chest if negative, CT scans of the abdomen and pelvis, and CT or magnetic resonance imaging (MRI) of the brain. Repeat curettage is not recommended because it does not often induce remission or influence treatment and may result in uterine perforation and hemorrhage.61
537
Once all information has been gathered, patients are categorized on the basis of anatomic extent of disease and likelihood of response to chemotherapy. Because many cases of GTN are identified by clinical criteria and no histologic diagnosis exists, treatment of these tumors is based on clinical prognostic factors. Thus, there is little difference between invasive mole and choriocarcinoma in terms of the way that the treatment strategy is planned. More important than histology is rapid diagnosis so that treatment may be administered promptly.1,56
STAGING CLASSIFICATION The classification of GTN has mostly relied upon relating prognostic factors to outcomes. In the early 1970s, investigators at the United States National Institutes of Health defined a clinical classification scheme on the basis of an analysis of patients treated for GTN. This system separated patients with nonmetastatic disease from those with metastatic disease. Of those with metastatic disease, a further breakdown was made between those with good and poor prognosis on the basis of factors that correlated with poor response to initial single-agent chemotherapy, including duration of disease >4 months, pretherapy hCG level >100 000 IU per 24 h urine or >40 000 mIU mL−1 serum, presence of liver or brain metastases, antecedent term pregnancy, and prior failed chemotherapy.62 This clinical classification system has remained popular for making treatment decisions in the United States. In 1976, Bagshawe et al.63 at the Charing Cross Hospital in London outlined a prognostic factor scoring system for trophoblastic neoplasia, which was adopted by the World Health Organization (WHO) scientific group on gestational trophoblastic disease in 1982, and ultimately modified in 2000 (see Table 1). Finally, the International Federation of Gynecology and Obstetrics (FIGO) adopted an anatomic staging system, as proposed by Song et al. from Beijing, China, in 1982 (see Table 2).64 All three of the above systems correlate with clinical outcomes of patients treated for GTN and identify patients at risk for failure to respond to chemotherapy.65 The use of these different classification schemes, however, has led to difficulties in comparing data between trophoblastic centers, raising questions concerning the applicability of newly emerging chemotherapy regimens. In an effort to find common ground, a proposal was made to the FIGO staging committee in 2000, consisting of a combined staging and risk factor scoring system that was accepted by FIGO in 2002.66 In the 2002 system, the FIGO stages are maintained, designated by a Roman numeral, followed the WHO prognostic score, designated by an Arabic numeral, separated by a colon. Patients with PSTTs are classified separately. The 2002 staging system also outlined the necessary components needed to diagnose postmolar mole GTN. These include at least one of the following: (i) hCG plateau for four values or more over at least three weeks (days 1, 7, 14, 21; plateau level to be determined by the physician); (ii) hCG rise of 10% or greater for three or more values over at least two weeks (days 1, 7, 14); (iii) presence of histologic
538
GYNECOLOGICAL CANCERS
Table 1 Modified WHO scoring system for gestational trophoblastic neoplasia.
Score Risk factor Age (year) Antecedent pregnancy Pregnancy event to treatment interval (months) Pretreatment hCG (mIU mL−1 ) Largest tumor mass, including uterus (cm) Site of metastases Number of metastases Previous failed chemotherapy
0
1
2
≤39 Hydatidiform mole <4 <103 – – – –
>39 Abortion 4–6 103 – 104 3–4 Spleen kidney 1–4 –
– Term 7 – 12 104 – 105 ≥5 GI tract 4–8 1 drug
4 – >12 >105 – Brain liver >8 ≥2 drugs
The total score for a patient is obtained by adding the individual scores for each prognostic factor: <7 = low risk; ≥7 = high risk.
Table 2 FIGO staging for gestational trophoblastic neoplasia.
Stage I Stage II Stage III Stage IV
Disease confined to the uterus Disease extends outside the uterus but is limited to genital structures (adnexa, vagina, broad ligament) Disease extends to lungs with or without genital tract involvement Disease involves other metastatic sites
choriocarcinoma; and (iv) hCG persistence 6 months after molar evacuation. For the diagnosis of metastases, the following requirements are: (i) for lung metastases, chest X ray is adequate and is used to count the number of metastases for risk score assessment; (ii) for intra-abdominal metastases, CT scanning is preferred; and (iii) for diagnosing brain metastases, MRI is superior to CT scanning. With regard to risk groups defined by the scoring system, there is a designated low-risk group with a score of 6 or less, and a high-risk group with a score of 7 or greater (see Table 1). Patients with nonmetastatic (FIGO stage I) and low-risk metastatic (FIGO stages II and III; WHO score <7) GTN can be treated with single-agent chemotherapy, with resulting survival rates approaching 100%. Patients classified as having high-risk metastatic disease (FIGO stage IV or WHO score ≥7) should be treated in a more aggressive manner with combination chemotherapy ± adjuvant radiation therapy or surgery to achieve cure rates of 80–90%.1
TREATMENT Gestational trophoblastic neoplasms have emerged as the most curable of all solid tumors since the initial discovery over 50 years ago that trophoblastic growth was inhibited by the folic acid antagonist methotrexate.67 This led investigators at the United States National Institutes of Health to spend almost a decade studying and documenting the effects of chemotherapy on gestational trophoblastic tumors. In patients with metastatic disease, they obtained complete sustained remissions in 47% using methotrexate alone and in 74% using methotrexate and actinomycin D (an antibiotic that inhibits DNA transcription). In patients with nonmetastatic disease, they achieved complete remissions in 93% using methotrexate without hysterectomy. Their work made major contributions to knowledge about cure of cancer by
chemotherapy, dosing regimens, and monitoring chemotherapy response. Chemotherapeutic agents other than methotrexate and actinomycin D have subsequently been found to be effective in GTN, including the alkylating agents cyclophosphamide and chlorambucil, the antimetabolite 5-fluorouracil, the vinca alkaloids vincristine and vinblastine, and the antitumor antibiotic bleomycin, as well as more recently etoposide, ifosfamide, the platinum agents (cisplatin and carboplatin), and paclitaxel.1
SINGLE-AGENT CHEMOTHERAPY Nonmetastatic Disease Essentially all patients with nonmetastatic GTN (FIGO stage I) can be cured, usually without compromising future fertility. Single-agent chemotherapy with either methotrexate or actinomycin D is the preferred treatment, with a number of outpatient protocols available, all yielding excellent and fairly comparable results. Hysterectomy may be used as part of primary therapy in patients who no longer desire fertility. The traditional and most effective protocol consists of methotrexate 0.4 mg kg−1 (max. 25 mg) i.v. or i.m. for 5 days every 2 weeks. This regimen was reviewed by Lurain and Elfstrand in 1995, analyzing patients with nonmetastatic disease treated over a 30-year period at the Brewer Trophoblastic Disease Center. The primary remission rate was 89%, with all 253 patients eventually being placed into permanent remission and only 2% requiring multiagent chemotherapy or hysterectomy. No life-threatening toxicities occurred, and the most common side effect was oropharyngeal stomatitis.68 Modifications of dosage and administration schedules of methotrexate that have been trialed and reported include (i) 1.0–1.5 mg kg−1 given i.m. every other day, alternating with folinic acid 0.1–0.15 mg kg−1 i.m. over a total of 8 days with 1 week rest in between courses;69 (ii) 30–50 mg m−2 i.m. weekly;70 and (iii) 100 mg m−2 i.v. push followed by 200 mg m−2 i.v. 12-h infusion with folinic acid 15 mg i.m. or p.o. every 12 h for four doses beginning 24 h after the start of methotrexate, with the interval between doses reliant on posttreatment hCG trends.71 Altogether, these three regimens have shown lower primary remission rates compared to 5-day every other week dosing, most likely because of the overall shorter exposure time to the chemotherapeutic agent. Actinomycin D given 10–12 mg kg−1 i.v. for 5 days every other week or as a single 1.25 mg m−2 i.v. dose every
GESTATIONAL TROPHOBLASTIC DISEASES
2 weeks is an acceptable alternative to methotrexate; however, it has a more toxic side-effect profile (nausea, alopecia) than methotrexate and produces local tissue injury if i.v. extravasation occurs. Therefore, actinomycin D is most often used as secondary therapy in the presence of methotrexate resistance or as primary therapy when patients have renal or hepatic disease contraindicating the use of methotrexate. Actinomycin D has produced primary remission rates ranging between 78 and 94% in patients with nonmetastatic GTN.68,72 Regardless of the protocol chosen, chemotherapy is continued until hCG values have returned to normal levels, with one to two additional courses of treatment given after the first normal hCG level is obtained. Patients with plateauing or increasing hCG levels during chemotherapy should be switched to an alternative single-agent regimen. Multiagent chemotherapy is reserved for those rare failures of single-agent treatment. Hysterectomy can be considered for disease that remains in the uterus and is refractory to chemotherapy.49 In summary, single-agent chemotherapy with either methotrexate or actinomycin D produces nearly 100% cure in patients with nonmetastatic disease. Approximately 85–90% of patients are cured by the initial chemotherapy regimen, with most of the remaining patients being placed into permanent remission with an alternate single agent. Rarely will patients require multiagent chemotherapy or surgery for cure.
Low-risk Metastatic Disease Patients with low-risk metastatic disease (FIGO stages II and III, WHO score <7) can usually be treated successfully with initial single-agent chemotherapy regimens. Most often, these consist of 5-day dosage schedules of methotrexate or actinomycin D, repeated every 2 weeks. If resistance or excessive toxicity develops with the initial single agent, treatment should be continued with another single agent. Resistance to two sequential single-agent chemotherapeutics requires combination chemotherapy with high-risk disease protocols. Hysterectomy may be performed as adjuvant treatment coincident with the institution of chemotherapy to shorten the duration of therapy or to eradicate persistent, chemotherapy-resistant disease in the uterus.49,73 Three centers specializing in the management of gestational trophoblastic disease in the United States have reported on outcomes in patients with low-risk metastatic disease undergoing treatment with either methotrexate or actinomycin D. Primary remission was achieved in 50 to 67% of patients with the initial single agent. Drug resistance developed in 30–50% of patients, necessitating a change to an alternate single agent. Toxicity was a less frequent reason to change therapy regimens. Eventually, 1–14% of patients needed multiagent chemotherapy with or without surgery to achieve remission. Characteristics that emerged as common among patients who failed initial single-agent chemotherapy were pretherapy hCG levels >100 000 mIU mL−1 , age >35, WHO score >4, and large vaginal metastases.73 – 75 In summary, single-agent chemotherapy with the 5-day dosing schedules of methotrexate or actinomycin D is the
539
preferred treatment for patients with low-risk metastatic GTN. With appropriate classification of low-risk patients and proper administration of chemotherapy, cure rates should approach 100%, and the greater morbidity associated with the use of multiagent chemotherapy avoided in most patients.
MULTIAGENT CHEMOTHERAPY High-risk Metastatic Disease In patients categorized as having high-risk metastatic GTN (FIGO stage IV or WHO score ≥ 7), initial treatment should consist of combination chemotherapy with or without adjuvant surgery or radiation therapy. Throughout the 1970s and 1980s, the primary multidrug regimen used was MAC (methotrexate, actinomycin D, and cyclophosphamide or chlorambucil), yielding cure rates of 63–71%.76 In 1982, Bagshawe’s group from the Charing Cross Hospital reported an 82% primary remission rate using the seven-drug CHAMOCA regimen, consisting of cyclophosphamide, hydroxyurea, actinomycin D, methotrexate with folinic acid, vincristine, and doxorubicin.77 However, in a randomized clinical trial comparing MAC and CHAMOCA for the primary treatment of high-risk patients, the Gynecologic Oncology Group found a higher cure rate and lower toxicity with the MAC regimen.78 After the discovery that etoposide was an effective agent for treatment of GTN, Bagshawe and Newlands formulated the EMA-CO regimen: etoposide, high-dose methotrexate with folinic acid, actinomycin D, cyclophosphamide, and vincristine (see Table 3).79 They originally reported a primary remission rate of 80% and an overall survival of 82% with minimal toxicity. Since then, many treatment centers around the world have reported complete response and long-term survival rates of >80%, making EMA-CO the preferred initial multidrug treatment regimen for patients with high-risk metastatic GTN.1 Updated studies on experience with EMA-CO have continued to show the success of this regimen, with complete remissions ranging between 71 and 91% and overall survival rates of 84–94%.80 – 82 The use of EMA without the CO component has been shown to yield similar results.83,84 Primary treatment of high-risk GTN using cisplatin/etoposide containing combinations has also been reported,85,86 but this often Table 3 EMA-CO chemotherapy regimen for high-risk gestational trophoblastic neoplasia.
Day
Drug
Dosing
1
Etoposide Actinomycin D Methotrexate
2
Etoposide Actinomycin D Folinic acid
8
Cyclophosphamide Vincristine
100 mg m−2 iv over 30 min 0.5 mg i.v. push 100 mg m−2 i.v. push, then 200 mg m−2 i.v. in 500 cc D5W over 12 h 100 mg m−2 i.v. over 30 min 0.5 mg i.v. push 15 mg i.m. or p.o. every 12 h for four doses beginning 24 h after the start of methotrexate 600 mg m−2 i.v. infusion 1.0 mg m−2 i.v. push
Repeat cycle on days 15, 16, and 22 (every 2 weeks).
540
GYNECOLOGICAL CANCERS
results in significant cumulative toxicity before a complete response is achieved and may compromise the ability to deliver adequate salvage chemotherapy. With high-risk metastatic disease, chemotherapy is continued for at least two to three courses after the first normal hCG. Adjuvant surgical procedures and radiation therapy are used in combination with chemotherapy in select patients. Hysterectomy and thoracotomy may be of use in resecting foci of chemotherapy-resistant disease. Surgery can also be useful in controlling metastatic tumor hemorrhage, relieving urinary or bowel obstruction, or treating infection. For brain metastases, both whole brain irradiation and steriotactic irradiation as well as intrathecal methotrexate in combination with systemic chemotherapy have been used, with cure rates of 50–80%.1
Secondary Treatment of High-risk Disease Despite multimodal primary therapy, approximately 20% of high-risk patients will fail first-line combination chemotherapy and up to 13% will develop recurrence after initial remission. Most of these patients have multiple metastases to sites other than the lung and vagina, and many have had inadequate chemotherapy. Salvage chemotherapy with drug regimens employing etoposide with platinum agents often combined with surgical resection of sites of persistent tumor will result in the cure of most of these high-risk patients with resistant disease.87 The EMA-EP regimen, substituting etoposide and cisplatin for cyclophosphamide and vincristine in the EMA-CO protocol, is considered the most appropriate therapy for patients who have responded to EMA-CO but have plateauing low hCG levels or who have developed reelevation of hCG levels after a complete response to EMA-CO. Newlands et al. reported an 88% cure rate in this population.88 In high-risk patients who have clearly developed resistance to methotrexate-containing protocols, drug combinations containing etoposide and platinum with bleomycin or ifosfamide have been found to be effective. The BEP (bleomycin, etoposide, cisplatin), VIP (etoposide, ifosfamide, cisplatin), and ICE (ifosfamide, carboplatin, and etoposide) protocols have all produced cures in some patients who have failed EMA regimens.87 Other approaches to the management of recurrent or refractory high-risk trophoblastic neoplasia include increasing the doses of already-proven chemotherapeutic agents in conjunction with peripheral stem cell support or using drugs that are currently being evaluated for activity against resistant GTN, such as paclitaxel, the topoisomerase I inhibitor topotecan, and the antimetabolite gemcitabine.1 In summary, intensive multimodality therapy with EMACO or some variation of it, along with adjuvant radiotherapy or surgery when indicated, results in cure rates of 80 to 90% in patients with high-risk metastatic GTN. Approximately 30% of high-risk patients will fail first-time therapy or relapse from remission. Salvage chemotherapy with drug regimens containing platinum agents, etoposide, and bleomycin or ifosfamide, often in conjunction with surgical resection of sites of persistent tumor, will result in the cure of most of these high-risk patients with resistant disease. Colonystimulating factors should be used to prevent treatment delays
and dose reductions. Newer anticancer drugs, such as paclitaxel and gemcitabine, or high-dose chemotherapy may have a role in the future management of select patients.
TREATMENT OF PLACENTAL-SITE TROPHOBLASTIC TUMORS Hysterectomy is the treatment of choice for PSTTs. Chemotherapy should be used in patients with metastatic PSTT and may be considered in patients with nonmetastatic disease with adverse prognostic factors. Although the optimum chemotherapy regimen for patients with PSTT remains to be defined, the current clinical impression is that the EMA-EP regimen is the treatment of choice. Advanced FIGO stage, long interval from last known pregnancy to diagnosis (>2 years), and high mitotic count (>5 mitoses per 10 hpfs) are adverse prognostic indicators for survival. The survival rate is approximately 100% for nonmetastatic disease and 50–60% for metastatic disease.27,88 – 90
FOLLOW-UP AFTER TREATMENT FOR GESTATIONAL TROPHOBLASTIC NEOPLASIA After hCG remission has been achieved, patients with GTN should undergo serial serum quantitative hCG determinations at 1-month intervals until monitoring has shown 1 year of normal hCG levels. The risk of recurrence after 1 year of remission is less than 1%. Physical exams are performed at 6–12-month intervals. Contraception should be maintained during treatment and for 1 year after the completion of chemotherapy; oral contraceptives are the preferred method. Because of the 1–2% risk of a second gestational trophoblastic disease event in subsequent pregnancies, early ultrasound is recommended for all future pregnancies, careful review of the products of conception or placenta should occur in subsequent pregnancies, and a serum hCG level should be measured 6 weeks after the completion of any pregnancy. The ability to successfully treat GTN has led to a large number of women who maintain their reproductive potential after chemotherapy. Many successful pregnancies have been reported subsequent to treatment for trophoblastic neoplasia. In general, these women do not experience an increase in the incidence of abortions, stillbirths, congenital anomalies, prematurity, or other obstetrical complications of pregnancies following the diagnosis of and treatment for GTN.91 Because many cancer drugs are known carcinogens, there is concern that the chemotherapy used to induce long-term remissions or cures of one cancer might induce second malignancies. There were no reports of increased susceptibility to the development of other malignancies after successful chemotherapy for GTN until recent reports of an increased risk of secondary malignancies (acute myelogenous leukemia, colon cancer, melanoma, and breast cancer) in women who received prolonged combination chemotherapy containing etoposide.92
SUMMARY GTN should be classified according to the 2002 FIGO stage: WHO prognostic scoring system into nonmetastatic, low-risk
GESTATIONAL TROPHOBLASTIC DISEASES
metastatic and high-risk metastatic categories. Nonmetastatic postmolar GTN (FIGO stage I) can be treated with a variety of single-agent methotrexate or actinomycin D protocols, resulting in the cure of essentially all patients. Metastatic low-risk GTN (FIGO stages II and III: WHO score <7) and nonmetastatic choriocarcinoma should be treated with 5-day dosage schedules of methotrexate or actinomycin D, with cure rates approaching 100%. Metastatic high-risk GTN (FIGO stage IV: WHO score ≥7) requires combination chemotherapy with EMA-CO with or without adjuvant radiation therapy and surgery, often with additional salvage chemotherapy employing drug regimens containing platinum agents and etoposide usually in conjunction with bleomycin (BEP) or ifosfamide (VIP, ICE), to achieve cure rates of 80–90%. Using this approach to management of GTN, the overall cure rate for these tumors should exceed 90% with reproductive function being preserved in most women.
REFERENCES 1. Lurain JR. Pharmacotherapy of gestational trophoblastic disease. Expert Opin Pharmacother 2003; 11: 2005 – 17. 2. Bracken MB. Incidence and aetiology of hydatidiform mole: an epidemiological review. Br J Obstet Gynaecol 1987; 94(12): 1123 – 35. 3. Palmer JR. Advances in the epidemiology of gestational trophoblastic disease. J Reprod Med 1994; 39(3): 155 – 62. 4. Atrash HK, Hogue CJR, Grimes DA. Epidemiology of hydatidiform mole during early gestation. Am J Obstet Gynecol 1986; 154(4): 906 – 9. 5. Bagshawe KD, Dent J, Webb J. Hydatidiform mole in England and Wales 1973-83. Lancet 1986; 2(8508): 673 – 7. 6. Takeuchi S. Incidence of gestational trophoblastic disease by regional registration in Japan. Hum Reprod 1987; 2(8): 729 – 34. 7. Smith HO. Gestational trophoblastic disease epidemiology and trends. Clin Obstet Gynecol 2003; 46(3): 541 – 56. 8. Parazzini F, La Vecchia C, Pampallona S. Parental age and risk of complete and partial hydatidiform mole. Br J Obstet Gynaecol 1986; 93(6): 582 – 5. 9. Sand PK, Lurain JR, Brewer JI. Repeat gestational trophoblastic disease. Obstet Gynecol 1984; 63(2): 140 – 4. 10. Berkowitz RS, et al. Gestational trophoblastic disease. Subsequent pregnancy outcome, including repeat molar pregnancy. J Reprod Med 1998; 43(1): 81 – 6. 11. Parazzini F, et al. Risk factors for gestational trophoblastic disease: a separate analysis of complete and partial hydatidiform moles. Obstet Gynecol 1991; 78(6): 1039 – 45. 12. Berkowitz RS, et al. Risk factors for complete molar pregnancy from a case-control study. Am J Obstet Gynecol 1985; 152(8): 1016 – 20. 13. Parazzini F, et al. Dietary factors and risk of trophoblastic disease. Am J Obstet Gynecol 1988; 158(1): 93 – 9. 14. Brinton LA, Bracken MB, Connelly RR. Choriocarcinoma incidence in the United States. Am J Epidemiol 1986; 123(6): 1094 – 100. 15. Smith HO, et al. Trends in gestational choriocarcinoma: a 27-year perspective. Obstet Gynecol 2003; 102(5 Pt 1): 978 – 87. 16. Moore KL, Persaud TVN. The Developing Human – Clinically Oriented Embryology. Philadelphia, Pennsylvania: W.B. Saunders, 1993. 17. Szulman AE, Surti U. The syndromes of hydatidiform mole. I. Cytogenetic and morphologic correlations. Am J Obstet Gynecol 1978; 131(6): 665 – 71. 18. Szulman AE, Surti U. The syndromes of hydatidiform mole. II. Morphologic evolution of the complete and partial mole. Am J Obstet Gynecol 1978; 132(1): 20 – 7. 19. Berkowitz RS, Goldstein DP, Bernstein MR. Evolving concepts of molar pregnancy. J Reprod Med 1991; 36(1): 40 – 4. 20. Lage JM, et al. A flow cytometric study of 137 fresh hydropic placentas: correlation between types of hydatidiform moles and nuclear DNA ploidy. Obstet Gynecol 1992; 79(3): 403 – 10.
541
21. Mosher R, et al. Complete hydatidiform mole. Comparison of clinicopathologic features, current and past. J Reprod Med 1998; 43(1): 21 – 7. 22. Paradinas FJ, et al. A clinical, histopathological and flow cytometric study of 149 complete moles, 146 partial moles and 107 non-molar hydropic abortions. Histopathology 1996; 28(2): 101 – 10. 23. Driscoll SG. Gestational trophoblastic neoplasia: surgical pathologic considerations with clinical emphasis. Clin Obstet Gynecol 1984; 27(1): 160 – 71. 24. Kurman RJ, Scully RE, Norris HJ. Trophoblastic pseudotumor of the uterus: an exaggerated form of “syncytial endometritis” simulating a malignant tumor. Cancer 1976; 38(3): 1214 – 26. 25. Scully RE, Young RH. Trophoblastic pseudotumor: a reappraisal. Am J Surg Pathol 1981; 5(1): 75 – 6. 26. Feltmate CM, et al. Advances in the understanding of placental site trophoblastic tumor. J Reprod Med 2002; 47(5): 337 – 41. 27. Papadopoulos AJ, et al. Twenty-five years’ clinical experience with placental site trophoblastic tumors. J Reprod Med 2002; 47(6): 460 – 4. 28. Newlands ES, et al. Management of placental site trophoblastic tumors. J Reprod Med 1998; 43(1): 53 – 9. 29. Fukunaga M, Ushigome S. Malignant trophoblastic tumors: immunohistochemical and flow cytometric comparison of choriocarcinoma and placental site trophoblastic tumors. Hum Pathol 1993; 24(10): 1098 – 106. 30. Kohorn EI. Molar pregnancy: presentation and diagnosis. Clin Obstet Gynecol 1984; 27(1): 181 – 91. 31. Curry SL, et al. Hydatidiform mole: diagnosis, management, and longterm follow-up of 347 patients. Obstet Gynecol 1975; 45(1): 1 – 8. 32. Berkowitz RS, Goldstein DP. Presentation and management of molar pregnancy. In Hancock W, Newlands ES, Berkowitz RS (eds) Gestational Trophoblastic Diseases, 2nd ed. London, Chapman & Hall, 2003: 206 – 228. 33. Montz FJ, Schlaerth JB, Morrow CP. The natural history of theca lutein cysts. Obstet Gynecol 1988; 72(2): 247 – 51. 34. Santos-Ramos R, Forney JP, Schwartz BE. Sonographic findings and clinical correlations in molar pregnancy. Obstet Gynecol 1980; 56(2): 186 – 92. 35. Hershman JM. Human chorionic gonadotropin and the thyroid: hyperemesis gravidarum and trophoblastic tumors. Thyroid 1999; 9(7): 653 – 7. 36. Twiggs LB, Morrow CP, Schlaerth JB. Acute pulmonary complications of molar pregnancy. Am J Obstet Gynecol 1979; 135(2): 189 – 94. 37. Soto-Wright V, et al. The changing clinical presentation of complete molar pregnancy. Obstet Gynecol 1995; 86(5): 775 – 9. 38. Berkowitz RS, Goldstein DP, Bernstein MR. Natural history of partial molar pregnancy. Obstet Gynecol 1985; 66(5): 677 – 81. 39. Czernobilsky B, Barash A, Lancet M. Partial moles: a clinicopathologic study of 25 cases. Obstet Gynecol 1982; 59(1): 75 – 7. 40. Szulman AE, Surti U. The clinicopathologic profile of the partial hydatidiform mole. Obstet Gynecol 1982; 59(5): 597 – 602. 41. Benson CB, et al. Sonographic appearance of first trimester complete hydatidiform moles. Ultrasound Obstet Gynecol 2000; 16(2): 188 – 91. 42. Fine C, et al. Sonographic diagnosis of partial hydatidiform mole. Obstet Gynecol 1989; 73(3 Pt 1): 414 – 8. 43. Genest D, et al. A clinicopathologic study of 153 cases of complete hydatidiform mole (1980 – 1990): histologic grade lacks prognostic significance. Obstet Gynecol 1991; 78(3 Pt 1): 402 – 9. 44. Cole LA, Butler S. Detection of hCG in trophoblastic disease. The USA hCG reference service experience. J Reprod Med 2002; 47(6): 433 – 44. Review. 45. Cole LA. hCG, its free subunits and its metabolites. Roles in pregnancy and trophoblastic disease. J Reprod Med 1998; 43(1): 3 – 10. 46. Berkowitz RS, et al. Human chorionic gonadotropin and free subunits’ serum levels in patients with partial and complete hydatidiform moles. Obstet Gynecol 1989; 74(2): 212 – 6. 47. Ozturk M, et al. Differential production of human chorionic gonadotropin and free subunits in gestational trophoblastic disease. Am J Obstet Gynecol 1988; 158(1): 193 – 8. 48. Tidy JA, et al. Gestational trophoblastic disease: a study of mode of evacuation and subsequent need for treatment with chemotherapy. Gynecol Oncol 2000; 78(3 Pt 1): 309 – 12.
542
GYNECOLOGICAL CANCERS
49. Soper JT. Surgical therapy for gestational trophoblastic disease. J Reprod Med 1994; 39(3): 168 – 74. 50. Kashimura Y, et al. Prophylactic chemotherapy for hydatidiform mole: 5 – 15 years follow-up. Cancer 1986; 58: 624 – 9. 51. Kim DS, et al. Effects of prophylactic chemotherapy for persistent trophoblastic disease in patients with complete hydatidiform mole. Obstet Gynecol 1986; 67(5): 690 – 4. 52. Lurain JR, et al. Natural history of hydatidiform mole after primary evacuation. Am J Obstet Gynecol 1983; 145(5): 591 – 5. 53. Goto S, et al. Development of postmolar trophoblastic disease after partial molar pregnancy. Gynecol Oncol 1993; 48(2): 165 – 70. 54. Deicas RE, et al. The role of contraception in the development of postmolar gestational trophoblastic tumor. Obstet Gynecol 1991; 78(2): 221 – 6. 55. Curry SL, et al. Hormonal contraception and trophoblastic sequelae after hydatidiform mole (a Gynecologic Oncology Group Study). Am J Obstet Gynecol 1989; 160(4): 805 – 9. 56. Lurain JR. Gestational trophoblastic tumors. Semin Surg Oncol 1990; 6(6): 347 – 53. 57. Lurain JR, Brewer JI. Invasive mole. Semin Oncol 1982; 9(2): 174 – 80. 58. Lathrop JC, et al. Clinical characteristics of placental site trophoblastic tumor (PSTT). Gynecol Oncol 1988; 31(1): 32 – 42. 59. Finkler NJ. Placental site trophoblastic tumor: diagnosis, clinical behavior and treatment. Reprod Med 1991; 36(1): 27 – 30. 60. Bagshawe KD, Harland S. Immunodiagnosis and monitoring of gonadotropin-producing metastases in the central nervous system. Cancer 1976; 38(1): 112 – 8. 61. Schlaerth JB, Morrow CP, Rodriguez M. Diagnostic and therapeutic curettage in gestational trophoblastic disease. Am J Obstet Gynecol 1990; 162(6): 1465 – 70; discussion 1470-1. 62. Hammond CB, et al. Treatment of metastatic trophoblastic disease: good and poor prognosis. Am J Obstet Gynecol 1973; 115(4): 451 – 7. 63. Bagshawe KD. Risk and prognostic factors in trophoblastic neoplasia. Cancer 1976; 38(3): 1373 – 85. 64. Kohorn EI. Negotiating a staging and risk factor scoring system for gestational trophoblastic neoplasia. A progress report. J Reprod Med 2002; 47(6): 445 – 50. 65. Lurain JR, et al. Prognostic factors in gestational trophoblastic tumors. Am J Obstet Gynecol 1991; 164: 611 – 6. 66. Kohorn EI. The new FIGO 2000 staging and risk factor scoring system for gestational trophoblastic disease: description and critical assessment. Int J Gynecol Cancer 2001; 11(1): 73 – 7. 67. Hertz R. Interference with estrogen-induced tissue growth in the chick genital tract by a folic acid antagonist. Science 1948; 107: 300 – 2. 68. Lurain JR, Elfstrand EP. Single-agent methotrexate chemotherapy for the treatment of non-metastatic gestational trophoblastic tumors. Am J Obstet Gynecol 1995; 172(2 Pt 1): 574 – 9. 69. Bagshawe KD, et al. The role of low-dose methotrexate and folinic acid in gestational trophoblastic tumours (GTT). Br J Obstet Gynaecol 1989; 96(7): 795 – 802. 70. Homesley HD, et al. Rapid escalation of weekly intramuscular methotrexate for nonmetastatic gestational trophoblastic disease: a Gynecologic Oncology Group study. Gynecol Oncol 1990; 39(3): 305 – 8. 71. Garrett AP, et al. Methotrexate infusion and folinic acid as primary therapy for nonmetastatic and low-risk metastatic gestational trophoblastic tumors. 15 years of experience. J Reprod Med 2002; 47(5): 355 – 62. 72. Osathanondh R, Goldstein DP, Pastorfide GB. Actinomycin D as the primary agent for gestational trophoblastic disease. Cancer 1975; 36(3): 863 – 6.
73. Dubeshter B, et al. Management of low-risk metastatic gestational trophoblastic tumors. J Reprod Med 1991; 36(1): 36 – 9. 74. Soper JT, et al. 5-day methotrexate for women with metastatic gestational trophoblastic disease. Gynecol Oncol 1994; 54(1): 76 – 9. 75. Roberts JP, Lurain JR. Treatment of low-risk metastatic gestational trophoblastic tumors with single-agent chemotherapy. Am J Obstet Gynecol 1996; 174(6): 1917 – 23. 76. Lurain JR. Advances in management of high-risk gestational trophoblastic tumors. J Reprod Med 2002; 47: 451 – 9. 77. Begent RHJ, Bagshawe KD. The management of high-risk choriocarcinoma. Semin Oncol. 1982; 9(2): 198 – 203. 78. Curry SL, et al. A prospective randomized comparison of methotrexate, dactinomycin, and chlorambucil versus methotrexate, dactinomycin, cyclophosphamide, doxorubicin, melphalan, hydroxyurea, and vincristine in “poor prognosis” metastatic gestational trophoblastic disease: a Gynecologic Oncology Group study. Obstet Gynecol 1989; 73(3 Pt 1): 357 – 62. 79. Newlands ES, et al. 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 1991; 98(6): 550 – 7. 80. Bower M, et al. EMA/CO for high-risk gestational trophoblastic tumors: results from a cohort of 272 patients. J Clin Oncol 1997; 15(7): 2636 – 43. Erratum in: J Clin Oncol 1997 Sep;15(9):3168. 81. Kim SJ, et al. Risk factors for the prediction of treatment failure in gestational trophoblastic tumors treated with EMA/CO regimen. Gynecol Oncol 1998; 71(2): 247 – 53. 82. Escobar PF, et al. Treatment of high-risk gestational trophoblastic neoplasia with etoposide, methotrexate, actinomycin D, cyclophosphamide and vincristine chemotherapy. Gynecol Oncol 2003; 91(3): 552 – 7. 83. Soto-Wright V, et al. The management of gestational trophoblastic tumors with etoposide, methotrexate, and actinomycin D. Gynecol Oncol 1997; 64: 156 – 9. 84. Matsui H, et al. Combination chemotherapy with methotrexate, etoposide, and actinomycin D for high-risk gestational trophoblastic tumors. Gynecol Oncol 2000; 78: 28 – 31. 85. Theodore c, et al. Treatment of high-risk gestational trophoblastic disease with chemotherapy combinations containing cisplatin and etoposide. Cancer 1989; 64: 1824 – 9. 86. Surwit EA, Childers JM. High-risk metastatic gestational trophoblastic disease. A new dose-intensive, multiagent chemotherapeutic regimen. J Reprod Med 1991; 36(1): 45 – 8. 87. Lurain JR, Nejad B. Secondary chemotherapy for high-risk gestational trophoblastic neoplasia. Gynecol Oncol 2005; 97(2): 618 – 23. 88. Newlands ES, et al. 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 2000; 18(4): 854 – 9. 89. Newland ES, et al. Management of placental site trophoblastic tumors. J Reprod Med 1998; 43(1): 53 – 9. 90. Hoekstra AV, Keh P, Lurain JR. Placental site trophoblastic tumor: a review of 7 cases and their implications for prognosis and treatment. J Reprod Med 2004; 49(6): 447 – 52. 91. Berkowitz RS, et al. Subsequent pregnancy experience in patients with gestational trophoblastic disease. New England Trophoblastic Disease Center, 1965 – 1992. J Reprod Med 1994; 39(3): 228 – 32. 92. Rustin GJS, et al. Combination but not single-agent methotrexate chemotherapy for gestational trophoblastic tumors increases the incidence of second tumors. J Clin Oncol 1996; 10: 2769 – 73.
Section 8 : Hematological Malignancies
48
Rare Leukemias
Attaphol Pawarode and Maria R. Baer
INTRODUCTION While all leukemias are uncommon, the leukemias chosen for inclusion in this chapter are those seen infrequently even by physicians specializing in leukemia in cancer centers and tertiary referral centers. The leukemias included in this chapter are shown in bold in Table 1 in the context of the World Health Organization (WHO) classification of myeloid and lymphoid neoplasms.1 Some are rare primary leukemias (e.g. chronic neutrophilic leukemia [CNL], acute megakaryocytic leukemia, hairy cell leukemia [HCL]), while others are rare primary leukemic presentations of neoplasms that do not usually present as leukemias (e.g. mantle cell leukemia, plasma cell leukemia [PCL]).
RARE MYELOPROLIFERATIVE DISEASES Chronic Neutrophilic Leukemia CNL is a myeloproliferative disorder (MPD) characterized by neutrophilic differentiation. Reported cases of CNL have likely also included leukemoid reactions to occult infections or malignancies and neutrophilic chronic myelogenous leukemia (CML-n), a recently described MPD characterized by the presence of the breakpoint cluster region/Abelson (BCR/ABL) p230 transcript rather than the classic p190 and p210 characteristic of Philadelphia chromosome-positive (Ph+ ) ALL (acute lymphoblastic leukemia) and CML, respectively. A critical review of case reports through the year 2001 found only 33 cases of true CNL.2 Information about pathogenesis is limited. CNL appears to arise in granulocyte-committed progenitors that are capable of spontaneous proliferation.3 Resistance of CNL neutrophils to apoptosis has been suggested.4 CNL is characterized by a persistent absolute neutrophilia, usually ≥25 × 109 L−1 , with normal hemoglobin and platelet counts.2 There is a male preponderance (2 : 1).2 The median age in an adult series from the Mayo Clinic was 66 years (range 54–86 years),5 but CNL is also seen in young adults and in children. Presenting symptoms included weight loss, fatigue, and pruritus, but not fever. All patients in this series had splenomegaly, with spleen tips palpable at a median of 4.5 cm below the left costal margin.
In the Mayo Clinic series,5 the median leukocyte and absolute neutrophil counts were 67.9 × 109 L−1 and 62.2 × 109 L−1 , respectively. The peripheral blood smear in CNL shows mature neutrophils, few immature cells, and no blasts, in contrast to the “left shift” seen in Ph+ CML. Neither basophilia nor eosinophilia is present. The neutrophils often contain toxic granules and/or D¨ohle bodies, while erythrocytes and platelets are morphologically normal. The leukocyte alkaline phosphatase (LAP) score is increased in CNL, in contrast to the low score in CML. Abnormalities in other laboratory values include elevated uric acid and serum vitamin B12 levels, as seen in other MPDs, and a low granulocyte colonystimulating factor (G-CSF) level, in contrast to the elevated level in leukemoid reactions. The bone marrow is hypercellular because of infiltration by neutrophils mainly ranging from metamyelocytes to mature segmented cells, without increased blasts. Other lineages appear normal. In contrast, expansion of all three lineages and of all levels of granulocytes is seen in Ph+ CML. Dysplasia, maturation arrest, and significant marrow fibrosis are absent in CNL, but a mild degree of reticulin fibrosis may be present and may differentiate CNL from a leukemoid reaction.5 Cytogenetic and molecular studies demonstrate absence of the t(9;22) and the BCR/ABL gene rearrangement that defines Ph+ CML. A normal karyotype is most commonly found in CNL, but clonal chromosome abnormalities include +8, +9, del(20q) and del(11q), all of which are also found in other myeloid neoplasms and thus are not characteristic of CNL. Histologic examination of the spleen and liver shows neutrophil infiltration of the splenic red pulp and the hepatic sinusoids and/or portal triads. Although only 20% of cases progress to acute myeloid leukemia (AML), the median survival time for CNL is only 30 months, and the 5-year survival rate 28%.2 Treatment is generally initiated to control neutrophilia, organ infiltration, and associated symptoms. Hydroxyurea is the usual first-line treatment and is very effective in controlling neutrophilia, splenomegaly and symptoms, but the disease eventually progresses and requires second-line therapy. Second-line treatments, including low-dose cytarabine, 2-chlorodeoxyadenosine (2-CdA, cladribine), 6-thioguanine
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
544
HEMATOLOGICAL MALIGNANCIES
Table 1 WHO classification of neoplastic diseases of the hematopoietic and lymphoid tissues.
Myeloproliferative diseases Chronic myelogenous leukemia, Philadelphia chromosome-positive (t(9;22)(qq34;q11), BCR/ABL) Chronic neutrophilic leukemia Chronic eosinophilic leukemia/hypereosinophilic syndrome Chronic idiopathic myelofibrosis Polycythemia vera Essential thrombocythemia Myeloproliferative disease, unclassifiable Myelodysplastic/myeloproliferative diseases Chronic myelomonocytic leukemia Atypical chronic myelogenous leukemia Juvenile myelomonocytic leukemia Myelodysplastic syndromes Refractory anemia With ringed sideroblasts Without ringed sideroblasts Refractory cytopenia (myelodysplastic syndrome) with multilineage dysplasia Refractory anemia (myelodysplastic syndrome) with excess blasts 5q syndrome Myelodysplastic syndrome, unclassifiable Acute myeloid leukemias AMLs with recurrent cytogenetic translocations AML with t(8;21)(q22;q22), AML1(core binding factor α (CBF-α)/eight-twenty-one (ETO)) Acute promyelocytic leukemia (AML with t(15;17)(q22;q11-12) and variants, promyelocytic leukemia/retinoic acid (PML/RAR)-α) AML with abnormal bone marrow eosinophils (inv(16)(p13q22) or t(16;16)(p13;q11), CBF-ß/MYH11X) AML with 11q23 (mixed lineage leukemia (MLL)) abnormalities AML with multilineage dysplasia With prior myelodysplastic syndrome Without prior myelodysplastic syndrome AML and myelodysplastic syndromes, therapy related Alkylating agent related Epipodophyllotoxin related (some may be lymphoid) Other types AML not otherwise categorized AML minimally differentiated AML without maturation AML with maturation Acute myelomonocytic leukemia Acute monocytic leukemia Acute erythroid leukemia Acute megakaryocytic leukemia Acute basophilic leukemia Acute panmyelosis with myelofibrosis B cell neoplasms Precursor B cell neoplasm Precursor B-lymphoblastic leukemia/lymphoma (precursor B cell acute lymphoblastic leukemia) Mature (peripheral) B cell neoplasms B cell chronic lymphocytic leukemia/small lymphocytic lymphoma B cell prolymphocytic leukemia Lymphoplasmacytic lymphoma Splenic marginal zone B cell lymphoma (+/− villous lymphocytes) Hairy cell leukemia Plasma cell myeloma/plasmacytoma Extranodal marginal zone B cell lymphoma of mucosa-associated lymphoid tissue (MALT) type Nodal marginal zone B cell lymphoma (+/− monocytoid B cells) Follicular lymphoma Mantle cell lymphoma Diffuse large B cell lymphoma Mediastinal large B cell lymphoma
Table 1 (continued).
Primary effusion lymphoma Burkitt’s lymphoma/Burkitt cell leukemia T cell and NK cell neoplasms Precursor T cell neoplasm Precursor T-lymphoblastic lymphoma/leukemia (precursor T cell acute lymphoblastic leukemia) Mature (peripheral) T cell neoplasms T cell prolymphocytic leukemia T cell granular lymphocytic leukemia Aggressive NK cell leukemia Adult T cell lymphoma/leukemia (HTLV1+) Extranodal NK/T cell lymphoma, nasal type Enteropathy-type T cell lymphoma Hepatosplenic γ -λ T cell lymphoma Subcutaneous panniculitis-like T cell lymphoma Mycosis fungoides/Sezary syndrome Anaplastic large cell lymphoma, T/null cell, primary cutaneous type Peripheral T cell lymphoma, not otherwise characterized Angioimmunoblastic T cell lymphoma Anaplastic large cell lymphoma, T/null cell, primary systemic type Hodgkin’s lymphoma (Hodgkin’s disease) Nodular lymphocyte-predominant Hodgkin’s lymphoma Classical Hodgkin’s lymphoma Nodular sclerosis Hodgkin’s lymphoma (grades 1 and 2) Lymphocyte-rich classical Hodgkin’s lymphoma Mixed cellularity Hodgkin’s lymphoma Lymphocyte depletion Hodgkin’s lymphoma Mast cell disease (mastocytosis) Cutaneous mastocytosis Indolent systemic mastocytosis Systemic mastocytosis with associated clonal, hematologic nonmast cell lineage disease Aggressive systemic mastocytosis Mast cell leukemia Mast cell sarcoma Extracutaneous mastocytoma NK, natural killer
(6-TG), and interferon-α (IFN-α), have yielded short-lived responses. The standard AML induction has resulted in prolonged cytopenias and death.2,5,6 Interestingly, a recent case report demonstrated clinical and cytogenetic remissions in patients with refractory CNL with t(15;19)(q13;p13.3) who had received imatinib mesylate, an ABL, plateletderived growth factor receptor (PDGFR), and KIT tyrosine kinase inhibitor; the mechanism is unknown and should be explored.7 Splenectomy is indicated for symptomatic resistant splenomegaly, but may worsen neutrophilia. Allogeneic hematopoietic stem cell transplantation following myeloablative therapy has been successful in patients of appropriate age with suitable donors.6,8 Transplantation is indicated in suitable patients due to the high incidence of secondary refractory disease, the progression to AML in 20%, and the short median survival. However, the optimal timing of transplantation has not been defined.
Chronic Eosinophilic Leukemia and Hypereosinophilic Syndrome Chronic erythrophilic leukemia (CEL) is a clonal MPD characterized by ≥1.5 × 109 L−1 circulating eosinophils
RARE LEUKEMIAS
and <20% myeloblasts in the blood and marrow.9 Idiopathic hypereosinophilic syndrome (HES) is characterized by ≥1.5 × 109 L−1 circulating eosinophils for >6 months and end-organ damage.10 HES is associated with the recently discovered FIP1L1-PDGFRA fusion gene, which is also present in a subgroup of CEL.11 CEL and HES comprise a heterogeneous spectrum of indolent-to-aggressive diseases and must be distinguished from other hematologic disorders associated with eosinophilia, either as a part of the malignant clone (CML, systemic mastocytosis, other MPDs, and some subtypes of AML) or as a reactive component because of cytokine production (Hodgkin’s disease, T cell non-Hodgkin’s lymphoma [NHL], and ALL).12 Brito-Babapulle categorized eosinophilia into reactive (nonclonal), clonal, and HES, which is a diagnosis of exclusion.13 The stem cell of origin may be pluripotent, multipotent, or a committed eosinophil precursor.9 In both CEL and HES, the blood shows mature eosinophils with few eosinophil precursors, which may look atypical or dysplastic. Neutrophilia and, less frequently, mild basophilia and monocytosis may be present. The marrow is hypercellular with various stages of eosinophils, frequently with Charcot-Leyden crystals. Eosinophils in both conditions have cyanide-resistant myeloperoxidase positivity.8 In a series by Simon et al.,14 approximately 25% of patients with HES had circulating T cells with an aberrant phenotype, half of which had a clonal T cell receptor (TCR) gene rearrangement. These abnormal T cells secreted large amounts of interleukin (IL)-5 in vitro. Roufosse et al.15 detected clonal T cells with increased in vitro secretion of IL-4 and IL-5 in another series. Most HES cases have a normal karyotype, but reported changes include the reciprocal translocations t(3;4)(p13;q12), t(4;7)(q11;q32), and t(4;7)(q11;13).12 The recently discovered constitutive gain-of-function fusion gene FIP1L1PDGFRA, generated by an interstitial deletion of chromosome 4q12, is present in a subgroup, indicating that other molecular mechanisms exist.11 The FIP1L1-PDGFRA fusion gene was present in 17% of cases by reverse transcription polymerase chain reaction (RT-PCR) and fluorescence in situ hybridization (FISH) analysis in a recent series.16 The TCR gene rearrangement was also noted in 31%,16 and in 47% by FISH in another series.17 Confirmation of clonality is difficult because most cases have a normal karyotype, not all have the molecular marker of FIP1L1-PDGFRA fusion gene, and the X-inactivation assay is not useful in males. CEL and HES have a male-to-female ratio of 9 : 1. Most patients are between 20 and 50 years of age. Ninety percent are symptomatic with clinical manifestations reflecting the effects of eosinophilic infiltration and of cytokines and humoral factors released from eosinophilic granules on diverse organs including the heart (Loeffler’s endomyocarditis, endomyocardial fibrosis, restrictive cardiomyopathy, mural thrombosis), lungs (pulmonary infiltrates, emboli, fibrosis, effusions), liver, spleen, skin (urticaria, angioedema, papules/nodules), gastrointestinal tract, and nervous system (thromboemboli, peripheral neuropathy, eosinophilic meningitis).12,13 Five- and 15-year survival rates of 80 and 42% have been reported in HES.18 Causes of death include
545
organ failure and transformation into more aggressive malignancies. In the past, treatment was aimed at alleviating or reversing end-organ damage by decreasing the number of circulating eosinophils.12,13 Corticosteroids, cytotoxic agents such as hydroxyurea, cyclophosphamide, vincristine, chlorambucil, and etoposide have short-term effects, whereas IFN-α induces long-term remissions by inhibition of eosinophil proliferation, differentiation, and secretion and by inhibition of the secretion of IL-5 by T cells. Imatinib mesylate, an ABL, PDGFR, and KIT tyrosine kinase inhibitor was recently found to yield complete response rates of 40–82%, with only 56–80% of responders carrying the FIP1L1-PDGFRA fusion gene, implicating other imatinib-responsive mechanisms.11,12,19 In the largest series, 10 of 11 patients treated with imatinib achieved hematologic response, with a 4-week median time to response (range, 1–12 weeks), and 9 had sustained responses, with a median duration of 7 months (range, 3–15 months).11 Elevated serum tryptase levels identify a subgroup of patients with an imatinib-responsive myeloproliferative variant of HES characterized by splenomegaly, increased serum vitamin B12 level, end-organ fibrosis, and an aggressive course.20 Imatinib resistance mediated by the PDGFRA T674I mutation, which is analogous to the T351I mutation in imatinib-resistant BCR-ABL+ CML, is overcome in a mouse model by the tyrosine kinase inhibitor PKC412, which inhibits FMS-like tyrosine kinase 3 (FLT3), protein kinase C (PKC), kinase insert domain-containing receptor (KDR), KIT, PDGFRα, and PDGFRβ.21 Imatinib is currently the first-line treatment for patients with and without the FIP1L1-PDGFRA+ disease, with or without increased serum tryptase levels. Pending availability of other tyrosine kinase inhibitors that may be effective in imatinib-resistant disease, IFN-α may be used for patients who fail imatinib therapy. Allogeneic transplantation should be considered for patients with aggressive disease.
RARE ACUTE MYELOID LEUKEMIAS Acute Erythroblastic Leukemia Acute erythroblastic leukemia is divided into two subtypes in the WHO classification: (i) erythroleukemia, defined by ≥50% erythroid precursors in marrow nucleated cells and ≥20% myeloblasts in nonerythroid cells; and (ii) pure erythroid leukemia, also known as DiGuglielmo disease, acute erythremic myelosis, true erythroleukemia, and minimally differentiated erythroleukemia, defined by ≥80% immature erythroid cells in the marrow nucleated cell population, without a significant myeloblastic component.22 Erythroleukemia likely arises in multipotential stem cells with varying degrees of erythroid and myeloid maturation, and pure erythroid leukemia arises in primitive stem cells committed to the erythroid lineage (BFU-E/CFU-E).22 Acute erythroblastic leukemia is subtype M6 in the FrenchAmerican-British (FAB) classification, defined by ≥50% erythroid precursors in marrow nucleated cells and ≥30% myeloblasts in nonerythroid cells. FAB M6 was further subdivided into M6a, corresponding to the traditional FAB M6
546
HEMATOLOGICAL MALIGNANCIES
category, M6b, which is pure erythroleukemia, and M6c, in which myeloblasts and pronormoblasts are each >30% of marrow cells.23 The incidence of acute erythroblastic leukemia peaks in younger (<20 years) and in older (>50 years) patients. Patients present with profound anemia, thrombocytopenia, and varying degrees of neutropenia. Hepatomegaly or hepatosplenomegaly may be present.24 – 26 The blood shows dysplastic erythroblasts, nucleated red blood cells, schistocytes, teardrop and pincered red cells, basophilic stippling, as well as myeloblasts, pseudo-Pelger-Huet cells, hypogranular neutrophils, and giant and hypogranular platelets.22 – 25 Marrow findings include dysplastic erythroid precursors with megaloblastoid and/or bi- or multinucleated nuclei (see Figure 1a). The cytoplasm may contain vacuoles that may coalesce and stain with periodic acid-Schiff (PAS) in a blocklike pattern (see Figure 1b). Ringed sideroblasts may be present. Myeloblasts are typical of those found in AML, and may contain Auer rods. Multilineage dysplasia is often
(a)
(b)
(c)
(d)
seen.22 – 25 At least one lineage is dysplastic in 100%, and two lineages in 86%.27 In erythroleukemia, erythroblasts stain with antibodies to glycophorin A and hemoglobin A (see Figure 1c), and do not express myeloid antigens. In pure erythroid leukemia, more mature erythroblasts stain for glycophorin A and hemoglobin A, while less mature forms express carbonic anhydrase 1 and CD36, the Gerbich blood group.22 Chromosome abnormalities are detected in three-fourths of cases of acute erythroblastic leukemia, and karyotypes are often complex. Chromosomes 5 and/or 7 are involved in two-thirds of cases.22 – 27 Acute erythroid leukemia responds poorly to AML therapy. Adverse prognostic factors include a high proerythroblastto-myeloblast ratio in the bone marrow, a high proliferative index, unfavorable karyotypes, expression of the multidrug resistance (MDR) protein P-glycoprotein (Pgp) and FAB M6b disease.28,29 Patients with normal versus abnormal karyotypes had complete remission (CR) rates of 73 versus 42%,27 and patients with chromosomes 5 and/or 7 abnormalities had median survivals of 16 versus 77 weeks for those without these chromosome anomalies.25 A Mayo Clinic series reported similar short survival in patients with erythroleukemia and pure erythrocytic leukemia (medians, 6 and 4 months).26 M6a and M6c had higher remission rates and longer survival with standard AML therapy, compared with M6b (medians, 31.4, 10.5, and 3.15 months), and M6b frequently expressed Pgp.23 Of note, a 95% sustained CR rate was reported in young patients with de novo disease, who mostly (85%) received bone marrow transplantation as a consolidation therapy, and the median survival of 2.9 years was similar to that of matched control non-M6 AML patients.30 In summary, acute erythroid leukemia is an unfavorable subtype of AML, though presence of a normal karyotype defines a small more favorable subgroup. Given the poor response to standard AML therapies, experimental therapies are justified, and enrollment on clinical trials is strongly encouraged. Allogeneic hematopoietic stem cell transplantation is indicated in eligible patients.
Acute Megakaryoblastic Leukemia
(e)
(f)
Figure 1 Rare acute myeloid leukemias. Pure erythroid leukemia (erythremic myelosis, AML FAB M6): (a) Marrow smear with over 80% dysplastic erythroid precursors (Wright-Giemsa stain, ×250). (b) Corresponding PAS stain showing marked granular and diffuse cytoplasmic reactivity (PAS stain, ×250). (c) Corresponding bone marrow biopsy showing islands of hemoglobin-positive erythroid precursors (Section immunohistochemistry, ×100). Acute megakaryoblastic leukemia (AML FAB M7): (d) Marrow smears with clusters of megakaryoblasts. (e) Micromegakaryoblasts with shedding of platelets (Wright-Giemsa stain, ×250). (f) Corresponding marrow biopsy section with sinusoidal sheets of megakaryoblasts (CD61 immunohistochemistry, B5-fixative, ×100) (Courtesy of Dr Maurice Barcos, Department of Pathology, Roswell Park Cancer Institute, Buffalo, NY).
Acute megakaryoblastic leukemia, defined as acute leukemia in which ≥50% of marrow blasts are megakaryoblasts,22 occurs in children and in older adults.31 – 33 The blasts are generally 12–18 µ in size, with basophilic cytoplasm that may have blebs and pseudopods and fine nuclear chromatin with one to three nucleoli (see Figure 1d), but they may be smaller and mimic lymphoblasts. Extensive marrow fibrosis may be present. The blood may show dysplastic neutrophils and giant platelets with hypogranulation. In addition to megakaryoblasts, fragments, and micromegakaryocytes may be detected (see Figure 1e). The blasts do not stain with Sudan Black B or myeloperoxidase, but may stain with PAS, acid phosphatase, or nonspecific esterase. Electron microscopy demonstrates platelet peroxidase in nuclear membranes and endoplasmic reticulum. Megakaryoblasts typically express CD41 [glycoprotein (gp) IIb/IIIa], CD61
RARE LEUKEMIAS
(gp IIIa), and CD36, and less frequently CD42 (gp Ib), and the myeloid markers CD13 and CD33 (see Figure 1f). They do not express CD34, CD45, or HLA-DR. CD7 may be expressed, but other lymphoid markers are not.22 The only specific chromosomal abnormality in acute megakaryoblastic leukemia is t(1;22)(p13;p13) in infancy.34 A large series (30 children and 23 adults) showed chromosomal abnormalities in 94%.32 These were more frequently complex in adults than in children (58.5 vs 38.5%). The most common recurring abnormalities were t(1;22) and +8 in children and t(9;22), 3q21q26 changes, including inv(3)(q21q26), chromosomes 5 and/or 7 abnormalities and i(12)(p10), which is associated with mediastinal germ cell tumors.35 The inv(3)(q21q26) is also found in other AML subtypes, usually with thrombocytosis. Mutations of the GATA1 gene on chromosome X and overexpression of the ERG gene on chromosome 21 have been found in acute megakaryoblastic leukemia.36,37 In children, acute megakaryoblastic leukemia is associated with both Down’s syndrome and the t(1;22)(p13;p13) chromosomal translocation and associated OTT-MAL fusion gene.32 The first group has a better prognosis, while the latter often presents with hepatosplenomegaly and aggressive disease that is refractory to induction chemotherapy. Children with t(1;22) or without Down’s syndrome fare poorly. Adult patients present with cytopenias and circulating megakaryoblasts. The reported median leukocyte count was 4.6 × 109 L−1 .32 Thrombocytosis may be present. Hepatosplenomegaly is uncommon in adults, and associated mediastinal germ cell tumors have been described in young men.35 Treatment outcome is poor in adults, with a reported CR rate of 33–43% and median survival of 4–5.4 months.32,33 Given the poor response to standard AML therapies, experimental therapies are justified, and enrollment on clinical trials is strongly encouraged. Allogeneic hematopoietic stem cell transplantation is also justified in appropriate patients.
Acute Basophilic Leukemia There are no definite criteria for diagnosis of this extremely rare entity. It includes both exclusive basophilic lineage involvement and multilineage involvement with basophilic predominance, which is associated with t(9;22) and may represent occult CML with basophilic blastic transformation.22 Basophilic myeloblasts are medium-sized, with an oval, round, and sometimes bilobed nucleus containing dispersed chromatin and one to three nucleoli. The cytoplasm is basophilic and contains variably irregular coarse basophilic granules characteristically staining with the metachromatic stain toluidine blue. Vacuoles may be seen. Erythroid precursors may show dysplastic changes. Electron microscopy shows structures characteristic of either basophil precursors or mast cells, corresponding to the basophilic granules seen by light microscopy.22,38 Blasts typically express CD34, HLA-DR, the myeloid antigens CD13 and CD33, and the specific markers CD9, CDw17, CD88, and CD 25, and do not express lymphoid markers.22,39 There are no specific cytogenetic abnormalities. Reported cytogenetic findings include normal karyotypes and t(9;22);22,38,40 – 42 while del(12)(p1113), 6q23 translocation, and t(6;9)(p23;q34) are associated
547
with other types of AML with bone marrow basophilia, rather than acute basophilic leukemia.22,41,42 Patients range from infant to elderly age-groups and present with cytopenias and circulating blasts with classical morphology.22,38,40 – 42 Some patients have a histaminerelease syndrome, which may include urticaria, hyperpigmentation, peptic ulcers, and anaphylactic shock.38 The liver and/or spleen may be enlarged. In a review of the literature between 1982 and 2002,41 most adult patients died within 1 year, while children fared somewhat better. There are no specific therapies. Given the poor outcome in adults, allogeneic hematopoietic stem cell transplantation is indicated when feasible.
Acute Panmyelosis with Myelofibrosis This very rare clinical entity is characterized by acute onset of severe pancytopenia, with a hypercellular dysplastic marrow with increased blasts and varying degrees of fibrosis.22,43 It is thought to arise from myeloid stem cells, with fibrotic proliferation as an epiphenomenon.22 Adults are mainly affected, but childhood cases have been reported. It can be de novo or therapy related.22,44,45 Patients usually present with acute onset of severe pancytopenia. In contrast to chronic idiopathic myelofibrosis, constitutional symptoms are common and splenomegaly is absent or minimal.22,43 The blood shows a leukoerythroblastic picture with minimal peripheral blasts, and, in contrast to chronic idiopathic myelofibrosis, teardrop poikilocytosis is absent or minimal. Dysgranulopoiesis and abnormal platelets are often seen. The bone marrow frequently cannot be aspirated due to the fibrosis. The bone marrow biopsy is hypercellular, with varying degrees of fibrosis, mostly reticulin rather than collagen. Multilineage dysplasia and increased blasts are present. Blasts express hematopoietic stem cell and early myeloid antigens, and less frequently, erythroid and/or megakaryocytic antigens in subpopulations.22,43 There are no specific cytogenetic abnormalities, although complex karyotypes with chromosomes 5 and/or 7 aberrations may be seen. The clinical course is very aggressive, and the median survival was only 2 months in a recent series43 and 9 months in another series, which may have included other entities.44 Accurate diagnosis is important. Since treatment outcome is poor, patients should be enrolled on clinical trials. Allogeneic hematopoietic stem cell transplantation should be considered, though fibrosis is a risk factor for engraftment failure.
RARE B CELL LEUKEMIAS Hairy Cell Leukemia HCL is an indolent chronic B cell lymphoproliferative disorder that accounts for 2% of adult leukemia. HCL was described in 1923 by Ewald, and then in 1958 as “leukemic reticuloendotheliosis” by Bouroncle, et al.46 The term hairy cell leukemia was first introduced by Schrek and Donnelly in 1966, emphasizing the characteristic microscopic feature.47
548
HEMATOLOGICAL MALIGNANCIES
(a)
(b)
(c)
(d)
(e) Figure 2 Rare B cell leukemias. Hairy cell leukemia: (a) Marrow smear showing classic hairy cells with abundant cottonlike cytoplasm with occasional vacuoles and round, oval, or reniform nuclei (Wright-Giemsa stain, ×250). (b) Corresponding bone marrow biopsy showing characteristic “fried egg” appearance of the leukemic cells with clear cytoplasm separating the nuclei (H&E, ×100). Plasma cell leukemia: (c) Peripheral blood smear with a preponderance of neoplastic plasma cells (Wright-Giemsa stain, ×250). Mantle cell lymphoma/leukemia: (d) Marrow biopsy section with extensive lymphomatous infiltrate (hematoxylin and eosin stain, ×80). (e) Corresponding lymph node biopsy section showing nuclear reactivity for cyclin-D1 (bcl-1) (immunohistochemistry, formalin fixative, ×80) (Courtesy of Dr Maurice Barcos, Department of Pathology, Roswell Park Cancer Institute, Buffalo, NY).
Hairy cells (HCs) are mature, but not terminally differentiated, B cells. They have a slightly eccentric round or ovoid nucleus with dispersed or stippled chromatin and indistinct nucleoli. The moderately abundant pale blue-gray cytoplasm (see Figure 2a) may contain a rod-shaped inclusion corresponding to the ribosomal lamella complex as seen by electron microscopy.48 – 50 The cytoplasmic membrane contains numerous irregular fine villi and broader ruffles up to 3 µm in length. Staining with tartrate-resistance acid phosphatase (TRAP) is specific for HCL.51 Peripheral blood and bone marrow are universally involved in HCL, though in cases with severe leukopenia, buffy coat preparations may be required to demonstrate hairy cells.51 Normochromic normocytic anemia, monocytopenia, and thrombocytopenia of varying degrees are routinely present. The bone marrow is hypercellular and diffusely replaced with HCs with an increase in reticulin fibers and eventual fibrosis.46,48,52,53 The abundant cytoplasm shrinks in biopsy sections, resulting in widely separated nuclei, yielding the classic “fried egg” appearance (see Figure 2b).
Extravasated erythrocytes in merging dilated sinusoids give rise to the typical angiomatous vascular lake. The spleen is involved in 80% of cases.53 HCs infiltrate and replace red pulp sinusoidal endothelial cells, forming congested pseudosinusoids that eventually merge into vascular lakes. The white pulp may also be infiltrated and is often atrophic. Hepatic sinusoid and periportal infiltrates are common.46,48,52,53 HCs typically express the B cell antigens CD19, CD20, CD22, CD79a, CD40, FMC7 and surface immunoglobulin, but not CD5, CD10, CD21, CD23, or CD79b. They also express the mucosal lymphocyte antigen CD103, the myelomonocytic marker CD11c and the IL-2 receptor CD25.46,48,52,53 There are no specific cytogenetic abnormalities,48 but multiple partially related clones with chromosomes 14 and 5 abnormalities may be present.46,53 In a variant form of HCL, morphology is a hybrid of classic HCL and B-prolymphocytic leukemia.54 Leukocytosis and lymphadenopathy are common, and the variant HCs do not stain with TRAP, and express CD11c, but generally not CD103 or CD25. A Japanese variant with polyclonal hairy lymphocytes, weakly TRAP+ staining, and CD25− has been described.55 HCL is mostly diagnosed in middle-aged Caucasians, with a 4 : 1 male preponderance. The median age is approximately 50 years. The classic presentation includes infections, splenomegaly, pancytopenia, circulating hairy cell lymphocytes, and inaspirable bone marrow.46,52,53 Ninety percent have splenomegaly, which may be massive.56 Symptoms may include discomfort, pain, and/or early satiety. Hepatomegaly and lymphadenopathy are infrequent. Twenty-five percent of patients are asymptomatic and are diagnosed incidentally by routine blood counts. Laboratory findings include pancytopenia in most cases and severe neutropenia in 30%.53 Fifty percent have pancytopenia, and the rest have involvement of at least one lineage. Monocytopenia is common. Approximately 20% of patients present with a leukemic phase that needs to be differentiated from the HCL variant.53 Dysregulated T cell function, neutropenia, and monocytopenia contribute to development of opportunistic infections. Organisms include Legionella pneumophila, mycobacteria (especially Mycobacterium kansasii ), fungi (candida, cryptococcus, aspergillus, histoplasma), and parasites (toxoplasma).46,52,53 Autoimmune phenomena or other immune derangements may occur, including polyclonal and monoclonal gammopathies, vasculitides (leukocytoclastic type, polyarteritis nodosa, and temporal arteritis), cryoglobulinemia, glomerulonephritis, autoimmune hemolytic anemia, idiopathic thrombocytopenic purpura, rheumatoid arthritis (RA), and virtually the entire spectrum of other autoimmune disorders.46,53 The risk of secondary malignancies following treatment is of special concern. A recent large series of patients who survived at least 7 years with a median follow-up of 108 months demonstrated an increase in incidence of secondary malignancies (observed-to-expected ratio = 2.03, 95% CI 1.49–2.71), the majority of which (95%) were solid tumors.57
RARE LEUKEMIAS
HCL is indolent, and asymptomatic patients do not require therapy. Indications for treatment include cytopenia(s) (Hb <10 g dL−1 , absolute neutrophil count <1 × 109 L−1 or platelets <100 × 109 L−1 ), recurrent infections, high circulating leukemic cells, symptomatic splenomegaly, and associated autoimmune phenomena, such as vasculitis.46 Several therapies are effective for HCL. Splenectomy rapidly reverses cytopenias in most patients, but the median response duration is only 8 months.58,59 IFN-α therapy, 3 million units m−2 subcutaneously daily, was first reported in 1984, with overall and complete response rates of 38–93 and 7–30%, respectively.46 Cytopenias resolve at medians of 2 to 5 months, and median response durations are 18–25 months.60 Nucleoside analogs effective in HCL include 2 -deoxycoformycin (pentostatin) and 2CdA.46,52,53,61 Pentostatin is generally administered at a dose of 4 mg m−2 every other week for 3–6 months. Overall and complete responses occur in 79–97% and 33–87% of patients and are durable.46 However, adverse reactions include nausea and vomiting, fever, photosensitivity, keratoconjunctivitis, and prolonged immunosuppression leading to opportunistic infections. 2-CdA is administered as a single continuous intravenous infusion course of 0.1 mg−1 kg−1 day−1 for 7 days. Overall and complete response rates are 89–100% and 75–91%, respectively, and responses are durable.46 Adverse reactions are neutropenia and neutropenic and opportunistic infections. Other therapies have been reported. Thomas et al.62 used anti-CD20 monoclonal antibody (Rituximab) in 15 patients with relapsed/refractory HCL, with overall and complete response rates of 80 and 53%. Kreitman et al.63 demonstrated impressive efficacy of the recombinant immunotoxin LMB-2, which consists of the Fv portion of the anti-CD25 antibody fused to a truncated form of Pseudomonas exotoxin A. 2-CdA is the current first-line therapy for patients with HCL. Patients with late relapses may be salvaged with the same agent, with high response rates.64 Refractory disease may be treated with pentastatin. IFN-α is indicated in patients with active infections or nucleoside analog-resistant disease. Rituximab and conjugated anti-CD25 immunotoxin are also active. Splenectomy may be used as described above. Follow-up should include monitoring for secondary malignancies.
Primary Plasma Cell Leukemia PCL accounts for 2–4% of plasma cell dyscrasias (see Chapter 50, Uncommon Presentations of Plasma Cell Dyscrasias).65 – 68 Primary plasma cell leukemia (PPCL), accounting for 60% of PCL,66 arises without preexisting multiple myeloma (MM) and is a distinct clinicobiologic entity,67 while secondary PCL is an end-stage manifestation of MM. PPCL is defined by the presence of ≥2 × 109 L−1 circulating plasma cells in the absence of prior known MM.66,69 It presents with bone pain, hepatosplenomegaly, marked anemia and thrombocytopenia, and renal failure.66,68,70 Circulating cells include mature plasma cells, plasmablasts, and anaplastic plasma cells (see Figure 2c). Bone marrow universally has greater than 30% involvement of plasma cells.
549
Plasma cell infiltration of liver and spleen is common, and involvement of other organs, including lymph nodes and myocardium, has been reported.65,66 In one series,67 PPCL and MM did not differ in median age, sex distribution, and extent of bone involvement, but PPCL had a higher prevalence of Salmon-Durie stage III disease, extramedullary disease, Bence Jones proteinuria, renal failure, hypercalcemia, severe anemia, thrombocytopenia, high serum β-2 microglobulin, C-reactive protein, elevated LDH levels and proportion of S-phase plasma cells, but a lower prevalence of immunoglobulin A (IgA) disease. Patients with PPCL had mean survival shorter than all MM patients but similar to that of poor-risk MM patients (8 vs 36 and 13 months). Combination chemotherapy is significantly more effective than melphalan and prednisone in controlling PPCL (18 vs 3 months). Favorable response to thalidomide and bortezomib has been reported.71,72 Consolidation with high-dose therapy, mainly melphalan, and stem cell rescue in selected cases has been associated with relapse-free survivals of 11 to 36 months.73,74 Allogeneic transplantation is less frequently employed because of high treatment-related mortality. The current optimal treatment approach consists of induction with a conventional chemotherapy regimen such as VAD (vincristine, adriamycin, and dexamethasone) or a regimen containing novel agents such as thalidomide or bortezomib, followed by high-dose therapy and autologous stem cell rescue. Effective palliation may be achieved with thalidomide or bortezomib with or without dexamethasone or with dexamethasone alone. Bisphosphonates such as pamidronate and zoledronic acid are an important adjuvant for the prevention and palliation of bone disease and hypercalcemia. Like MM and secondary PCL, PPCL is an incurable disease, and enrollment on a clinical trial is warranted.
Mantle Cell Leukemia Mantle cell lymphoma is a distinct clinico-biological subtype of B cell NHL arising from peripheral B lymphocytes of the lymph node inner mantle zone and characterized by an aggressive nature and short survival.75 – 77 Cyclin D1 is constitutively overexpressed as a result of fusion of the CCND1 gene to the IgH gene promoter in the t(11;14)(q13;q32) universally present in this unique NHL subtype. Mantle cell lymphoma affects middle-aged to elderly population with male predominance, and typically manifests with B symptoms, widespread lymphadenopathy, and extranodal disease including hepatosplenomegaly, involvement of the gastrointestinal tract, and frequent marrow (see Figure 2d,e) and blood involvement.78 Mantle cell leukemia, a rare variant, is defined as ≥4 × 109 L−1 absolute lymphocyte count in a patient with mantle cell lymphoma.79 In one series,80 only 14.3% of patients with blood involvement fulfilled the criterion for mantle cell leukemia. Almost all patients with mantle cell leukemia present without prior known mantle cell lymphoma. A large Royal Marsden series classified mantle cell leukemia into small cell, typical, and blastoid types, representing 15, 46, and 39% of cases.81 Small cells are smallto medium-sized lymphocytes with a regular or slightly
550
HEMATOLOGICAL MALIGNANCIES
indented nucleus, condensed chromatin and scant-to-medium amount of cytoplasm, resembling small lymphocytes in chronic lymphocytic leukemia/small lymphocytic lymphoma. Typical mantle cells are small- to medium-sized cells with a slightly-to-markedly irregular indented nucleus with speckled chromatin and an indistinct nucleolus, resembling centrocytes. The blastoid type is defined by ≥20% large cells with reticular chromatin and one or two nucleoli or medium-sized cells with dispersed chromatin, resembling lymphoblasts. Mantle cells typically express surface IgM and/or sIgD, CD5, CD20, CD22, CD43, bcl-1 and bcl-2, but not CD23 or CD10, but atypical mantle cells lack CD5 expression.82 Cytogenetic analysis and FISH reveal t(11;14)(q13;q32) and the CCND1/IgH fusion gene in 75% and almost 100%, respectively.81 The immunoglobulin heavy and light chain genes are rearranged, but variable region genes are germline in most cases.82 The Royal Marsden group reported the largest clinicopathologic study of mantle cell leukemia (n = 58) with ≥5 × 109 L−1 clonal absolute B lymphocytosis.81 Median age was 62 years, with a male-to-female ratio of 2.4 : 1. Common presenting manifestations included splenomegaly (74%), anemia (55%), lymphadenopathy (45%), hepatomegaly (17%); 10% were asymptomatic. The median absolute lymphocyte count was 58 × 109 L−1 . The typical CD5+ CD23− immunophenotype was detected in 68%, and atypical CD5− CD23− and CD5− CD23+ in 17 and 15%, respectively. Ninety-two percent were cyclin D1+ by immunohistochemistry and/or flow cytometry. The t(11;14) was detected by conventional analysis in 100% of tested cases and by FISH analysis of peripheral blood in 83%. Almost half had complex karyotypes, mostly involving chromosomes 1, 17q, 12, and 17p. The splenic and blastoid variants of mantle cell leukemia have been increasingly reported.79,83,84 The former is characterized by splenomegaly without lymphadenopathy and responds well to splenectomy.83 In contrast, blastoid disease has a very poor prognosis, with a median survival of 3 months.84 Data on treatment of mantle cell leukemia are limited. In mantle cell lymphoma, purine analogs (fludarabine or cladribine) with or without anthracycline or anthracenedione (idarubicin or mitoxantrone) resulted in overall and complete response rates of 33–100% and 0–44%, with response durations of 6–24 months.75,76 More intensive regimens followed by high-dose therapy and autologous stem cell rescue have yielded 3-year survival rates as high as 72 and 92%. In particular, sequential intensive high-dose chemotherapy, in vivo purging with rituximab, autologous stem cell transplantation, and posttransplant maintenance with rituximab have led to durable CRs in two patients with primary refractory mantle cell leukemia.85 Allogeneic stem cell transplantation has induced long-term remissions, even in relapses after autologous transplantation. Targeted small molecules such as the proteasome inhibitor bortezomib, thalidomide plus rituximab, the cyclin inhibitor flavopiridol, or the mammalian target of rapamycin (mTOR) inhibitor CCI-779 have shown encouraging clinical responses in mantle cell lymphoma, but no data are available for mantle cell leukemia.75,86
As with other rare leukemias there is no standard treatment, and patients should participate in clinical trials. Splenectomy is effective for the splenic variant. Young patients or patients with refractory disease and good performance status are eligible for high-dose chemotherapy and stem cell rescue. Incorporation of rituximab into treatment regimens is reasonable, and mTOR inhibitor therapy may hold promise.
RARE T CELL AND NATURAL KILLER (NK) CELL LEUKEMIAS T-Large Granular Lymphocyte Leukemia T-large granular lymphocyte (T-LGL) leukemia was first described in the 1970s.87 Brouet et al. reported T cell chronic lymphocytic leukemia (T-CLL) with circulating lymphocytes resembling LGLs, and McKenna et al. reported circulating LGLs with chronic neutropenia. Loughran proposed the term T cell LGL leukemia in 1993. T-LGLs are antigen-activated cytotoxic T lymphocytes expressing CD3, CD8, and CD57 which are mostly in the G0/G1 phase of the cell cycle.87,88 Accumulation of cells is due to dysregulation of Fas-mediated apoptosis. The most salient clinical feature in T-LGL leukemia is neutropenia, which may be due to multiple mechanisms including antibody-mediated peripheral destruction, decreased neutrophil survival due to increased Fas ligand-mediated apoptosis and suppression of hematopoiesis by lymphokines including IFN-γ and tumor necrosis factor (TNF)-α.87 LGLs are medium to large cells with abundant cytoplasm, round to oval eccentric nuclei, low nuclear –cytoplasmic ratio, and numerous azurophilic granules (see Figure 3), corresponding to membrane-bound electron-dense granules as seen by electron microscopy.87 T-LGLs characteristically express CD3, CD8, and CD57, but not CD4, CD5, CD7, CD25, or CD56.87,88 Expression of CD16 is variable. CD52 is also expressed. Most T-LGL cases have a normal karyotype. Clonality is confirmed by southern blot or polymerase chain-reaction analysis of TCR gene rearrangements.
Figure 3 Large granular lymphocyte (LGL) leukemia. Peripheral smear with leukemic LGL showing coarse cytoplasmic granules (Wright-Giemsa stain, ×250) (Courtesy of Dr Maurice Barcos, Department of Pathology, Roswell Park Cancer Institute, Buffalo, NY).
RARE LEUKEMIAS
Marrow involvement may be diffuse or nodular, and the degree of LGL infiltration does not correlate with the degree of cytopenias. Rarely, the bone marrow appears normal or is aplastic, in which case flow cytometry is helpful in establishing diagnosis. Granulopoiesis is usually normal but maturation arrest has been reported.89 LGLs may infiltrate the splenic red pulp cords and hepatic sinusoids, but lymph node involvement is very rare. T-LGL patients have a median age of 55 to 65 years, with equal sex distribution.87 The diagnosis of LGL leukemia should be considered in patients with increased circulating LGL and/or unexplained neutropenia. Patients may present with neutropenia-related infections, mostly bacterial, but approximately 25% are asymptomatic. Absolute LGL counts range from 0.1–50 × 109 L−1 (median 4 × 109 L−1 ). Neutropenia is present in 85%, and is severe in 50%.90 Anemia and thrombocytopenia are present in 20–50 and 20% respectively. The proposed diagnostic criteria for T-LGL leukemia include presence of ≥2 × 109 L−1 cells with LGL morphology in blood smears with a CD3+ CD8+ CD57+ immunophenotype or presence of ≥0.5 × 109 L−1 of these cells with clonal T CRβ rearrangement.87 Various immune processes, attributed to dysregulated B cell function, are associated with T-LGL leukemia. Clinical findings may include RA, pure red cell aplasia, autoimmune hemolytic anemia, and idiopathic thrombocytopenic purpura.87,88,90 Laboratory abnormalities include presence of rheumatoid factor, antinuclear antibodies, antineutrophil antibodies, polyclonal and monoclonal gammopathies, hypogammaglobulinemia, and elevated β-2 microglobulin. T-LGL leukemia is an indolent disease. Morbidity and mortality are due to neutropenia, rather than progressive lymphoproliferation. A Mayo Clinic series reported an overall survival of 90% at a median follow-up of 44 months and an actuarial median survival of 161 months.91 Treatment of T-LGL leukemia should be individualized. Nearly one-third of patients do not require treatment. Indications for treatment include severe neutropenia, neutropenia with recurrent infections, anemia, and/or thrombocytopenia.87 Treatments have included splenectomy, corticosteroids, cyclophosphamide, hematopoietic growth factors, cyclosporine A (CsA), low-dose methotrexate (MTX), nucleoside analogs, and allogeneic stem cell transplantation. Splenectomy, corticosteroids, and growth factors provide transient and/or partial, if any, improvement in cytopenias, with persistence of the clonal LGL population.87 Splenectomy may increase the number of circulating LGL postoperatively, and corticosteroids are associated with significant toxicity. Pure red cell aplasia associated with T-LGL leukemia may be effectively treated with single-agent oral cyclophosphamide. Successful use of nucleoside analogs including fludarabine, cladribine, and pentostatin has been reported anecdotally. The current best treatments are CsA and MTX. CsA improves neutropenia and anemia in most patients despite persistence of the LGL clone.90 Time to response is 4 to 9 weeks. The initial dose is 1 to 1.5 mg kg−1 orally every 12 hours, with subsequent dose adjustments based on trough
551
levels. The dose may be tapered after attainment of response, but ongoing low-dose maintenance therapy is generally required.90 HLA-DR4 may be a predictor of response to CsA.92 Low-dose oral MTX may induce long-term clinical remissions, with resolution of cytopenias, regression of the leukemic clone, and molecular remissions in some patients.87 MTX is especially effective for T-LGL leukemia with RA,93 resulting in improvement of both rheumatologic and hematologic manifestations. MTX responses are gradual, and maintenance treatment may be needed.
NK-Large Granular Lymphocyte Leukemia This clinical entity occurs primarily in younger patients in Far Eastern Asia, and Epstein-Barr virus (EBV) infection is implicated in its pathogenesis.87 In contrast to TLGL leukemia, NK-LGL leukemia is an aggressive malignancy, with B symptoms, hepatosplenomegaly, visceral organ involvement, and coagulopathy. Neutropenia is less severe than in T-LGL leukemia, but anemia and thrombocytopenia are more common. Autoimmune phenomena are not seen. NK-LGL cells express CD56, CD2, CD7, and CD16, but not CD3, and TCR genes are germline.87 Complex karyotypes may be present, with frequent chromosome 6q abnormalities. There is no standard treatment strategy for NK-LGL leukemia. Chemotherapy is generally ineffective, and most patients die within months of diagnosis. Successful treatment with allogeneic stem cell transplantation has been reported,94,95 and should be considered in appropriate patients.
Mast Cell Leukemia Mast cell leukemia, defined by ≥20% mast cells in the marrow and ≥10% in the blood, is the most aggressive form of systemic mastocytosis.96 It may occur as an initial presentation or as a late manifestation of other types of systemic mastocytosis.97 The marrow is typically diffusely infiltrated by immature mast cells, including blast-like cells with few metachromatic granules and hypogranulated promastocytes with bi- or multilobed nuclei, typically without fibrosis or osteosclerosis.96 – 98 Mast cells stain strongly for histamine and c-KIT and variably with metachromatic and chloroacetate esterase stains, but not with myeloperoxidase. They express CD2, CD25, and CD117 (KIT), the receptor for the stromal cell-derived cytokine stem cell factor that regulates mast cell growth and function. The gain-of-function D816V cKIT mutation is prevalent in benign mastocytosis, but is less frequently present in mast cell leukemia.96,98 Electron microscopy demonstrates multiple surface projections, a lobulated nucleus, and granule-like structures containing tryptase-immunogold reactive material. Mast cell leukemia occurs primarily in adults and runs an aggressive course. Patients generally present with histaminerelease syndrome (abdominal pain, nausea/vomiting, flushing, pruritus, diarrhea, headache, and/or hypotension), constitutional symptoms, hepatosplenomegaly, lymphadenopathy,
552
HEMATOLOGICAL MALIGNANCIES
anemia, and bleeding tendency, but cutaneous mastocytosis is absent.97 – 99 The peripheral blood smear may show a leukoerythroblastic picture. Coagulopathy may be present. Rapidly progressive multiorgan failure may occur, and autopsies show widespread infiltration of organs including the heart, lungs, gastrointestinal tract, and kidneys. Treatment for mast cell leukemia is generally ineffective, and the Mayo Clinic review showed a median survival of <6 months despite combination chemotherapy.99 Successful allogeneic stem cell transplantation has been reported.100 Cladribine is effective in systemic mastocytosis and aleukemic mast cell leukemia,101,102 which is defined by <10% peripheral blood involvement; but its activity in classic mast cell leukemia is unknown. Imatinib mesylate, an ABL, PDGFR, and KIT tyrosine kinase inhibitor, is especially effective in systemic mastocytosis with eosinophilia without the D816V c-KIT or T674I PDGFRA mutations.21,103 PKC412 is effective in acute mast cell leukemia with D816V c-KIT.104 In summary, enrollment on clinical trials incorporating a c-KIT inhibitor is appropriate, and allogeneic stem cell transplantation is the treatment of choice in appropriate patients. Antihistamines and corticosteroids should be given during the treatment for prophylaxis of mediator-release syndrome.
10.
11.
12.
13. 14. 15.
16.
17.
18. 19.
20.
CONCLUSION Knowledge of diagnostic criteria will allow identification of rare leukemias and application of specific management considerations. Patients with rare leukemias should generally be enrolled on clinical trials. Stem cell transplantation appears to have a role in some of these conditions, but may have unexpected patterns of morbidity, and should thus be implemented in specialist centers. Recently developed specific and targeted therapies have significantly improved outcome in some of these diseases.
REFERENCES 1. Jaffe ES, et al. (eds). World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press, 2001. 2. Reilly JT. Chronic neutrophilic leukaemia: a distinct clinical entity? Br J Haematol 2002; 116: 10 – 8. 3. Yanagisawa K, et al. Neoplastic involvement of granulocytic lineage, not granulocytic-monocytic, monocytic, or erythrocytic lineage, in a patient with chronic neutrophilic leukemia. Am J Hematol 1998; 57: 221 – 4. 4. Hara K, et al. Apoptosis resistance of mature neutrophils in a case of chronic neutrophilic leukaemia. Eur J Haematol 2001; 66: 70 – 1. 5. Elliott MA, et al. Chronic Neutrophilic Leukemia (CNL): a clinical, pathologic and cytogenetic study. Leukemia 2001; 15: 35 – 40. 6. Piliotis E, Kutas G, Lipton JH. Allogeneic bone marrow transplantation in the management of chronic neutrophilic leukemia. Leuk Lymphoma 2002; 43: 2051 – 4. 7. Choi IK, et al. Efficacy of imatinib mesylate (STI571) in chronic neutrophilic leukemia with t(15;19): case report. Am J Hematol 2004; 77: 366 – 9. 8. Hasle H, et al. Chronic neutrophil leukaemia in adolescence and young adulthood. Br J Haematol 1996; 94: 628 – 30. 9. Bain B, et al. Chronic eosinophilic leukaemia and the hypereosinophilic syndrome. In Jaffe ES, et al. (eds) World Health Organization Classification of Tumours. Pathology and Genetics of Tumours
21.
22.
23.
24. 25.
26.
27.
28. 29.
30. 31. 32.
33.
of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press, 2001: 29 – 31. Chusid MJ, et al. The hypereosinophilic syndrome: analysis of fourteen cases with review of the literature. Medicine (Baltimore) 1975; 54: 1 – 27. Cools J, et al. A tyrosine kinase created by fusion of the PDGFRA and FIP1L1 genes as a therapeutic target of imatinib in idiopathic hypereosinophilic syndrome. N Engl J Med 2003; 348: 1201 – 14. Gotlib J, et al. The FIP1L1-PDGFRalpha fusion tyrosine kinase in hypereosinophilic syndrome and chronic eosinophilic leukemia: implications for diagnosis, classification, and management. Blood 2004; 103: 2879 – 91. Brito-Babapulle F. The eosinophilias, including the idiopathic hypereosinophilic syndrome. Br J Haematol 2003; 121: 203 – 23. Simon HU, et al. Abnormal clones of T cells producing interleukin-5 in idiopathic eosinophilia. N Engl J Med 1999; 341: 1112 – 20. Roufosse F, et al. Clonal Th2 lymphocytes in patients with the idiopathic hypereosinophilic syndrome. Br J Haematol 2000; 109: 540 – 8. Roche-Lestienne C, et al. Molecular characterization of the idiopathic hypereosinophilic syndrome (HES) in 35 French patients with normal conventional cytogenetics. Leukemia 2005; 19: 792 – 8. Vandenberghe P, et al. Clinical and molecular features of FIP1L1PDFGRA (+) chronic eosinophilic leukemias. Leukemia 2004; 18: 734 – 42. Lefebvre C, et al. Prognostic factors of hypereosinophilic syndrome. Study of 40 cases Ann Med Interne (Paris) 1989; 140: 253 – 7. Stone RM, Gilliland DG, Klion AD. Platelet-derived growth factor receptor inhibition to treat idiopathic hypereosinophilic syndrome. Semin Oncol 2004; 31(2 Suppl 6): 12 – 7. Klion AD, et al. Elevated serum tryptase levels identify a subset of patients with a myeloproliferative variant of idiopathic hypereosinophilic syndrome associated with tissue fibrosis, poor prognosis, and imatinib responsiveness. Blood 2003; 101: 4660 – 6. Cools J, et al. PKC412 overcomes resistance to imatinib in a murine model of FIP1L1-PDGFRalpha-induced myeloproliferative disease. Cancer Cell 2003; 3: 459 – 69. Brunning RD, et al. Acute myeloid leukaemias. In Jaffe ES, et al. (eds) World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press, 2001: 75 – 107. Mazzella FM, et al. Acute erythroleukemia: evaluation of 48 cases with reference to classification, cell proliferation, cytogenetics, and prognosis. Am J Clin Pathol 1998; 110: 590 – 8. Park S, Picard F, Dreyfus F. Erythroleukemia: a need for a new definition. Leukemia 2002; 16: 1399 – 401. Olopade OI, et al. Clinical, morphologic, and cytogenetic characteristics of 26 patients with acute erythroblastic leukemia. Blood 1992; 80: 2873 – 82. Goldberg SL, et al. The erythroid leukemias: a comparative study of erythroleukemia (FAB M6) and Di Guglielmo disease. Am J Clin Oncol 1998; 21: 42 – 7. Davey FR, et al. Morphologic characteristics of erythroleukemia (acute myeloid leukemia; FAB-M6): a CALGB study. Am J Hematol 1995; 49: 29 – 38. Mazzella FM, et al. The acute erythroleukemias. Clin Lab Med 2000; 20: 119 – 37. Kowal-Vern A, et al. Diagnosis and characterization of acute erythroleukemia subsets by determining the percentages of myeloblasts and proerythroblasts in 69 cases. Am J Hematol 2000; 65: 5 – 13. Killick S, et al. Acute erythroid leukemia (M6): outcome of bone marrow transplantation. Leuk Lymphoma 1999; 35: 99 – 107. Gassmann W, Loffler H. Acute megakaryoblastic leukemia. Leuk Lymphoma 1995; 18(Suppl 1): 69 – 73. Dastugue N, et al. Groupe Francais d’Hematologie Cellulaire. Cytogenetic profile of childhood and adult megakaryoblastic leukemia (M7): a study of the Groupe Francais de Cytogenetique Hematologique (GFCH). Blood 2002; 100: 618 – 26. Oki Y, et al. Adult acute megakaryocytic leukemia: an analysis of 37 patients treated at M.D. Anderson Cancer Center. Blood 2005; Aug 25, [Epub ahead of print].
RARE LEUKEMIAS 34. Carroll A, et al. The t(1;22) (p13;q13) is nonrandom and restricted to infants with acute megakaryoblastic leukemia: a Pediatric Oncology Group Study. Blood 1991; 78: 748 – 52. 35. Nichols CR, et al. Hematologic neoplasia associated with primary mediastinal germ-cell tumors. N Engl J Med 1990; 322: 1425 – 9. 36. Stachura DL, Chou ST, Weiss MJ. An early block to erythromegakaryocytic development conferred by loss of transcription factor GATA-1. Blood 2005; Sep 6, [Epub ahead of print]. 37. Rainis L, et al. The proto-oncogene ERG in megakaryoblastic leukemias. Cancer Res 2005; 65: 7596 – 602. 38. Shvidel L, et al. Acute basophilic leukaemia: eight unsuspected new cases diagnosed by electron microscopy. Br J Haematol 2003; 120: 774 – 81. 39. Duchayne E, et al. Diagnosis of acute basophilic leukemia. Leuk Lymphoma 1999; 32: 269 – 78. 40. Peterson LC, et al. Acute basophilic leukemia. A clinical, morphologic, and cytogenetic study of eight cases. Am J Clin Pathol 1991; 96: 160 – 70. 41. Seth T, et al. Acute basophilic leukemia with t(8;21). Leuk Lymphoma 2004; 45: 605 – 8. 42. Giagounidis AA, et al. Acute basophilic leukemia. Eur J Haematol 2001; 67: 72 – 6. 43. Suvajdzic N, et al. Acute panmyelosis with myelofibrosis: clinical, immunophenotypic and cytogenetic study of twelve cases. Leuk Lymphoma 2004; 45: 1873 – 9. 44. Thiele J, et al. Acute panmyelosis with myelofibrosis: a clinicopathological study on 46 patients including histochemistry of bone marrow biopsies and follow-up. Ann Hematol 2004; 83: 513 – 21. 45. Thiele J, Kvasnicka HM, Schmitt-Graeff A. Acute panmyelosis with myelofibrosis. Leuk Lymphoma 2004; 45: 681 – 7. 46. Wentworth N, Saven A. Hairy cell leukemia: Biology, diagnosis, and treatment. In Schiller GJ, (ed) Chronic Leukemias And Lymphomas. Biology, Pathophysiology, and Clinical Management. Totowa, New Jersey: Humana Press, 2003: 55 – 78. 47. Schrek R, Donnelly WJ. “Hairy” cells in blood in lymphoreticular neoplastic disease and “flagellated” cells of normal lymph nodes. Blood 1966; 27: 199 – 211. 48. Foucar K, Catovsky D. Hairy cell leukaemia. In Jaffe ES, et al. (eds) World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press, 2001: 138 – 141. 49. Bartl R, et al. Bone marrow histology in hairy cell leukemia. Identification of subtypes and their prognostic significance. Am J Clin Pathol 1983; 79: 531 – 45. 50. Katayama I. Bone marrow in hairy cell leukemia. Hematol Oncol Clin North Am 1988; 2: 585 – 602. 51. Yam LT, et al. Cytochemistry of tartrate-resistant acid phosphatase: 15 years’ experience. Leukemia 1987; 1: 285 – 8. 52. Goodman GR, Bethel KJ, Saven A. Hairy cell leukemia: an update. Curr Opin Hematol 2003; 10: 258 – 66. 53. Polliack A. Hairy cell leukemia: biology, clinical diagnosis, unusual manifestations and associated disorders. Rev Clin Exp Hematol 2002; 6: 366 – 88. 54. Sainati L, et al. A variant form of hairy cell leukemia resistant to alpha-interferon: clinical and phenotypic characteristics of 17 patients. Blood 1990; 76: 157 – 62. 55. Machii T, et al. Polyclonal B-cell lymphocytosis with features resembling hairy cell leukemia-Japanese variant. Blood 1997; 89: 2008 – 14. 56. Flandrin G, et al. Hairy cell leukemia: clinical presentation and followup of 211 patients. Semin Oncol 1984; 4(Suppl 2): 458 – 71. 57. Goodman GR, et al. Extended follow-up of patients with hairy cell leukemia after treatment with cladribine. J Clin Oncol 2003; 21: 891 – 6. 58. Golomb HM, Vardiman JW. Response to splenectomy in 65 patients with hairy cell leukemia: an evaluation of spleen weight and bone marrow involvement. Blood 1983; 61: 349 – 52. 59. Golde DW. Therapy of hairy-cell leukemia. N Engl J Med 1982; 307: 495 – 6. 60. Ratain MJ, et al. Relapse after interferon alfa-2b therapy for hairycell leukemia: analysis of prognostic variables. J Clin Oncol 1988; 6: 1714 – 21.
553
61. Savoie L, Johnston JB. Hairy cell leukemia. Curr Treat Options Oncol 2001; 2: 217 – 24. 62. Thomas DA, et al. Rituximab in relapsed or refractory hairy cell leukemia. Blood 2003; 102: 3906 – 11. 63. Kreitman RJ, et al. Responses in refractory hairy cell leukemia to a recombinant immunotoxin. Blood 1999; 94: 3340 – 8. 64. Saven A, et al. Long-term follow-up of patients with hairy cell leukemia after cladribine treatment. Blood 1998; 92: 1918 – 26. 65. Hayman SR, Fonseca R. Plasma cell leukemia. Curr Treat Options Oncol 2001; 2: 205 – 16. 66. Blade J, Kyle RA. Nonsecretory myeloma, immunoglobulin D myeloma, and plasma cell leukemia. Hematol Oncol Clin North Am 1999; 13: 1259 – 72. 67. Garcia-Sanz R, et al. Primary plasma cell leukemia: clinical, immunophenotypic, DNA ploidy, and cytogenetic characteristics. Blood 1999; 93: 1032 – 7. 68. Bernasconi C, et al. Plasma cell leukemia: a report on 15 patients. Eur J Haematol 1989; 51: 76 – 83. 69. Grogan TM, et al. Plasma cell neoplasms. In Jaffe ES, et al. (eds) World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press, 2001: 142 – 156. 70. Costello R, et al. Primary plasma cell leukaemia: a report of 18 cases. Leuk Res 2001; 25: 103 – 7. 71. Wohrer S, et al. Effective treatment of primary plasma cell leukemia with thalidomide and dexamethasone-a case report. Hematol J 2004; 5: 361 – 3. 72. Esparis-Ogando A, et al. Bortezomib is an efficient agent in plasma cell leukemias. Int J Cancer 2005; 114: 665 – 7. 73. Saccaro S, et al. Primary plasma cell leukemia: report of 17 new cases treated with autologous or allogeneic stem-cell transplantation and review of the literature. Am J Hematol 2005; 78: 288 – 94. 74. Hovenga S, et al. Consolidation therapy with autologous stem cell transplantation in plasma cell leukemia after VAD, high-dose cyclophosphamide and EDAP courses: a report of three cases and a review of the literature. Bone Marrow Transplant 1997; 20: 901 – 4. 75. Williams ME, Densmore JJ. Biology and therapy of mantle cell lymphoma. Curr Opin Oncol 2005; 17: 425 – 31. 76. Decaudin D. Mantle cell lymphoma: a biological and therapeutic paradigm. Leuk Lymphoma 2002; 43: 773 – 81. 77. Argatoff LH, et al. Mantle cell lymphoma: a clinicopathologic study of 80 cases. Blood 1997; 89: 2067 – 78. 78. Cohen PL, et al. Bone marrow and peripheral blood involvement in mantle cell lymphoma. Br J Haematol 1998; 101: 302 – 10. 79. Dunphy CH, et al. Blastic mantle cell leukemia: a previously undescribed form. J Clin Lab Anal 1999; 13: 112 – 5. 80. Wong KF, et al. Mantle cell lymphoma in leukemic phase: characterization of its broad cytologic spectrum with emphasis on the importance of distinction from other chronic lymphoproliferative disorders. Cancer 1999; 86: 850 – 7. 81. Matutes E, et al. The leukemic presentation of mantle-cell lymphoma: disease features and prognostic factors in 58 patients. Leuk Lymphoma 2004; 45: 2007 – 15. 82. Swerdlow SH, et al. Mantle cell lymphoma. In Jaffe ES, et al. (eds) World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press, 2001: 168 – 170. 83. Angelopoulou MK, et al. The splenic form of mantle cell lymphoma. Eur J Haematol 2002; 68: 12 – 21. 84. Viswanatha DS, et al. Blastic mantle cell leukemia: an unusual presentation of blastic mantle cell lymphoma. Mod Pathol 2000; 13: 825 – 33. 85. Oyan B, Koc Y, Kansu E. Successful salvage with high-dose sequential chemotherapy coupled with in vivo purging and autologous stem cell transplantation in 2 patients with primary refractory mantle cell lymphoma presenting in the leukemic phase. Int J Hematol 2005; 81: 155 – 8. 86. Witzig TE, et al. Phase II trial of single-agent temsirolimus (CCI-779) for relapsed mantle cell lymphoma. J Clin Oncol 2005; 23: 5347 – 56. 87. Benekli M, Baer MR. Large granular lymphocytic proliferative diseases. In Schiller GJ, (ed) Chronic Leukemias And Lymphomas.
554
88.
89.
90.
91.
92.
93.
94.
HEMATOLOGICAL MALIGNANCIES Biology, Pathophysiology, and Clinical Management. Totowa, New Jersey: Humana Press, 2003: 55 – 78. Chan WC, et al. T-cell large granular lymphocyte leukaemia. In Jaffe ES, et al. (eds) World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press, 2001: 197 – 198. Chan WC, et al. A morphologic and immunologic study of the large granular lymphocyte in neutropenia with T lymphocytosis. Blood 1984; 63: 1133 – 40. Sood R, et al. Neutropenia associated with T-cell large granular lymphocyte leukemia: long-term response to cyclosporine therapy despite persistence of abnormal cells. Blood 1998; 91: 3372 – 8. Dhodapkar MV, et al. Clinical spectrum of clonal proliferations of T-large granular lymphocytes: a T-cell clonopathy of undetermined significance? Blood 1994; 84: 1620 – 7. Battiwalla M, et al. HLA-DR4 predicts haematological response to cyclosporine in T-large granular lymphocyte lymphoproliferative disorders. Br J Haematol 2003; 123: 449 – 53. Hamidou MA, et al. Low-dose methotrexate for the treatment of patients with large granular lymphocyte leukemia associated with rheumatoid arthritis. Am J Med 2000; 108: 730 – 2. Murashige N, et al. Allogeneic haematopoietic stem cell transplantation as a promising treatment for natural killer-cell neoplasms. Br J Haematol 2005; 130: 561 – 7.
95. Ebihara Y, et al. Successful treatment of natural killer (NK) cell leukemia following a long-standing chronic active Epstein-Barr virus (CAEBV) infection with allogeneic bone marrow transplantation. Bone Marrow Transplant 2003; 31: 1169 – 71. 96. Valent P, et al. Diagnostic criteria and classification of mastocytosis: a consensus proposal. Leukemia 2001; 25: 603 – 25. 97. Sperr WR, et al. Clinical and biologic diversity of leukemias occurring in patients with mastocytosis. Leuk Lymphoma 2000; 37: 473 – 86. 98. Valent P, et al. Mastocytosis: pathology, genetics, and current options for therapy. Leuk Lymphoma 2005; 46: 35 – 48. 99. Travis WD, et al. Mast cell leukemia: report of a case and review of the literature. Mayo Clin Proc 1986; 61: 957 – 66. 100. Spyridonidis A, et al. Evidence for a graft-versus-mast-cell effect after allogeneic bone marrow transplantation. Bone Marrow Transplant 2004; 34: 515 – 9. 101. Kluin-Nelemans HC, et al. Cladribine therapy for systemic mastocytosis. Blood 2003; 102: 4270 – 6. 102. Penack O, et al. Cladribine therapy in a patient with an aleukemic subvariant of mast cell leukemia. Ann Hematol 2005; 84(10): 692 – 3. 103. Pardanani A, et al. Imatinib for systemic mast-cell disease. Lancet 2003; 362: 535 – 6. 104. Gotlib J, et al. Activity of the tyrosine kinase inhibitor PKC412 in a patient with mast cell leukemia with the D816V KIT mutation. Blood 2005; 106(8): 2865 – 70 .
Section 8 : Hematological Malignancies
49
Rare Lymphomas Graham A. R. Young
INTRODUCTION The focus of this chapter has changed somewhat from the previous edition, in that “Lymphoma at Uncommon Sites” has been updated to “Rare Lymphomas”. However, the reader should find the two chapters complementary, and is encouraged to read both. In attempting to address the new title, it has been important to obtain prevalent data that are, at best, incomplete and show that there are major geographical variations in incidence.1 This highlights the complex interplay of genetic and environmental factors. The chapter ‘Lymphoma at uncommon sites’ in the previous edition reviewed unusual presentations from an anatomical perspective, whereas this chapter will take a histopathological view, and, where possible, include aspects of biology and treatment. From this perspective, the classification of lymphoid neoplasms remains a dynamic and ever-changing process, and as new technologies emerge, new classifications have also emerged. In the era of gene profiling and proteomics, it is almost getting to the stage that each individual lymphoma will be characterized by a unique signature at the molecular level, as will the patient, and the concept of “rare lymphomas” will become obsolete. Nevertheless, at the time of writing, the classification most current and most widely used is the one devised by the World Health Organisation (WHO)1 to cover hematological malignancies in general, and is the one that evolved from the “Revised European American Classification of Lymphoid Neoplasms” (REAL),2 which was published in 1994. The WHO has published a superb monograph1 covering the pathology and genetics of these conditions, and this monograph should be viewed as “the reference text” for such conditions. The WHO classification recognizes three major categories of lymphoid neoplasms: B cell neoplasms, T and natural killer (NK) cell neoplasms, and Hodgkin’s lymphoma (HL) (see Table 1). The authors of the WHO classification have recognized that within these three major categories there are a large number of lymphoproliferative diseases that are associated with distinct epidemiological, etiological, genetic, and clinical features, and that while it is helpful
for comparative studies, the issue of clinical grouping can be misleading, and for this reason have avoided such groupings. Nevertheless, in addressing the question of “Rare Lymphomas” the WHO classification has been used as the framework for this chapter, as it is almost universally accepted, and incidence figures on which to base a label of “rare” are available in most cases.3 For the purpose of this chapter, a definition of “rare” has been arbitrarily set at less than 5% of the total population of lymphomas. Finally, the literature still abounds with single-case reports and small series of patients with interesting lymphomas,4,5 and to do the topic justice would, as before, require a book rather than a chapter. This chapter has been written with the general oncologist in mind rather than the specialist “lymphomaniac”. Thus, many of the caveats covered in the previous edition chapter ‘Lymphoma at uncommon sites’ still apply, and in particular, selectivity and author bias. Where possible, review articles are cited, therefore many articles will not be referenced; and to the authors, apologies are again offered.
HISTORICAL BACKGROUND The history of lymphoma makes fascinating reading though it is difficult to accurately pinpoint when it all began. An obvious landmark is the often quoted paper6 of Thomas Hodgkin written in 1832 and titled “On Some Morbid Appearances of the Absorbent Glands and Spleen” in which seven patients were described, and it is worth remembering that the concept of lymphoma was first enunciated by Virchov in 1864.7 However, it could also be argued that the invention of the microscope by Robert Hooke in 1665 or the description of lymphocytes by Hewson in 1774 were also of fundamental importance. Three excellent recent historical reviews by Aisenberg,7 DeVita and Canellos8 (on lymphoma), and Bonadonna9 (on Hodgkin’s Disease) are worthy of detailed reading and bring to life the historical aspects of these diseases, and how the classification systems have evolved.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
556
HEMATOLOGICAL MALIGNANCIES
Table 1 WHO classification.
Table 1 (continued).
B CELL NEOPLASMS Precursor B cell NHL
ICD-O code
Lymphoblastic lymphoma – B cell
9728/3
Mature B cell NHL Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) B cell prolymphocytic leukemia Lymphoplasmacytic lymphoma Splenic marginal zone lymphoma Hairy cell leukemia Plasma cell myeloma Solitary plasmacytoma of bone Extraosseous plasmacytoma Extranodal marginal zone (MALT) lymphoma Nodal marginal zone lymphoma Follicular lymphoma Mantle cell lymphoma Diffuse large B cell lymphoma Mediastinal (thymic) large B cell lymphoma Intravascular large B cell lymphoma Primary effusion lymphoma Burkitt’s lymphoma
Percentage of NHL
ICD-O code
Lymphomatoid papulosis
9718/3
<1%
9823/3 9670/3
6 – 7%
9833/3
<1%
9671/3
1%
9689/3
<1%
9940/3 9732/3 9731/3
– – –
9734/3 9699/3
– 7.6%
9699/3
1.8%
9690/3 9673/3 9680/3
22% 6% 31%
9679/3
2.4%
9680/3
<1%
9678/3 9687/3
<1% 2%
HODGKIN’S LYMPHOMA
Nodular lymphocyte predominant Hodgkin lymphoma (NLPHL) Classical Hodgkin’s lymphoma (CHL) Nodular sclerosis CHL (NSCHL) Mixed cellularity CHL (MCCHL) Lymphocyte-rich CHL (LRCHL) Lymphocyte-depleted CHL (LDCHL)
Percentage of HL
9659/3
5%
9650/3
95%
9663/3
70%
9652/3
20%
9651/3
<5%
9653/3
<5%
IMMUNODEFICIENCY ASSOCIATED LYMPHOPROLIFERATIVE DISORDERS IALD
ICD-O code
Lymphoproliferative diseases associated with primary immune disorders Human immunodeficiency virus-related lymphoma Post transplant lymphoproliferative disorders Polymorphic type Methotrexate-associated lymphoproliferative disorder
Percentage of NHL
–
<1%
–
<1%
9970/1
<1%
–
<1%
ICD-O, International Classification of Diseases for Oncology.
Precursor T cell NHL
ICD-O code
Lymphoblastic lymphoma – T cell Blastic NK cell lymphoma
9729/3
1.7%
9727/3
<1%
ICD-O code 9834/3
Percentage of NHL <1%
ICD-O code
T CELL and NK CELL NEOPLASMS
T cell and NK cell NHL T cell prolymphocytic leukemia T cell large granular lymphocytic leukemia Aggressive NK cell leukemia Adult T cell leukemia/lymphoma Extranodal NK/T cell lymphoma, nasal type Enteropathy-type T cell lymphoma Hepatosplenic T cell lymphoma Subcutaneous panniculitis-like T cell lymphoma Mycosis fungoides Sezary syndrome Primary cutaneous anaplastic large cell lymphoma Peripheral T cell lymphomas, unspecified Angioimmunoblastic T cell lymphoma Anaplastic T cell lymphoma
T cell proliferation of uncertain malignant potential
Percentage of NHL
Percentage of NHL –
9831/3
–
9948/3 9827/3
– <1%
9719/3
1.4%
9717/3
<1%
9716/3
<1%
9708/3
<1%
9700/3 9701/3 9718/3
<1% <1% <1%
9702/3
15 – 20%
9705/3
1.2%
9714/3
2.4%
BIOLOGICAL ASPECTS The WHO classification is, to a large extent, based on grouping lymphomas according to the presumed cell of origin of the tumor, and in recognizing that in most cases the lymphomatous cell has arisen from a normal counterpart. With the advent of immunophenotyping, cytogenetics, and molecular genetics, and combining these technologies with clinical correlates, it has been possible to build up a picture, like a jigsaw puzzle, of lymphoma. The picture that has emerged classifies lymphomas into B cell neoplasms, T and NK cell neoplasms, and Hodgkin’s lymphoma, and for convenience this will form the basis of this chapter (see Table 1). Before proceeding, however, it is important to mention the advances that have been made in our understanding of lymphocyte homing and trafficking,10 – 12 which may go some way to explaining the tissue distribution of some lymphomas. For example, it appears that while differential expression of CD44 molecules and the α4 β7 integrin play a role in lymphocyte homing, expression of chemokine receptors like CXCR3 may also be important,13 particularly in rare lymphomas involving the gastrointestinal tract.
RARE LYMPHOMAS
557
Lymphoblastic lymphomas (LBL) constitute about 5% of all non-Hodgkin’s lymphomas (NHLs) in adults, and of these only 10–15% are of B cell phenotype. In contrast, LBL constitutes about 33% of childhood NHL. Thus, 64% of cases in a recent review14 were less than 18 years old and there was a male predominance. The most frequent sites of organ involvement in B-LBL are the skin(particularly multiple nodules), bone, soft tissue, and lymph nodes, and in distinction to T cell lymphoblastic lymphoma (T-LBL), mediastinal involvement is not common.14 In this series with literature review, TdT, CD10, CD19, CD79a and HLA-DR were the most frequently expressed antigens, while CD20 and CD45 were expressed in only two thirds of the cases. Cytogenetic analysis showed additions to 21q as a recurring chromosomal abnormality, and as expected most series show an overlap with precursor B cell acute lymphoblastic leukemia (ALL). Treatment strategies stress the importance of intensive multiagent chemotherapy protocols, and in younger patients an ALL-like protocol is often recommended.15,16 Although high remission rates are seen in children, the prognosis in adults is not as good,16 and the role of transplantation remains to be clearly defined.
adults, particularly males. Nodal involvement is most common, with the liver and spleen also involved. With bone marrow involvement, the distinction with Waldenstrom’s macroglobulinemia becomes blurred, although there are clear distinctions from extranodal mucosa-associated lymphoid tissue (MALT) lymphomas, splenic marginal zone lymphomas (SMZL), and nodal marginal zone lymphomas (NMZL) based on immunophenotyping and genetic abnormalities.17 The finding of t(1;14)(q22;q32) or t(11;18)(q21;q21) is against a diagnosis of LPL,18 and the initial reports19 of a specific cytogenetic abnormality [t(9;14)(p13;q32)] associated with the PAX5 gene in LPL have been recently contested.20 Characteristically an immunoglobulin M (IgM) paraprotein is found, and this may give rise to troublesome hyperviscosity symptoms or result in cryoglobulinemia. Of particular recent interest is the finding that there is an association between type II mixed cryoglobulinemia and hepatitis C infection, and whether this link will prove to be of an etiological nature remains to be clarified. Nevertheless, treatment of patients with cryoglobulinemia by interferon21 or rituximab22 has been beneficial. In addition, this IgM paraprotein may act as an autoantibody and give rise to diverse clinical features depending on the specificity of the antibody binding. For example, neuropathies may occur in up to 10% of cases due to deposition in myelin sheaths, and symptoms of diarrhea may result from binding in the gastrointestinal tract. Similarly, coagulopathies can result from binding of the antibody to coagulation factors. The tumor is characterized by the finding of small B lymphocytes, plasmacytoid lymphocytes, and plasma cells infiltrating diffusely through a lymph node. The cells usually express CD19, CD20, and CD79a, but are CD5− , a feature that aids distinction from CLL/SLL, and CD10 and CD23 are also normally not expressed. Treatment of LPL depends usually on the presence of symptoms, as asymptomatic patients are usually not treated. If treatment is required then alkylating agents or purine analogs have been the mainstay of therapy, with plasmapheresis used for IgM-associated problems. More recently, rituximab in extended doses has been advocated.22 Median survival is about 5 years, with transformation to a diffuse large BCL which occurs in about 10% of patients heralding a poor prognosis.
Lymphoplasmacytic Lymphoma (LPL)
Splenic Marginal Zone Lymphoma
This condition is often considered as the tissue equivalent of Waldenstrom’s macroglobulinemia (WM), in the same way that small lymphocytic lymphoma (SLL) is considered the tissue equivalent of chronic lymphatic leukemia (CLL), or LBL is considered the tissue equivalent of ALL. While there may be subtle differences in the genetic make-up of these counterparts, the differential expression of adhesion molecules on the cell surface, for example, CD18 (LFA 1), may simply explain the different tissue distribution. As gene expression profiling explores the differences in adhesion molecule expression, it is likely that the picture will become much clearer. In any event, LPL is a rare lymphoma and comprises about 1.5% of nodal lymphomas,1 occurring predominately in older
This is again a rare disorder comprising less than 1% of lymphomas, and as its name suggests, is a disease which predominately affects the spleen, with infiltration of both the white pulp germinal centers, and also the red pulp. There are several excellent recent reviews.23,24 Patients are usually over 50 years old and present with symptoms of splenomegaly, but autoimmune blood dyscrasias may also be the presenting feature. Lymphadenopathy at the splenic hilum is frequently seen as is bone marrow involvement, and when villous lymphocytes were seen in the blood, the term splenic lymphoma with villous lymphocytes (SLVL) was coined. These villous lymphocytes are presumed to be the tumor cells, and are characterized by short cytoplasmic projections that can mimic the appearance of hairy cells. Immunophenotyping of the
B CELL NEOPLASMS The WHO classification divides B cell neoplasms into those derived from “immature” or precursors cells and those derived from more mature cells. Thus, precursor B lymphoblastic leukemia/lymphoma is recognized as a disease of small to medium-sized lymphoblasts (i.e. immature cells), committed to the B cell lineage, and makes the distinction between the two names on the basis of the predominant tissue involved. If the disease affects the blood and marrow predominantly then the term “leukemia” is preferred, whereas “lymphoma” designates the disease where a mass lesion and minimal blood and bone marrow involvement is seen. Although there can be significant overlap in the two conditions, and indeed the two conditions may simply represent ends of a clinical spectrum, in deference to the chapter title, only precursor B cell lymphoblastic lymphoma (B-LBL) will be discussed further.
Precursor B Cell Lymphoblastic Lymphoma
558
HEMATOLOGICAL MALIGNANCIES
cells usually reveals positivity for CD20, CD79a, and surface IgM and IgD. The cells are usually negative for CD5 and CD43 (excluding CLL and mantle cell lymphoma), CD10, CD23, and CD103 (the latter being helpful in excluding hairy cell leukemia, although some series25 challenge this). Cytogenetic features are heterogeneous,25 with abnormalities of chromosomes 3, 5, and 7 (particularly deletions of 7q31–32) all being described, and there is variability in the mutation pattern of immunoglobulin variable heavy (VH) chain gene expression.26 In common with other indolent BCLs, there is a high prevalence of hepatitis C seropositivity, particularly in a recent Italian series.27 In terms of management, this is an indolent lymphoma, and splenectomy is probably the treatment of choice,28 although alkylating agents, purine analogs, and monoclonal antibody therapy may be useful in selected cases.29 As in other lowgrade BCLs, transformation to more aggressive forms occurs rarely and indicates a poor prognosis.
Extranodal Marginal Zone B Cell Lymphoma of Mucosa-associated Lymphoid Tissue (MALT Lymphoma) As the name suggests, these are diseases that affect mucosal surfaces which, often as a result of chronic inflammation and antigen stimulation, become infiltrated by lymphoid cells and eventually become antigen-independent and undergo malignant transformation. Many sites can be involved but the gastrointestinal tract (>50%), lung (14%), head and neck (14%), ocular adnexae (12%), and skin (11%) are common sites.1 Technically speaking, MALT lymphomas are not that “rare” when all sites of disease are taken into account, and may comprise up to 7–8% of all BCLs, thus falling outside the scope of this chapter. Although they do arise in a variety of tissues, reflecting the homing patterns of normal lymphocytes, they often share pathological and clinical features, and there has been some recent interesting biological perspectives in subsets of patients that warrant discussion,30,31 and for this reason some aspects will be discussed, albeit superficially here. Gastric MALT Lymphoma
Gastrointestinal MALT lymphomas comprise over 50% of all MALT lymphomas, stretching from the tongue32 to the rectum;33 but by far the most intensely studied are those in the stomach, which are associated with Helicobacter pylori infection.34 Although the precise mechanism remains unclear, there is evidence that infection triggers a wide range of inflammatory responses, including neutrophil activation and the release of reactive oxygen species, leading to an orderly progression of cytological changes culminating in malignant transformation. At a molecular level, the t(11;18)(q32;21) which involves the API2 gene and the MALT1 gene, is common in both gastric and pulmonary MALT disease, but rare in those lymphomas occurring in other extranodal sites like the salivary glands, skin, or ocular adnexal tissues, where the t(14;18)(q32;q21) translocation is more frequently seen.35 Gastric MALT lymphomas are often localized at presentation, although it is not uncommon to
find adjacent lymphadenopathy. Morphologically, the cells are usually small to medium-sized lymphocytes although the presence of large cells is not uncommon and may lead to the diagnosis of a large cell lymphoma. In any event, early stage disease may be eradicated in over 60% of cases by antibiotics used to treat H.pylori infection even in the presence of large cells. While radiation therapy gives excellent results36 in localized disease, a recent randomized trial of surgery, radiotherapy, and chemotherapy, favored the use of chemotherapy with overall survival predicted to be 87% at 10 years in the chemotherapy arm.37 Nevertheless, late relapses are well documented and lifelong follow-up is advocated.38 Nongastric MALT Lymphoma
As mentioned above, MALT lymphomas can occur in many sites other than the stomach,39 and this has become an area of intense study with the finding of particular infectious associations.30 Specifically, ocular adnexal MALT lymphoma has been associated with Chlamydia psittaci infection,40 skin MALT lymphoma with Borrelia burgdorferi infection,41 small intestinal MALT lymphoma with Campylobacter jejuni infection,42 and salivary gland MALT lymphoma with hepatitis C infection.43 It will be most interesting to see if other infectious agents are associated with other MALT or indeed non-MALT lymphomas, and whether anti-infective strategies will eradicate infection and lymphoma. Nevertheless, patients with asymptomatic disease may be observed without initial therapy, while those with localized disease should have radiation therapy44 and those with symptomatic or progressive disease may benefit from single agent alkylating agent therapy, nucleoside analogs, or monoclonal antibody therapy.45
Nodal Marginal Zone B Cell Lymphoma This condition is the nodal equivalent of its splenic or extranodal counterpart, and is indeed less common, comprising less than 2% of all lymphoid neoplasms.46 In many respects, the disease behaves as an indolent lymphoma, usually presenting with localized disease, and although no randomized trials are available to guide therapy, irradiation for localized disease seems a reasonable option.44
Mediastinal Large B Cell Lymphoma This disease has been classified as a subtype of diffuse large cell lymphoma arising in the mediastinum and often associated with sclerosis; however, as gene expression profiling has emerged, it has been recognized to have a unique molecular signature and many feel that it is a disease more biologically related to Hodgkin’s lymphoma.47,48 It represents about 2–4% of lymphomas and is most frequently seen in young middle-aged women, though pediatric and adolescent patients are well recognized.49 Patients usually present with localized disease and symptoms relating to an enlarging mediastinal mass, and radiological features may help to differentiate other types of mediastinal lymphomas.50 The tumor cells are heterogeneous in size, but usually have abundant
RARE LYMPHOMAS
559
pale cytoplasm and are surrounded by bands of fibrous tissue, which may help to contain rapid enlargement. The cells show a B cell phenotype with CD19 and CD20 positivity, but are usually negative for CD5 and CD10. The presence of CD23 positivity in over 70% of cases has been to support tumor origin from activated dendritic thymic B cells.51 Treatment strategies have promoted the use of multiagent chemotherapy regimens like MACOP-B,52 with the addition of post chemotherapy mediastinal radiotherapy52 and stem cell transplantation in poor risk patients.53 Using such regimens, patients have projected 10-year overall survival rates of greater than 70% in one series.54
Treatment is likewise often difficult66 and even with highdose chemotherapy and stem cell transplantation the disease may be refractory,67 and attempts to control disease with inhibitors of viral replication have been proposed.68 Clinical variants of PEL are described, including post transplant presentations, and in elderly subjects of eastern European descent, these are even rarer.62 Recently reports have emerged of HHV-8 negative PEL in immunocompetent patients. These are again very rare, and appear to be related to chronic EBV infection, particularly in Japan.62 These are again aggressive diseases with a poor prognosis.
Intravascular Large B Cell Lymphoma
Burkitt Lymphoma
This is a rare lymphoma first described in 195955 and accounting for less than 1% of all lymphomas. As its name suggests, it is characterized by the presence of lymphoma cells in the lumina of small blood vessels, often capillaries, and usually at extranodal sites. Case reports are the most common description in the literature with involvement of the central nervous system (CNS),56 skin,57 breast,58 and uterus59 being recently described. An Asian variant has been proposed60 characterized by hemophagocytosis. Because of the predilection to form within the blood vessels, the clinical presentation is dictated by the site of tissue involvement, but skin and neurological symptoms are said to be most common; however, there may be multiple organ involvement including the bone marrow, and pancytopenia and disseminated intravascular coagulopathy have been reported.1 The tumor cells are usually positive for mature B cell markers CD19, CD20 and CD23, and in a subset of cases the CD5 antigen is also positive.60 A recent report has suggested that prostatic acid phosphatase may be an appropriate tumor marker.61 Treatment is empirical as there are no large clinical trials, but this is an aggressive tumor, and multiagent chemotherapy is recommended.
Burkitt lymphoma (BL) is named after Denis Burkitt, an English surgeon and pathologist, who studied the epidemiology of this highly aggressive tumor in central Africa. The disease now is recognized to have three main clinical variants, which in total represents 2% of all lymphomas. The classical endemic BL occurs predominantly in young children (aged less than 7 years), in equatorial Africa and Papua New Guinea, and has been correlated to areas of endemic malaria. Sporadic BL, on the other hand, is seen throughout the world, and occurs in children and adults. The third clinical variant is BL associated with immunodeficiency states, particularly in patients with HIV infection. The face and jaw bones are common sites of involvement in endemic BL, whereas abdominal involvement is more commonly seen in sporadic BL, although extranodal involvement, particularly the CNS is commonly seen in all three variants. When marrow involvement occurs, a leukemic picture can present with circulating blasts that show a phenotype of mature B cells, with positivity for CD19, CD20, CD22, CD79a, and surface immunoglobulin. This is a tumor with an interesting molecular etiology,69 which has been intricately involved with EBV infection. In the endemic form, the EBV genome has been found in the tumor cells in all cases, as distinct from the other two variants where it is found in only about 25–30% of cases. Nevertheless, all cases have a translocation involving the MYC gene at band q24 on chromosome 8 to the Ig heavy chain region on chromosome 14 at band q32, or less commonly to one of the Ig light chain loci on chromosome 2p12 or chromosome 22q11. In endemic cases, the breakpoint on chromosome 14 involves the heavy chain joining region, whereas in sporadic cases the Ig switch region is involved, leading to constitutive expression of the MYC gene. This is a tumor with a rapid doubling time, and so the clinical picture can change dramatically. Patients often present with a short history, and can have advanced stage disease, with high uric acid and lactic dehydrogenase (LDH) levels, and urgent treatment is required. Having said that, it is important to have in place measures to anticipate tumor lysis syndrome,70 which is particularly likely to happen when treatment begins due to the high proliferative fraction (often 100%) and the sensitivity of the tumor cells to chemotherapy. All patients should be assessed for their ability to undergo
Primary Effusion Lymphoma This again is a rare disease occurring in less than 1% of all lymphomas, and is almost universally associated with human herpes virus-8 (HHV-8)/Kaposi sarcoma herpes virus (KSHV) infection in an immunocompromised host, often infected with the human immunodeficiency virus (HIV). The disease is part of a larger clinical grouping known as body cavity lymphomas,62 where in some cases there is no evidence of HHV-8 infection. Primary effusion lymphoma (PEL) was first recognized in 199563 when HHV-8 DNA sequences were found in a subgroup of AIDS-related lymphomas with pleural, pericardial, or peritoneal effusions. Typically, only one site is involved, and patients present with symptoms referable to the site of fluid accumulation. The tumor cells are positive for HHV-8/KSHV in all cases of PEL, and may also be coinfected with Epstein-Barr Virus (EBV), which confers different cellular phenotypes.64 These cells are usually large cells and harbor mutations in BCL-665 and IgH genes and express CD45 and CD138, but are often negative for CD19 and CD20 making lineage assignment often difficult.
560
HEMATOLOGICAL MALIGNANCIES
intensive, but short duration, multiagent chemotherapy,71 such as CODOX –M/IVAC (cyclophosphamide, vincristine, doxorubicin, high-dose methotrexate/ifosfamide, etoposide and high-dose cytarabine), which is best tolerated in children. The role of cranial irradiation has been recently reviewed in the light of previous excessive neurotoxicity72 as has the need for better treatment protocols in older patients.71 Despite this, the outcome can be reasonable and a recent report quoted an actuarial 3-year event-free survival of 69% and overall survival of 76% at 3 years after autologous transplantation.73 The optimal treatment for acquired immunodeficiency syndrome (AIDS) – related lymphoma remains unclear, and although the use of highly active antiretroviral therapy (HAART) seems intuitively worthy of recommendation this remains a controversial area.74 Similarly, a randomized trial of rituximab-cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) versus CHOP alone, conducted by the AIDS Malignancies Consortium, showed no event-free survival benefit in a preliminary analysis for the addition of rituximab, and indeed a higher incidence of septic deaths was noted in the rituximab-containing arm.75
are T lymphoblasts, which are medium-sized cells with a high nuclear-to-cytoplasmic ratio. These cells express TdT, a defining feature, and usually also express cytoplasmic CD3 and CD7. There is often coexpression of CD4 and CD8, and CD10 may also be expressed. Sometimes, there is aberrant expression of myeloid markers such as CD13 and CD33. Genetically, the most common abnormalities involve translocations of the T cell receptor gene (the α and λ gene loci are at 14q11.2, the β locus at 7q35, and the γ locus at 7p14–15) with partner genes like MYC on 8q24.1, leading to transcriptional disruption of the partner gene. Treatment of lymphoblastic lymphoma in children usually involves the use of multiagent chemotherapy protocols as used in high risk ALL.15 Similarly in adults, aggressive protocols are advocated16 as, although no randomized trials have been reported, CHOP-like regimens without CNS treatment appear inadequate. Mediastinal radiotherapy should be directed at appropriate masses, and overall survival at 5 years has been reported at over 50%.78 High-dose chemotherapy with autologous stem cell rescue appears to be effective therapy but in a small randomized trial was not shown to be superior to standard chemotherapy.79
T CELL AND NK CELL NEOPLASMS
Blastic NK Cell Lymphoma
T cells and NK cells are thought to be derived from a common precursor cell and because of this and the fact that neoplasms derived from these cells show overlapping phenotypic and clinical features, they are diseases that are often grouped together, and are so in the WHO classification. Nevertheless, these diseases are not common in the Western world, and indeed are also very heterogeneous, and as we learn more about their genetic make-up and protein expression, it is likely that the classification of these diseases will evolve rapidly in the next few years. Two excellent reviews on NK cell lymphomas have just been published,76 including one on pediatric patients.77 As with BCLs, the T cell and NK cell lymphomas have been divided in the WHO classification into precursor and mature neoplasms reflecting the degree of differentiation of the tumor cells (see Table 1).
This is an extremely rare lymphoma, only recently recognized,76 and usually found in middle-aged or elderly patients, though children have been affected.77 An alternative term is CD4+ /CD56+ hematodermic neoplasm,80 as this describes the phenotype of the neoplastic cells and recognizes the high incidence of cutaneous as well as lymph node and bone marrow involvement (see Figure 1). It has been proposed that the tumor cells are derived from a plasmacytoid dendritic cell (DC2),81 and recent microarray analysis may dictate reclassification.82 The cells are usually negative for CD3, B cell and myeloid antigens, and the combination of CD4 and CD56 positivity is characteristic. This is an aggressive disease, particularly if the disease extends beyond the skin, and usually responds poorly to conventional lymphoma regimens, with survival quoted at less than 3 years.83 The hyper-CVAD regimen appears helpful in elderly patients,84 and L-asparaginase therapy has been successfully applied in children.85
Precursor T Cell Neoplasms Precursor T Lymphoblastic Leukemia/Lymphoblastic Lymphoma
Like its B cell counterpart, the distinction between these two terms is somewhat arbitrary, and is determined by the predominant tissue involvement. With extensive bone marrow involvement, the term ALL is used, whereas if there is less than 25% lymphoblasts in the marrow and a significant mass lesion, the term lymphoblastic lymphoma is preferred. Precursor T lymphoblastic lymphoma comprises about 1.7% of all lymphomas in a large series,3 and about 85% of all LBLs. It is most frequently seen in adolescent males, and patients typically present with a large mediastinal mass, in distinction to patients with B cell disease, who rarely do not. Pleural effusions are commonly seen as is nodal involvement, and rarely extranodal disease may be the presenting feature. The tumor cells that define this disease
Figure 1 Lymph node smear of blastic NK cell lymphoma/leukemia.
RARE LYMPHOMAS
561
Mature T Cell and NK Cell Neoplasms In total, mature T cell and NK cell neoplasms represent only about 12% of all lymphomas in a large international series,3 but it is important to reemphasize that there is a wide geographical variation, and that T cell and NK cell neoplasms are more commonly seen in oriental rather than occidental populations.86 The reasons for this are not totally clear and are likely to be multifactorial, but genetic make-up and viruses, including the EBV, are sure to be major players. Returning to the WHO classification, only those diseases that are considered to be lymphomas, rather than leukemias, will be discussed. Having said that, the first condition is indeed a hybrid disorder, reflected in its name.
Lymphoma
Adult T Cell Leukemia/Lymphoma (ATLL)
This is a fascinating disease, caused with the human T cell lymphotrophic virus (HTLV-1), which is an oncoretrovirus and is said to infect 15–20 million people worldwide.87 The virus can be transmitted in three ways: horizontally, vertically (i.e. mother to child), and through blood transfusion, causing two major diseases – ATLL and tropical spastic paraparesis. While most individuals infected with HTLV-1 remain asymptomatic carriers, about 1–5% will develop ATLL87 often after a long latency period, and the reasons for this long latency and low lymphoma incidence are the subjects of much controversy. The disease has a well-described geographical distribution, paralleling the distribution of endemic HTLV-1 infection, and is prevalent in Japan, the Caribbean, and parts of central Africa. Clinically, the disease is heterogeneous with four major subtypes being recognized: smoldering and chronic forms with median survivals of 2 years, and acute and lymphomatous forms that are more aggressive with median survivals of less than 12 months.88 The most common variant is the acute type, presenting with a leucocytosis due to the presence of large multilobated lymphoid cells, sometimes called “flower cells”. Hypercalcemia, skin involvement, and generalized lymphadenopathy are also features of this acute subtype. In the lymphomatous presentation, lymphadenopathy predominates and hypercalcemia and leucocytosis are less likely. Whatever subtype presents, this is an aggressive disease and it is quite common for the more indolent forms to transform into a more aggressive entity, resulting in poor survival. New therapeutic approaches have been recently proposed89 and include the use of arsenic trioxide, proteasome inhibitors and an interesting approach targeting the CC chemokine receptor 4.90 Extranodal NK/T Cell Lymphoma, Nasal Type
This is a lymphoma highly associated with EBV infection, and found more commonly in Asian communities and people of central and South American descent.76 It used to be called “lethal midline granuloma” or “angiocentric T cell lymphoma”. In south east Asia, up to 8% of lymphomas may be of this subtype,1 whereas in western series the incidence is much less. Adult males over 50 are most commonly affected, and, as the name suggests, the most common presentation is a destructive nasal or midline facial lesion causing nasal
Figure 2 Computed tomography (CT) scan of skull in case of extranodal NK/T cell lymphoma, nasal type.
obstruction (see Figure 2), although other extranodal sites like the skin, testes, and upper respiratory tract are also commonly involved. Hemophagocytosis can be seen as a late complication and often heralds a rapid downhill course. Morphologically the tumor cells may show a spectrum of size characteristics from small to large and anaplastic, and there may be an admixture of reactive, inflammatory cells. Typically, the tumor cells are positive for CD2, CD56, and cytoplasmic CD3, but negative for surface CD3, CD4, CD8, CD16, and CD57. Cytogenetically, deletions involving the long arm of chromosome 6, [del(6)(q21q25)], are most common, but the significance of this is unclear. Again, the prognosis for this disease is poor, and while radiation therapy may be beneficial in localized disease, the response to current chemotherapy protocols is disappointing and new treatments are required.91 Biological correlates of response are interesting, with a report that CD94 expression in these diseases may confer a better prognosis,92 although p53 overexpression did not correlate with survival.93 A rare, predominately cutaneous variant seen in children in the same geographical regions has been described94 and called “Hydroa vacciniforme-like lymphoma”. Enteropathy-type T Cell Lymphoma (ETTL)
This is a rare tumor of intraepithelial T cells, accounting for less than 1% of all lymphomas, and most commonly affects the jejunum or ileum.95 It has long been associated with coeliac disease, and while a small proportion of patients have a history of childhood coeliac disease, often the diagnosis of adult onset coeliac disease is made at the same time as the lymphoma presents. Interestingly, however, most lymphomas associated with coeliac disease in a large population-based study in Sweden were not enteropathy-type T cell lymphoma (ETTL), but other T cell lymphomas and B cell disease.96 The neoplastic cells are positive for CD3 and CD7, but usually negative for CD4, CD5, and CD8, and chromosome 9 harbors interesting abnormalities in this disease, with loss of heterozygosity at 9p21 being frequently found,97 as well as amplification of NOTCH1 and ABL1 genes at 9q34.98
562
HEMATOLOGICAL MALIGNANCIES
The tumor characteristically ulcerates the intestinal mucosa and eventually invades the intestinal wall, causing abdominal pain and intestinal obstruction. Positron emission tomography (PET) scan has been recommended for assessing disease,99 and surgical intervention is usually required, often on a recurrent basis. Obviously, on a background of coeliac disease, patients may be malnourished, and this adds an extra dimension to an already difficult management situation. In a recent Canadian series of T cell lymphomas,100 patients with ETTL were in the worst prognostic group with a 5-year overall survival of less than 25%, and obviously new treatment strategies are required. Perhaps with the advent of genetic expression profiling, new molecular targets will be identified. Hepatosplenic T Cell Lymphoma
This again is a rare lymphoma accounting for less than 5% of peripheral T cell lymphomas, and is thought to originate in postthymic cytotoxic T cells, similar to those found in ETTL, and expressing the γ δ T cell receptor (TCR).101 In contrast to most other T cell lymphomas, this is a disease most commonly seen in adolescents, particularly males, and patients usually present, as the name suggests, with marked hepatosplenomegaly without lymphadenopathy.102 The hepatosplenomegaly is due to extensive sinusoidal infiltration without parenchymal involvement of CD2, CD3, and CD7 positive neoplastic lymphoid cells. Typically, these cells have γ δT CR gene rearrangements; however, a rarer variant condition, more commonly seen in females, appears to have αβT CR gene rearrangements.103 Isochromosome 7q (i7q) and trisomy 8 are consistent chromosomal abnormalities.104 Patients may present with fatigue, anemia, and thrombocytopenia due to the combination of marrow involvement and hypersplenism. Again, the clinical course is very aggressive, with median survival figures of less than 12 months, and there is no standard effective treatment. As in similar cases with an aggressive course in a younger age-group, highdose chemotherapy with stem cell support seems a reasonable option, but new treatment strategies are required.
CUTANEOUS LYMPHOMAS Perhaps, of all the lymphomas, the cutaneous lymphomas have been most studied in addressing the question of why different lymphomas present in different tissues.105 While many answers remain to be determined, there is compelling evidence for the selective use of certain adhesion molecules and chemokine receptors in these conditions.106 In a similar vein, there may be selective use of V β genes as detected in the circulating neoplastic T cells of cutaneous T cell lymphoma (CTCL) in the leukemic phase.107 Gene expression profiling is already being used to unravel the vagaries of why some neoplastic T cells have a predilection for cutaneous tissues.108 Although the WHO classification recognizes a subgroup of T cell lymphomas as CTCLs, like all classifications constant revision is required, and has resulted in a recent update following collaboration with the European Organization for Research and Treatment of Cancer (EORTC). This new classification, termed the WHO-EORTC classification,109
has brought together previous classifications and has been applied to 1905 patients with primary cutaneous lymphomas from Dutch and Austrian registries to illustrate the utility of the new classification. It is likely that this new classification will become “the gold standard”, but until it receives more widespread acknowledgement the current WHO classification will be used here.
Subcutaneous Panniculitis-like T Cell Lymphoma (SPLTL) This disease represents less than 1% of all lymphomas, and like the previous two conditions, is thought to be derived from postthymic T cells, but this disease appears to have a predilection for subcutaneous tissues. The sex distribution is said to be equal, and the age distribution spans all decades. A recent excellent review110 of 156 patients in the English medical literature provides a comprehensive picture of the pathology and clinical features of this condition. Most patients present with multiple subcutaneous nodules (see Figure 3), usually larger than 1 cm and occurring on the trunk or limbs, and in 37% of the cases reviewed above, hemophagocytosis was observed at presentation and shown to be a negative prognostic factor. Biopsy typically shows an infiltrate of αβ T cells, positive for CD3 and CD4 in the subcutaneous tissues, with sparing of the dermis and epidermis. This is a disease with a poor prognosis using conventional therapy for lymphoma, with median survival in one series of 27 months,110 but in another was only 5 months,111 and as before new therapeutic strategies are needed.
Mycosis Fungoides (MF) MF is typically an indolent skin lymphoma, and usually presents with erythematous patches or plaques which may progress to cutaneous tumors, generalized erythrodermia and, if the cells spill into the blood, then this is termed S´ezary syndrome (SS). This is the most common CTCL, and taken in conjunction with SS, accounts for about 45% of all cutaneous lymphomas,112 but less than 0.5% of all lymphomas. What causes MF remains unknown, but the HTLV-1 virus has been implicated in a North American series,87 but not in a recent series from Poland,113 suggesting geographical variability. The cells that infiltrate the skin are cerebriform mature T cells that are usually CD3, CD4, and CD5 positive, but CD7 and CD8 negative. Clinically, patients usually present with isolated scaly erythematous patches, which may not come to medical attention for months or even years. These may evolve to more extensive patches or plaques with lymphadenopathy, to frank tumors with ulceration, and eventually to a disseminated disease with visceral involvement. A comprehensive analysis of staging, prognosis, and treatment has been recently published.112 In summary, prognosis is related to the extent of disease with patients who present with early stage disease (T1) having an excellent prognosis (100% at 10 years), and even patients with advanced stage disease at presentation, the so-called T4 disease, having a 10-year survival of over 40%.112
RARE LYMPHOMAS
563
decetylase inhibitors, monoclonal antibodies, and fusion toxins to allogeneic stem cell transplantation and even vaccine strategies. These options have been recently reviewed112 and the fact that so many exist probably reflects the unsatisfactory outcomes.
Primary Cutaneous Anaplastic Large Cell Lymphoma (PCALCL)
Figure 3 Ulcers in case of subcutaneous panniculitis-like T cell lymphoma.
A major consideration is the potential for transformation to a large cell lymphoma, occurring in up to 39% of 115 cases in one series,114 especially those patients with cutaneous tumors. Such transformation appears more likely in patients with advanced stage disease, and has a major impact on survival, and hence biopsy is mandatory in patients with rapidly progressive lymphadenopathy.
S´ezary Syndrome (SS) This is often regarded as a clinically aggressive variant of MF where there is usually generalized pruritic erythrodermia, lymphadenopathy, and abnormal cerebriform T cells, the socalled S´ezary cells, in the blood.112 Visceral involvement is common in the terminal stages of this disease but remarkably the bone marrow is spared. This is an aggressive disease and deserves separate consideration, not only as a distinct entity in the WHO classification, but also because it is the subtype with a significantly inferior survival to MF (5-year survival of 33% vs 87% for MF).115 Treatment strategies range from conventional lymphoma chemotherapy including nucleoside analogs, through histone
This is the most common of the primary cutaneous CD30+ T cell lymphoproliferative disorders (the others being lymphomatoid papulosis [LyP] and the so-called “borderline lesions”), which in total represent about 25% of CTCLs.116 PCALCL is typically a disease of elderly males; however, pediatric cases have been reported and recently reviewed.117 The disease, as its name suggests, should be confined to the skin, and extracutaneous involvement should result in reclassification to systemic anaplastic large cell lymphoma (ALCL). Patients usually present with solitary, or at least localized, nodules, tumors, or rarely papules, and multifocal disease is uncommon but associated with a poorer outcome.118 The tumor cells usually express CD4 and CD30; however, the anaplastic lymphoma kinase (ALK) protein typically seen in the cells of systemic ALCL is negative. Treatment is directed at local disease control with surgical excision or radiotherapy, and disease-specific survival rates of over 90% at 5 years for cases with localized disease have been reported,119 whereas it is only 50% for generalized disease.120 Multiagent chemotherapy may be required for more aggressive cases, while the use of anti-CD30 monoclonal antibody shows promise.118 LyP is a similar condition characterized by papules, which spontaneously come and go, often within 3 to 6 weeks, and which on biopsy show an atypical T cell infiltrate mimicking lymphoma. Although this is viewed as a benign condition with an excellent prognosis (overall survival in one series was 92% at 10 years120 ), some patients may progress on to a frank lymphoma of varying histological subtypes, particularly MF or Hodgkin’s lymphoma.
Angioimmunoblastic T Cell Lymphoma (AITL) This rare lymphoma accounts for about 15–20% of peripheral T cell lymphomas, or 1–1.5% of all NHL. It occurs in adult patients with an equal sex distribution, and presents as a systemic disease often in an advanced stage, with lymphadenopathy, hepatosplenomegaly, and a pruritic skin rash.121 A polyclonal hypergammaglobulinemia may accompany autoimmune laboratory features, for example, cold agglutinins and hemolysis. The diagnosis is usually established by finding a polymorphous lymphoid infiltrate and the proliferation of new blood vessels in the tissues. Although these cells may be a mixture of CD4 and CD8 cells, it is the finding of CD21+ follicular dendritic cells (FDCs) surrounding the arborizing blood vessels that clinches the diagnosis. Of some recent interest is the finding that there is increased expression of the CXC chemokine receptor 3 (CXCR3) in AILT,122 although the significance of this remains unclear, and the neoplastic T cells express CD10.123 The prognosis in this condition remains poor with conventional lymphoma chemotherapy, and median survival
564
HEMATOLOGICAL MALIGNANCIES
figures of less than 3 years are quoted. A retrospective European study reporting on 29 patients treated with highdose chemotherapy and autologous stem cell transplantation reported a 5-year overall survival of 44%.124
Anaplastic Large Cell Lymphoma (ALCL) ALCL accounts for about 2.4% of NHLs, and is the most common pediatric mature T cell lymphoma.77 The disease is characterized by the proliferation of CD30+ neoplastic lymphoid cells, which typically (60–80%) express the ALK protein, although ALK-negative cases are also well recognized. The normal function of the full length ALK protein remains to be fully established, but it is a receptor tyrosine kinase which appears to be important in neural and muscular development;125 however, its expression in these lymphoma cells is due to chromosomal translocation leading to the formation of an ALK-derived oncogenic fusion protein.126 The ALK locus is on chromosome 2, and several translocation partners have been identified, but the most common is the nucleophosmin (NPM) gene on chromosome 5, giving rise to a t(2;5)(p23;35) abnormality which can be detected by reverse transcriptase polymerase chain reaction (RT-PCR). Variant translocations have been identified involving the ALK locus and partner genes on chromosomes 1, 2, 3, and 17. Although all these translocations result in increased ALK protein expression, cytochemically there is variation in the site of such overexpression, with the typical t(2;5) translocation resulting in both nuclear and cytoplasmic localization, whereas variants give rise to only cytoplasmic staining. ALK-positive ALCL is most commonly seen in men under 30 years, who typically present with systemic symptoms, and have nodal disease as well as extranodal involvement. The skin is commonly involved, and such a presentation must be distinguished from the primary cutaneous form previously described, which is usually a more indolent disease. Clusterin has been recently identified by microarray assays as a gene that may be differentially expressed in systemic ALCL but not in primary cutaneous ALCL.127 Bone, bone marrow, soft tissue, and lung are other common sites of the disease.128 1
Overall Survival
ALK+
Probability
0.75
n = 53 0.50
p <0.0007 0.25
ALK−
n = 25 0
0
1
2
3
4
5
6 7 Years
8
9
10 11 12 13
Figure 4 Survival of ALK+ ALCL and ALK–ALCL (Falini B, et al. Blood 1999; 93:2697 – 706). (From Clift RA, Martin P, Fisher L, Buckner CD, Thomas ED. Allogenic marrow transplantation for CML in phase-risk factors for survival. Blood 1987; 70: 104 – 112. American Society of Hematology).
On the other hand, ALK-negative cases are more commonly seen in older patients who do not show the same male predominance, and less extranodal disease.128 ALK positivity is of significance in terms of survival (see Figure 4) and appears to confer a more favorable prognosis, with overall survival of close to 80% at 5 years compared to 40% in ALKnegative cases.129 Of some interest is the recent report of ALCL in breast tissue of patients who had received silicone implants.130
HODGKIN’S LYMPHOMAS Hodgkin’s lymphomas,131 formerly Hodgkin’s disease, comprise about 25% of all lymphomas and like NHL there is a wide geographical variation, with high rates in Europe and North America and low rates in Asia and Africa.132 The WHO classification identifies two separate entities – nodular lymphocyte predominant Hodgkin’s lymphoma (NLPHL) and classical Hodgkin’s lymphoma (CHL), which has four histopathological subtypes (see Table 1). As with NHL, this chapter will focus on the pathological subtypes, rather than anatomical localization of disease, which formed the basis of the chapter ‘Lymphoma at uncommon sites’ in the previous edition, and again the chapters should be considered complementary and read accordingly. Before discussing those “rare” subtypes (<5%), it is worth mentioning that recent reviews of Hodgkin’s lymphoma in the stomach,133 rectum,134 bone,135 mandible,136 and CNS137 and as an endobronchial presentation138 have been published.
Nodular Lymphocyte Predominant Hodgkin’s Lymphoma (NLPHL) As detailed in Table 1, this condition represents about 5% of all HL, and is characterized by the presence of large neoplastic cells known as “popcorn” or “L and H” cells, which reside in follicles and are surrounded by reactive lymphocytes. These “L and H” cells are CD20, CD45, and CD79a+ , but usually CD30− , and many investigators view this condition as a low-grade B cell NHL. When the pattern is diffuse rather than nodular, the condition may be indistinguishable from T cell rich B cell lymphoma (TCRBCL).139 The disease is usually found in patients aged 30–50 years, and there is a male predominance. Most patients present in an indolent manner, with localized nodal disease in the head and neck, axilla, or groin. The prognosis of this condition is excellent and indeed is superior to other classical variants, at least in part due to the young age and other favorable prognostic factors at presentation.140 Most patients with early stage disease are cured by their primary therapy, and excellent long-term survival has been achieved in stage I –IIA with extended field radiotherapy.141 Optimal chemotherapy has not been established through clinical trials, but standard HL regimens are effective, and rituximab shows promise.142
Lymphocyte-rich Classical Hodgkin’s Lymphoma (LRCHL) As its name suggests, this subtype of CHL is characterized by the presence of scattered Reed–Sternberg cells in a
RARE LYMPHOMAS
background of small lymphocytes without reactive myeloid cells. It comprises about 5% of all Hodgkin’s lymphoma, and is typically found in older males. Patients usually present with peripheral lymphadenopathy and early stage disease (I/II), but without classical “B” symptoms. This is a subtype with a good prognosis, and indeed may have survival rates superior to other types of CHL and similar to NLPHL.143
Lymphocyte-depleted Classical Hodgkin’s Lymphoma (LDHL) This subtype is the rarest form of CHL and indeed, in the past, many cases may have been cases of NHL, particularly ALCL or peripheral T cell lymphoma, and gene expression profiling has been helpful in exploring difficult cases.144 Even with this caveat, this subtype comprises less than 5% of all cases of HL. It is a disease more commonly found in developing countries, and is the subtype associated with HIV infection. Patients are typically middle-aged males, who present with symptomatic advanced stage disease often involving retroperitoneal nodes, and with an absence of peripheral lymphadenopathy. The prognosis is the poorest of all subtypes, not only reflecting the advanced stage at presentation but also the aggressive biological behavior.
MISCELLANEOUS LYMPHOMAS In concluding this chapter, it seems reasonable to acknowledge that many types of lymphomas have not been covered, or even mentioned, and so a final few words and references are required. Many of the references in my previous chapter (Textbook of Uncommon Cancer, 2nd edition) related to lymphomas occurring in unusual anatomical sites, often in association with immunodeficiency states, particularly those associated with HIV infection. Such HIV-associated cases continue to be reported, though it appears that, in general, the distribution of cases in extranodal sites, particularly the gastrointestinal tract and CNS, has not changed dramatically from the pre-HAART to the post-HAART era.145 Nevertheless, immunodeficiency remains a major risk factor for lymphoma and the WHO classification recognizes four broad categories of immunodeficiency associated lymphoproliferative disorders (IALD) (see Table 1), and these are covered in detail elsewhere.1 The most common primary immunodeficiency states associated with lymphoproliferative disorders are ataxia telangiectasia, Wiskott–Aldrich syndrome, common variable immunodeficiency, severe combined immunodeficiency, and X-linked lymphoproliferative disorder.1 Of some interest is the finding that T cell lymphomas are more common in ataxia telangiectasia than B cell tumors which are more commonly found associated with the other immunodeficiency states. Nevertheless, because the immunodeficiency states are relatively rare disorders the overall incidence of associated lymphomas is low. The overall prognosis is generally poor as these disorders are clinically aggressive, and the added morbidity of the underlying immunodeficiency state adds to the complexity of treatment, but allogeneic transplantation may be helpful.146
565
HIV-associated lymphomas are a complex and heterogeneous topic and several excellent reviews exist.147,148 The common histological types are diffuse large B cell lymphoma (often involving the CNS), BL, PEL, and plasmablastic lymphoma of the oral cavity, and while major therapeutic advances have been made,145 many challenges remain,148 particularly primary CNS lymphoma. The risks associated with solid organ transplantation are numerous and include the development of NHL – the socalled post transplant lymphoproliferative disorder (PTLD). The highest relative risk (RR) appears to be in recipients of heart-lung transplants (RR 239.5), although in absolute terms there are more cases associated with kidney grafts because of the greater number of renal transplants.149 It is interesting to note that a pattern of preferential lymphoma localization has been noted near the transplant, and this appears to be of prognostic significance. The mortality rate is surprisingly high, with 12-month mortality rates of 40% in renal transplants and 50% in heart transplant patients,149 and treatment options including the use of rituximab have been recently canvassed.150 Whether there is an increased risk of lymphoma in patients treated with methotrexate remains controversial, but a prospective study in France indicated that the incidence of NHL was not significantly increased in patients with rheumatoid arthritis treated with methotrexate, whereas the incidence of HL appeared to be higher than in the French population.151
CONCLUSION Lymphoma remains a fascinating and enigmatic disease with a wide variety of pathological and clinical presentations that continue to enthral its devotees. This chapter provides only a snapshot.
REFERENCES 1. Jaffe ES et al., (eds). World Health Organisation Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press, 2001. 2. Harris NL, et al. A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood 1994; 84: 1361 – 92. 3. The Non-Hodgkin’s Lymphoma Classification Project. A clinical evaluation of the International Lymphoma Study Group classification of Non-Hodgkin’s lymphoma. Blood 1997; 89: 3909 – 18. 4. Lapkuviene O, Forchetti D, Roepke JE. Unusual sites of involvement by hematologic malignancies. Case 1. Intravascular large B-cell lymphoma presenting with CNS symptoms. J Clin Oncol 2001; 19: 3988 – 91. 5. Quintini G, et al. Uncommon presentations of non-Hodgkin’s lymphoma: Case 1. Intravascular large B-cell lymphoma: diagnosis on prostate biopsy. J Clin Oncol 2003; 21: 564 – 5. 6. Hodgkin T. On some morbid appearances of the absorbent glands and spleen. Med Chir Trans 1832; 17: 69 – 97. 7. Aisenberg AC. Historical review of lymphomas. Br J Haematol 2000; 109(3): 466 – 76. 8. De Vita VT Jr, Canellos GP. The lymphomas. Semin Hematol 1999; 36(4): Suppl 7 84 – 94. 9. Bonadonna G. Historical view of Hodgkin’s disease. Br J Haematol 2000; 110: 504 – 11. 10. Drillenburg P, Pals ST. Cell adhesion receptors in lymphoma dissemination. Blood 2000; 95: 1900 – 10.
566
HEMATOLOGICAL MALIGNANCIES
11. Tanaka T, et al. Molecular determinants controlling homeostatic recirculation and tissue-specific trafficking of lymphocytes. Int Arch Allergy Immunol 2004; 134: 120 – 34. 12. Jaffe ES. Lymphoid lesions of the head and neck: a model of lymphocyte homing and lymphomagenesis. Mod Pathol 2002; 15: 255 – 63. 13. Bende RJ, et al. Primary follicular lymphoma of the small intestine: alpha4beta7 expression and immunoglobulin configuration suggest an origin from local antigen-experienced B-cells. Am J Pathol 2003; 162: 105 – 13. 14. Maitra A, et al. Precursor B-cell lymphoblastic lymphoma. A study of nine cases lacking blood and bone marrow involvement and review of the literature. Am J Clin Pathol 2001; 115: 868 – 75. 15. Thomas DA, Kantarjian HM. Lymphoblastic lymphoma. Hematol Oncol Clin North Am 2001; 15: 51 – 95. 16. Hoelzer D, Gokbuget N. Treatment of lymphoblastic lymphoma in adults. Baillieres Best Pract Res Clin Haematol 2002; 15: 713 – 28. 17. Pangalis GA, Kyrtsonis MC, Kontopidou FN., et al. Differential diagnosis of Waldenstrom’s macroglobulinemia from other low-grade B-cell lymphoproliferative disorders. Semin Oncol 2003; 30: 201 – 5. 18. Ye H, et al. t(1;14)and t(11;18) in the differential diagnosis of Waldenstrom’s macroglobulinemia. Mod Pathol 2004; 17: 1150 – 4. 19. Iida S, et al. Chromosomal rearrangement of the PAX-5 locus in lymphoblastic lymphoma with t(9;14)(9p13;q32). Leuk Lymphoma 1999; 34(1 – 2): 25 – 33. 20. Cook JR, et al. Lack of PAX-5 rearrangements in lymphoplasmacytic lymphomas: reassessing the reported association with t(9;14). Hum Pathol 2004; 35: 447 – 54. 21. Mazzaro C, et al. Regression of monoclonal B-cell expansion in patients affected by mixed cryoglobulinaemia responsive to alphainterferon therapy. Cancer 1996; 77: 2604 – 13. 22. Ghobrial IM, et al. Initial increase in the cryoglobulin level after rituximab therapy for type II cryoglobulinemia secondary to Waldenstrom’s macroglobulinemia does not indicate failure of response. Am J Hematol 2004; 77: 329 – 30. 23. Franco V, Florena AM, Iannitto E. Splenic marginal zone lymphoma. Blood 2003; 101: 2464 – 72. 24. Oscier D, Owen R, Johnson S. Splenic marginal zone lymphoma. Blood Rev 2005; 19: 39 – 51. 25. Ocio EM, et al. Immunophenotypic and cytogenetic comparison of Waldenstrom’s macroglobulinemia with splenic marginal zone lymphoma. Clin Lymphoma 2005; 5: 241 – 5. 26. Travese-Glehen A, et al. Analysis of VH genes in marginal zone lymphoma reveals marked heterogeneity between splenic and nodal tumors and suggests the existence of clonal selection. Haematologica 2005; 90: 470 – 8. 27. Arcaini L, et al. Splenic and nodal marginal zone lymphomas are indolent disorders at high hepatitis C virus seroprevalence with distinct presenting features but similar morphologic and phenotypic profiles. Cancer 2004; 100: 107 – 15. 28. Parry-Jones N, et al. Prognostic features of splenic lymphoma with villus lymphocytes: a report on 129 patients. Br J Haematol 2003; 120: 759 – 64. 29. Arcaini L, et al. Combination of rituximab, cyclophosphamide, and vincristine induces complete hematologic remission of splenic marginal zone lymphoma. Clin Lymphoma 2004; 4: 250 – 2. 30. Jaffe ES. Common threads of mucosa-associated lymphoid tissue lymphoma pathogenesis: From infection to translocation. J Natl Cancer Inst 2004; 96: 571 – 3. 31. Isaacson PG. Update on MALT lymphomas. Baillieres Best Pract Res Clin Haematol 2005; 18(Special Issue 1): 57 – 68. 32. Goteri G, et al. Primary MALT lymphoma of the tongue. Med Oral 2004; 9: 459 – 63. 33. Yamamoto R, et al. A case of primary rectal mucosa-associated lymphoid tissue lymphoma treated by endoscopic mucosal resection. Dig Endosc 2005; 17: 172 – 4. 34. Farinha P, Gascoyne RD. Helicobacter pylori and MALT lymphoma. Gastroenterology 2005; 128: 1579 – 605. 35. Streubel B, et al. T(14;18)(q32;q21) involving IGH and MALT1 is a frequent chromosomal aberration in MALT lymphoma. Blood 2003; 101: 2335 – 9.
36. Schechter NR, Portlock CS, Yahalom J. Treatment of mucosaassociated lymphoid tissue lymphoma of the stomach with radiation alone. J Clin Oncol 1998; 16: 1916 – 21. 37. Aviles A, et al. Mucosa-associated lymphoid tissue (MALT) lymphoma of the stomach: Results of a controlled clinical trial. Med Oncol 2005; 22: 57 – 62. 38. Raderer M, et al. High relapse rate in patients with MALT lymphoma warrants lifelong follow-up. Clin Cancer Res 2005; 11: 3349 – 52. 39. Thieblemont C, de la Fouchardiere A, Coiffier B. Nongastric mucosaassociated lymphoid tissue lymphomas. Clin Lymphoma 2003; 3: 212 – 24. 40. Ferreri AJM, et al. Evidence for an association between Chlamydia psittaci and ocular adnexal lymphoma. J Natl Cancer Inst 2004; 96: 586 – 94. 41. Cerroni L, et al. Infection by Borrelia burgdorferi and cutaneous B-cell lymphoma. J Cutan Pathol 1997; 24: 457 – 61. 42. Lecuit M, et al. Immunoproliferative small intestinal disease associated with Campylobacter jejuni. N Engl J Med 2004; 350: 239 – 41. 43. Ambrosetti A, et al. Most cases of primary salivary mucosaassociated lymphoid tissue lymphoma are associated either with Sjoegren syndrome or hepatitis C virus infection. Br J Haematol 2004; 126: 43 – 9. 44. Tsang RW, et al. Localised mucosa-associated lymphoid tissue lymphoma treated with radiation therapy has excellent clinical outcome. J Clin Oncol 2003; 21: 4157 – 64. 45. Conconi A, et al. Clinical activity of rituximab in extranodal marginal zone B-cell lymphoma of MALT type. Blood 2003; 102: 2741 – 5. 46. Armitage JO, Weissenberger DD. New approach to classifying non-Hodgkin’s lymphoma: clinical features of the major histologic subtypes. Non-Hodgkin’s Lymphoma Classification Project. J Clin Oncol 1998; 16: 2780 – 95. 47. Savage KJ, et al. The molecular signature of mediastinal large B-cell lymphoma differs from that of other diffuse large B-cell lymphomas and shares features with classical Hodgkin lymphoma. Blood 2003; 102: 3871 – 9. 48. Calvo KR, et al. Molecular profiling provides evidence of primary mediastinal large B-cell lymphoma as a distinct entity related to classic Hodgkin lymphoma: implications for mediastinal gray zone lymphomas as an intermediate form of B-cell lymphoma. Adv Anat Pathol 2004; 11: 227 – 38. 49. Seidemann K et al., NHL Berlin-Frankfurt-Munster Group. Primary mediastinal large B-cell lymphoma with sclerosis in pediatric and adolescent patients: treatment and results from three therapeutic studies of the Berlin-Frankfurt-Munster Group. J Clin Oncol 2003; 21: 1782 – 9. 50. Tateishi U, et al. Primary mediastinal lymphoma: characteristic features of the various histological subtypes on CT. J Comput Assist Tomogr 2004; 28: 782 – 9. 51. Calaminici M, et al. CD23 expression in mediastinal large B-cell lymphomas. Histopathology 2004; 45: 619 – 24. 52. Todeschini G, et al. Primary mediastinal large B-cell lymphoma (PMLBCL): long-term results from a retrospective multicentre Italian experience in 138 patients treated with CHOP or MACOP-B/VACOPB. Br J Cancer 2004; 90: 372 – 6. 53. Cairoli R, et al. Efficacy of an early intensification treatment integrating chemotherapy, autologous stem cell transplantation and radiotherapy for poor risk primary mediastinal large B cell lymphoma with sclerosis. Bone Marrow Transplant 2002; 29: 473 – 7. 54. Zinzani PL et al., International Extranodal Lymphoma Study Group (IELSG). Induction chemotherapy strategies for primary mediastinal large B-cell lymphoma with sclerosis: a retrospective multinational study on 426 previously untreated patients. Haematologica 2002; 87: 1258 – 64. 55. Pfleger VL, Tappeiner J. Zur kenntnis der systemiserten endotheliomatose der cutanen blutgefasse (reticulendotheliose?). Hautarzt 1959; 10: 359 – 63. 56. Imai H, et al. Intravascular large B-cell lymphoma presenting with mass lesions in the central nervous system: a report of five cases. Pathol Int 2004; 54: 231 – 6. 57. Ferreri AJ et al., International Extranodal Lymphoma Study Group (IELSG). Intravascular lymphoma: clinical presentation, natural
RARE LYMPHOMAS
58. 59.
60. 61.
62. 63.
64.
65.
66.
67.
68. 69. 70. 71. 72.
73.
74.
75.
76. 77. 78.
79.
80.
history, management and prognostic factors in a series of 38 cases, with special emphasis on the ‘cutaneous variant’. Br J Haematol 2004; 127: 173 – 83. Monteiro M, et al. Intravascular large B-cell lymphoma of the breast. Breast 2005; 14: 75 – 8. Yamada N, et al. CD5 + Epstein-Barr virus-positive intravascular large B-cell lymphoma in the uterus co-existing with huge myoma. Am J Hematol 2005; 78: 221 – 4. Tokura T, et al. Asian variant of CD5 + intravascular large B-cell lymphoma with splenic infarction. Intern Med 2003; 42: 105 – 9. Seki K, et al. Prostatic acid phosphatase is a possible tumor marker for intravascular large B-cell lymphoma. Am J Surg Pathol 2004; 28: 1384 – 8. Ascoli V, Lo-Coco F. Body Cavity Lymphoma. Curr Opin Pulm Med 2002; 8: 317 – 22. Cesarman E, et al. Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas. N Engl J Med 1995; 332: 1186 – 91. Hamoudi R, et al. Distinct cellular origins of primary effusion lymphoma with and without EBV infection. Leuk Res 2004; 28: 333 – 8. Gaidano G, et al. Genetic characterization of HHV-8/KSHV-positive primary effusion lymphoma reveals frequent mutations of BCL6: implications for disease pathogenesis and histogenesis. Genes Chromosomes Cancer 1999; 24: 16 – 23. Simonelli C, et al. Clinical features and outcome of primary effusion lymphoma in HIV-infected patients: a single-institution study. J Clin Oncol 2003; 21: 3948 – 54. Waddington TW, Aboulafia DM. Failure to eradicate AIDS-associated primary effusion lymphoma with high-dose chemotherapy and autologous stem cell reinfusion: case report and literature review. AIDS Patient Care STDS 2004; 18: 67 – 73. Luppi M, et al. Treatment of herpesvirus associated primary effusion lymphoma with intracavity cidofovir. Leukemia 2005; 19: 473 – 6. Hecht JL, Aster JC. Molecular biology of Burkitt’s lymphoma. J Clin Oncol 2000; 18: 3707 – 21. Cairo MS, Bishop M. Tumour lysis syndrome: new therapeutic strategies and classification. Br J Haematol 2004; 127: 3 – 11. Kasamon YL, Swinnen LJ. Treatment advances in adult Burkitt lymphoma and leukemia. Curr Opin Oncol 2004; 16: 429 – 35. Rizzieri DA, et al. Intensive chemotherapy with and without cranial radiation for Burkitt leukemia and lymphoma: final results of Cancer and Leukemia Group B Study 9251. Cancer 2004; 100: 1438 – 48. Van Imhoff GW, et al. Short intensified therapy and autologous stem cell transplantation in adult Burkitt lymphoma. Excellent results without high-dose MTX. Blood 2002; 100: 182a, Abstract 681. Little RF, et al. Highly effective treatment of acquired immunodeficiency syndrome-related lymphoma with dose-adjusted EPOCH: impact of antiretroviral therapy suspension and tumor biology. Blood 2003; 101: 4653 – 9. Kaplan LD, Lee J, Scadden DT. No benefit from rituximab in a randomized phase lll trial of CHOP with or without rituximab for patients with HIV-associated non-Hodgkins lymphoma: updated data from AIDS Malignancies Consortium Study 010. Blood 2003; 102: 409a, Abstract 1488. Nava V, Jaffe ES. The pathology of NK-Cell lymphomas and leukemias. Adv Anat Pathol 2005; 12: 27 – 34. Jaffe ES. Mature T-cell and NK-cell lymphomas in the pediatric age group. Am J Clin Pathol 2004; 122(Suppl): S110 – 21. Dabaja BS, et al. The role of local radiation therapy for mediastinal disease in adults with T-cell lymphoblastic lymphoma. Cancer 2002; 94: 2738 – 44. Sweetenham JW, et al. High-dose therapy and autologous stem-cell transplantation versus conventional – dose consolidation/maintenance therapy as post remission therapy for adult patients with lymphoblastic lymphoma: results of a randomized trial of the European Group for Blood and Marrow Transplantation and the United Kingdom Lymphoma Group. J Clin Oncol 2001; 19: 2927 – 36. Petrella T, et al. CD4+ CD56+ cutaneous neoplasms: a distinct hematological entity? Groupe Francais d’Etude des Lymphomes Cutanes (GFELC). Am J Surg Pathol 1999; 23: 137 – 46.
567
81. Chaperot L, et al. Identification of a leukemic counterpart of the plasmacytoid dendritic cells. Blood 2001; 97: 3210 – 7. 82. Choi YL, et al. DNA microarray analysis of natural killer cell-type lymphoproliferative disease of granular lymphocytes with purified CD3-CD56+ fractions. Leukemia 2004; 18: 556 – 65. 83. Jacob MC, et al. CD4+ CD56+ lineage negative malignancies: a new entity developed from malignant early plasmacytoid dendritic cells. Haematologica 2003; 88: 941 – 55. 84. Shapiro M, et al. Complete remission in advanced blastic NKcell lymphoma/leukemia in elderly patients using the hyper-CVAD regimen. Am J Hematol 2003; 74: 46 – 51. 85. Hyakuna N, et al. Childhood blastic NK cell leukemia successfully treated with L-asparagenase and allogeneic bone marrow transplantation. Pediatr Blood Cancer 2004; 42: 631 – 4. 86. Anderson JR, Armitage JO, Weissenberger DD. Epidemiology of the non-Hodgkin’s lymphomas: distribution of the major subtypes differ by geographic locations. Non-Hodgkin’s Lymphoma Classification Project. Ann Oncol 1998; 9: 717 – 20. 87. Nicot C. Current views in HTLV-1 associated adult T-cell leukemia/lymphoma. Am J Hematol 2005; 78: 232 – 9. 88. Ohshima K, et al. Survival of patients with HTLV-1 associated lymph node lesions. J Pathol 1999; 189: 539 – 45. 89. Bazarbachi A, et al. New therapeutic approaches for adult T-cell leukaemia. Lancet 2004; 5: 664 – 72. 90. Ishida T, et al. The CC chemokine receptor 4 as a novel specific molecular target for immunotherapy in adult T-Cell leukemia/lymphoma. Clin Cancer Res 2004; 10: 7529 – 39. 91. Tse E, Liang RH. Natural killer cell neoplasms. Clin Lymphoma 2004; 5: 197 – 201. 92. Lin CW, et al. CD94 transcripts imply a better prognosis in nasal-type extranodal NK/T-cell lymphoma. Blood 2003; 102: 2623 – 31. 93. Ng SB, et al. Nasal-type extranodal natural killer/T-cell lymphomas: a clinicopathologic and genotypic study of 42 cases in Singapore. Mod Pathol 2004; 17: 1097 – 107. 94. Chen HH, Hsiao CH, Chiu HC. Hydroa vacciniforme-like primary cutaneous CD8-positive T-cell lymphoma. Br J Dermatol 2002; 147: 587 – 91. 95. Isaacson PG, Du MQ. Gastrointestinal lymphoma: where morphology meets molecular biology. J Pathol 2005; 205: 255 – 74. 96. Smedby KE, et al. Malignant lymphomas in coeliac disease: evidence of increased risks for lymphoma types other than enteropathy-type T cell lymphoma. Gut 2005; 54: 54 – 9. 97. Obermann EC, et al. Loss of heterozygosity at chromosome 9p21 is a frequent finding in enteropathy-type T-cell lymphoma. J Pathol 2004; 202: 252 – 62. 98. Cejkova P, et al. Amplification of NOTCH1 and ABL1 gene loci is a frequent aberration in enteropathy-type T-cell lymphoma. Virchows Arch 2005; 446: 416 – 20. 99. Hoffmann M, et al. 18F-fluoro-deoxy-glucose positron emission tomography (18F-FDG-PET) for assessment of enteropathy-type Tcell lymphoma. Gut 2003; 52: 347 – 51. 100. Savage KJ, et al. Characterization of peripheral T-cell lymphomas in a single North American institution by the WHO classification. Ann Oncol 2004; 15: 1467 – 75. 101. Gaulard P, Belhadj K, Reyes F. Gammadelta T-cell lymphomas. Semin Hematol 2003; 40: 233 – 43. 102. Weidmann E. Hepatosplenic T cell lymphoma. A review on 45 cases since first report describing the disease as a distinct lymphoma entity in 1990. Leukemia 2000; 14: 991 – 7. 103. Macon WR, et al. Hepatosplenic [alpha][beta] T-cell lymphomas. A report of 14 cases and comparison with hepatosplenic [gamma][delta] T-cell lymphomas. Am J Surg Pathol 2001; 25: 285 – 96. 104. Coventry S, et al. Consistency of isochromosome 7q and trisomy 8 in hepatosplenic gammadelta T-cell lymphoma: detection by fluorescence in situ hybridization of a splenic touch preparation from a pediatric patient. Pediatr Dev Pathol 1999; 2: 478 – 83. 105. Kim EJ, et al. Immunopathogenesis and therapy of cutaneous T cell lymphoma. J Clin Invest 2005; 115: 798 – 812. 106. Sokolowska-Wojdylo M, et al. Circulating clonal CLA(+) and CD4(+) T cells in Sezary syndrome express the skin-homing chemokine receptors CCR4 and CCR10 as well as the lymph nodehoming chemokine receptor CCR7. Br J Dermatol 2005; 152: 258 – 64.
568
HEMATOLOGICAL MALIGNANCIES
107. Vonderheid EC, et al. Evidence for restricted Vbeta usage in the leukemic phase of cutaneous T cell lymphoma. J Invest Dermatol 2005; 124: 651 – 61. 108. Hinshaw MA, Wood GS. cDNA microarrays: revolutionary technology for the diagnosis, prognosis and treatment of cutaneous T-cell lymphoma. Int J Derm 2005; 44: 181 – 3. 109. Willemze R, et al. WHO-EORTC classification for cutaneous lymphomas. Blood 2005; 105: 3768 – 85. 110. Go RS, Wester SM. Immunophenotypic and molecular features, clinical outcomes, treatments and prognostic factors associated with subcutaneous panniculitis-like T-cell lymphoma: a systematic analysis of 156 patients reported in the literature. Cancer 2004; 101: 1404 – 13. 111. Bekkenk MW, et al. CD56+ hematological neoplasms presenting in the skin: a retrospective analysis of 23 new cases and 130 cases from the literature. Ann Oncol 2004; 15: 1097 – 108. 112. Foss F. Mycosis fungoides and the S´ezary syndrome. Curr Opin Oncol 2004; 16: 421 – 8. 113. Pawlaczyk M, et al. No evidence of HTLV-I infection in patients with mycosis fungoides and S´ezary syndrome. Neoplasma 2005; 52(1): 52 – 5. 114. Diamandidou E, et al. Transformation of mycosis fungoides/S´ezary syndrome: clinical characteristics and prognosis. Blood 1998; 92: 1150 – 9. 115. Fink-Puches R, et al. Primary cutaneous lymphomas:applicability of current classification schemes (European Organization for Research and Treatment of Cancer, World Health Organization) based on clinicopathologic features observed in a large group of patients. Blood 2002; 99: 800 – 5. 116. Willemze R, Meijer CJ. Primary cutaneous CD30-positive lymphoproliferative disorders. Hematol Oncol Clin North Am 2003; 17: 1319 – 32. 117. Kumar S, et al. Primary cutaneous CD-30 positive anaplastic large cell lymphoma in childhood: report of 4 cases and review of the literature. Pediatr Dev Pathol 2005; 8: 52 – 60. 118. Shehan JM, et al. Management of multifocal primary cutaneous CD30 anaplastic large cell lymphoma. J Am Acad Dermatol 2004; 51: 103 – 10. 119. Bekkenk MW, et al. Primary and secondary cutaneous CD30+ lymphoproliferative disorders: a report from the Dutch Cutaneous Lymphoma Group on the long-term follow-up data of 219 patients and guidelines for diagnosis and treatment. Blood 2000; 95: 3653 – 61. 120. Liu HL, et al. CD30+ cutaneous lymphoproliferative disorders: the Stanford experience in lymphomatoid papulosis and primary cutaneous anaplastic large cell lymphoma. J Am Acad Dermatol 2003; 49: 1049 – 58. 121. Dogan A, Attygalle AD, Kyriakou C. Angioimmunoblastic T-cell lymphoma. Br J Haematol 2003; 121: 681 – 91. 122. Ishida T, et al. CXC chemokine receptor 3 and CC chemokine receptor 4 expression in T-cell and NK-cell lymphomas with special reference to clinicopathological significance for peripheral T-cell lymphoma, unspecified. Clin Cancer Res 2004; 10: 5494 – 500. 123. Attygalle AD, et al. CD10 expression in extranodal dissemination of angioimmunoblastic T-cell lymphoma. Am J Surg Pathol 2004; 28: 54 – 61. 124. Schetelig J, et al. Long-term disease-free survival in patients with angioimmunoblastic T-cell lymphoma after high dose chemotherapy and autologous stem cell transplantation. Haematologica 2003; 88: 1272 – 8. 125. Pulford K, Morris SW, Turturro F. Anaplastic lymphoma kinase proteins in growth control and cancer. J Cell Physiol 2004; 199: 330 – 58. 126. Kutok JL, Aster JC. Molecular biology of anaplastic lymphoma kinase-positive anaplastic large-cell lymphoma. J Clin Oncol 2002; 20: 3691 – 702. 127. Wellmann A, et al. Detection of differentially expressed genes in lymphomas using cDNA arrays: identification of clusterin as a new
128. 129.
130.
131. 132. 133. 134. 135.
136. 137. 138.
139.
140.
141. 142. 143.
144. 145.
146. 147. 148.
149.
150.
151.
diagnostic marker for anaplastic large-cell lymphoma. Blood 2000; 96: 398 – 404. Kadin ME, Carpenter C. Systemic and primary cutaneous anaplastic large cell lymphomas. Semin Hematol 2003; 40: 244 – 56. Weissenberger DD, et al. Systemic anaplastic large-cell lymphoma: results from the non-Hodgkin’s lymphoma classification project. Am J Hematol 2001; 67: 172 – 8. Gaudet G, et al. Breast lymphoma associated with breast implants: two case-reports and a review of the literature. Leuk Lymphoma 2002; 43: 115 – 9. Re D, et al. From Hodgkin disease to Hodgkin lymphoma: biologic insights and therapeutic potential. Blood 2005; 105: 4553 – 60. Cartwright RA, Watkins G. Epidemiology of Hodgkin’s disease: a review. Hematol Oncol 2004; 22: 11 – 26. Venizelos I, et al. Primary gastric Hodgkin’s lymphoma: a case report and review of the literature. Leuk Lymphoma 2005; 46: 147 – 50. Valbuena JR, et al. Classical Hodgkin lymphoma arising in the rectum. Ann Diagn Pathol 2005; 9: 38 – 42. Gebert C, et al. Primary multifocal osseous Hodgkin disease: a case report and review of the literature. J Cancer Res Clin Oncol 2005; 131: 163 – 8. Warburton G, et al. Hodgkin’s lymphoma: a case report involving the mandible. J Oral Maxillofac Surg 2003; 16: 1492 – 6. Hirmiz K, et al. Intracranial presentation of systemic Hodgkin’s disease. Leuk Lymphoma 2004; 45: 1667 – 71. Kinai B, Magro CM, Ross P. Endobronchial presentation of Hodgkin lymphoma: a review of the literature. Ann Thorac Surg 2003; 76: 967 – 72. Aki H, et al. T-cell-rich B-cell lymphoma: a clinicopathologic study of 21 cases and comparison with 43 cases of diffuse large B-cell lymphoma. Leuk Res 2004; 28: 229 – 36. Diehl V, et al. Clinical presentation, course and prognostic factors in lymphocyte-predominant Hodgkin’s disease and lymphocyte-rich classical Hodgkin’s disease: report from the European Task Force on Lymphoma Project on Lymphocyte-Predominant Hodgkin’s Disease. J Clin Oncol 1999; 17: 776 – 83. Schlembach PJ, et al. Radiotherapy alone for lymphocyte-predominant Hodgkin’s disease. Cancer J 2002; 8: 377 – 83. Ekstrand BC, et al. Rituximab in lymphocyte-predominant Hodgkin disease: results of a phase 2 trial. Blood 2003; 101: 4285 – 9. Anagnostopoulos I, et al. European Task Force on Lymphoma project on lymphocyte predominance Hodgkin disease: histologic and immunohistologic analysis of submitted cases reveals 2 types of Hodgkin disease with a nodular growth pattern and abundant lymphocytes. Blood 2000; 96: 1889 – 99. Trempat P, et al. Gene expression profiling in anaplastic large cell lymphoma and Hodgkin’s disease. Leuk Lymphoma 2004; 45: 2001 – 6. Robotin MC, et al. Clinical features and predictors of survival of AIDS-related non-Hodgkin’s lymphoma in a population-based case series in Sydney, Australia. HIV Med 2004; 5: 377 – 84. Hadzic N, et al. Correction of the hyper-IgM syndrome after liver and bone marrow transplantation. N Engl J Med 2000; 342: 320 – 4. Noy A. Update in HIV-associated lymphoma. Curr Opin Oncol 2004; 16: 450 – 4. Aboulafia DM, Pantanowitz L, Dezube BJ. AIDS-related non-Hodgkin lymphoma: still a problem in the era of HAART. AIDS Read 2004; 14: 605 – 17. Opalz G, D¨ohler B. Lymphoma after solid organ transplantation: a collaborative transplant study Report. Am J Transplant 2003; 4: 222 – 30. Davis JE, Moss DJ. Treatment options for post-transplant lymphoproliferative disorder and other Epstein-Barr virus-associated malignancies. Tissue Antigens 2004; 63: 285 – 92. Mariette X, et al. Lymphomas in rheumatoid arthritis patients treated with methotrexate: a 3-year prospective study in France. Blood 2002; 99: 3909 – 15.
Section 8 : Hematological Malignancies
50
Uncommon Presentations of Plasma Cell Dyscrasias Rachid Baz and Mohamad A. Hussein
PLASMA CELL LEUKEMIA Biology Primary plasma cell leukemia (PCL) refers to the de novo presence of greater than 20% plasma cells in the peripheral blood and/or an absolute plasma cell count of greater than 2000/µL.1,2 Secondary PCL is thought to represent leukemic transformation of known multiple myeloma, which is thought to be a terminal event and implies a median survival of about 2–3 months.3 While about 40% of PCL is secondary, only 1% of patients with multiple myeloma develop it. PCL accounts for about 4% of patients with multiple myeloma.2,3
Presentation Patients with primary PCL present on average 10 years earlier than those with their myeloma counterpart. PCL has a more aggressive clinical behavior than multiple myeloma. Accordingly, extramedullary involvement, renal dysfunction, severe anemia, and thrombocytopenia are more common at presentation. On the other hand, bony involvement is less likely than with multiple myeloma.2,4,5
Management and Prognosis Consistent with the more aggressive behavior, primary PCL implies a poor prognosis with lower response rates and median survivals of 8–10 months.2,4,5 The treatment of PCL has traditionally been similar to the treatment of multiple myeloma. The more aggressive course often implies a more intensive chemotherapy regimen than melphalan and prednisone. Nonrandomized retrospective studies suggest a possibly improved outcome with autologous stem cell transplantation.6,7
NONSECRETORY MULTIPLE MYELOMA Biology Nonsecretory multiple myeloma accounts for 1 to 5% of patients with multiple myeloma.3 Differences in its prevalence are probably related to referral biases and differences
in the diagnostic evaluations among different series.8 – 10 Diagnostic and therapeutic challenges in the management of patients with nonsecretory multiple myeloma are mainly related to the absence of a clear disease parameter to follow. The labeling of “nonsecretory” is used to define patients who do not have an adequate amount of monoclonal protein to follow i.e., patients who demonstrate positive monoclonal protein by immunofixation and yet lack any significant M protein as measured by serum protein electrophoresis or 24-hour urinalysis. Nonsecretory multiple myeloma is defined by the absence of monoclonal protein in the serum or urine in a patient with other characteristic signs and symptoms of multiple myeloma (including increased plasma cell in the bone marrow or lytic bone lesions).3 Immunofluorescence studies must be performed in all patients in whom this diagnosis is suspected and in large subset of patients, a cytoplasmic immunoglobulin is identified. The remainder of patients are defined as having truly nonsecretory or non-producer multiple myeloma. Several hypotheses have been proposed for the lack of an M protein in patients with nonsecretory multiple myeloma, including the inability to excrete the immunoglobulin out of the plasma cell, production of a rapidly degradable immunoglobulin and low synthetic capability of the plasma cell clone.3
Presentation Dreicer and Alexanian reported on the presenting features of 29 patients with nonsecretory multiple myeloma.10 They noted a lower age at diagnosis as compared to patients with a monoclonal protein. Patients were also less likely to have anemia, renal dysfunction, or hypercalcemia at presentation.10 Other smaller series did not identify differences in the presentation of patients with nonsecretory myeloma, as compared to patients with an M protein with the exception of the lack of renal dysfunction at the time of presentation.8,9 Unpublished data from the Southwest Oncology Group show that approximately 5% of multiple myeloma
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
570
HEMATOLOGICAL MALIGNANCIES
patients will have monoclonal protein detected mostly by immunofixation, and the main presenting feature is skeletal disease. Such patients are known to have a better survival, despite significant skeletal morbidity.
noted chromosome 13q deletions (a poor prognostic factor in patients with classical multiple myeloma) in three out of the nine tested patients. Other series did not report on the frequency of chromosome 13q deletions.14
Management and Prognosis
Management and Prognosis
The natural history of patients with nonsecretory multiple myeloma is not thought to be different from other patients with multiple myeloma. In our practice, treatment of nonsecretory multiple myeloma is the same as the treatment of multiple myeloma with a measurable M protein. In this setting, a response to treatment is defined as improvement in the signs and symptoms of bone pains, anemia, hypercalcemia, and the absence of new lytic bone lesions. A decrease in bone marrow plasmacytosis is thought to be an objective marker of response. Conversely, progressive disease is characterized by progressive bony disease and an increase in bone marrow plasmacytosis.3
The survival of patients with IgD multiple myeloma is thought to be less than their counterparts with classical multiple myeloma. This observation may be biased by the difference in the dates of diagnosis in different series, with more contemporary series describing a more similar survival.11 – 14,16 Chemotherapeutic regimens used for the treatment of patients with classical multiple myeloma are thought to have efficacy in the treatment of IgD myeloma. Wechalekar et al. suggest a possible benefit from autologous stem cell transplantation when compared to conventional chemotherapy.14 However, the retrospective, nonrandomized study design and the small sample size may have biased the findings.
IMMUNOGLOBULIN D MYELOMA Biology and Presentation Commonly involved immunoglobulins in multiple myeloma are immunoglobulin G, A or M. Less likely, immunoglobulin D (IgD) is the monoclonal protein in about 2% of patients.11,12 The presenting clinical features of IgD myeloma are similar to those in patients with other immunoglobulin myelomas, with the exception of a few noted associations. As compared to patients with the more common myeloma, patients with IgD myeloma are on average younger at diagnosis (about 54 years of age).11 – 14 Lymph node enlargement, an unusual sign in patients with classical multiple myeloma, was found in about 10% of patients with IgD myeloma.11,12 The prevalence of extramedullary plasmacytomas in patients with IgD myeloma is about 15–20% (higher than patients with classical myeloma).11,12 The electrophoretic pattern of the serum and urine of patients with IgD myeloma is not different from that of patients with light chain multiple myeloma: usually an M spike of less than 2 g/dL, and light chain proteinuria.11,13 The detection of an IgD monoclonal protein in the serum or urine is not synonymous with the diagnosis of IgD myeloma; rare cases of monoclonal gammopathy of undetermined significance (MGUS) with IgD production has been described in literature.15 λ light chains are thought to be more commonly associated with IgD myeloma than are κ light chains. Hence patients with λ light chain myeloma with a discrete M spike on serum electrophoresis should have immunofixation for IgD.3 Wechalekar et al. reported that the monoclonal IgD was detectable by immunofluorescence only in 25 of the 26 patients, with only one patient having measurable serum IgD.14 The same authors noted several associated hematologic disorders in their cohort of patients (total 26 patients) with IgD myeloma: one patient with chronic lymphocytic leukemia, one patient with hair cell leukemia, and three patients with increased bone marrow reticulin stains. They
¨ MACROGLOBULINEMIA WALDENSTROM Biology and Presentation Waldenstr¨om’s macroglobulinemia (WM) is a B cell lymphoproliferative disorder characterized by the production of a monoclonal immunoglobulin of the IgM subtype and by intertrabecular bone marrow infiltration with a lymphoplasmacytic infiltrate.17 The second international workshop on Waldenstr¨om’s macroglobulinemia has proposed the following diagnostic criteria: • an IgM monoclonal protein of any concentration, • bone marrow infiltration with small lymphocytes exhibiting plasmacytoid differentiation, • a suggestive immunophenotype (expression of surface IgM, CD19, CD20, CD25, CD27, FMC7, and CD 138 without the expression of CD5, CD10, CD23, and CD103).18 Symptoms attributable to WM are related to tumor infiltration or to the monoclonal protein. The former results in constitutional symptoms (fevers, sweats, and weight loss), cytopenias (secondary to bone marrow involvement), lymphadenopathy, and hepatosplenomegaly.17,19 Symptoms related to the monoclonal protein include symptoms related to hyperviscosity, cryoglobulinemia, cold agglutinin, neuropathy, and amyloidosis. Hyperviscosity develops because the IgM molecules are large intravascular structures with carbohydrate content that tends to bind to water and increases the resistance to blood flow in the microcirculations.20,21 The Bing-Neel syndrome refers to confusion, memory loss, and eventual coma secondary to long-standing hyperviscosity with resultant increase in vascular permeability and deposition of lymphoplasmacytic cells.22 Increase in cryoglobulins is noted in 20% of patients with WM; however, clinical signs of cryoglobulinemia (skin ulcerations, Raynaud’s phenomenon) occur in only 5% of patients.23,24
UNCOMMON PRESENTATIONS OF PLASMA CELL DYSCRASIAS
Deposition of the M protein in tissues may result in organ dysfunction: such deposits have been described in glomerular loops and may result in proteinuria and uremia which can be reversed by aphaeresis of the monoclonal protein; deposition into the skin results in flesh colored skin papules called macroglobulinemia cutis; and deposits in the gastrointestinal tract may result in malabsorption and diarrhea.17,20,25 Less commonly, amyloidosis may also result from deposition of the M protein in tissues and may result in cardiac dysfunction, neuropathy, and renal dysfunction.26 A distal symmetric peripheral neuropathy may result from auto-antibodies to glycolipid on nerve sheets.27 Occasional patients are diagnosed by chance and are asymptomatic at presentation.28
Management and Prognosis The median survival of patients with WM ranges from 5 to 10 years.17 Prognostic factors identified in several series of patients include the following: patient age, the presence of anemia, leucopenia or thrombocytopenia, serum β2 microglobulin, and the level of the monoclonal protein.29 – 33 Treatment should be instituted in symptomatic patients and should be considered in patients with cytopenias, bulky adenopathy, or visceromegaly. The level of the monoclonal protein should not be used as an indication for treatment.34 Treatment responses, for comparison of clinical trials, have been defined by the second international workshop on WM.35 A complete response is defined as the disappearance of the monoclonal protein, and by the resolution of infiltration of lymph node and visceral organs confirmed on two separate evaluations 6 weeks apart. A partial response is defined as greater than 50% reduction in the monoclonal protein, and greater than 50% reduction in lymphadenopathy with the resolutions of symptoms related to WM. Progressive disease is defined as a greater than 25% increase in the monoclonal protein, worsening of cytopenias, organ infiltration, or disease-related symptoms.35 The evidence underlying the choice of treatment recommendations is mostly based on case series, phase II trials and only two randomized controlled trials. Accordingly, the following guidelines for treatment should be viewed as recommendations and treatment should be individualized whenever possible according to the patient presentation and preference. Consideration for referral to an academic center experienced in the treatment of patients with WM or enrollment in clinical trials should be sought. Plasmapheresis is an effective method for the rapid reduction of the monoclonal protein when it results in hyperviscosity, peripheral neuropathy, or cryoglobulinemia. A single session will often result in greater than 50% reduction in serum viscosity.36,37 The benefits of plasmapheresis are short lived, and it is only indicated in the long term in patients with hyperviscosity who are resistant to systemic treatment. Splenectomy has been found to result in long remission in case reports of patients resistant to systemic treatment.38 The beneficial effects of splenectomy have been attributed to cytoreduction of T cells necessary for the B lymphocyte
571
differentiation into IgM producing cells.17 Splenectomy cannot be routinely recommended in the management of patients with WM, but may be considered in select situations. Traditionally, oral alkylating agents were used for the systemic control of WM. A prospective randomized trial comparing low dose daily chlorambucil to high intermittent dosing did not note a survival advantage to either approach and patients had a median overall survival of approximately 5 years.39 While complete responses are rare with the use of alkylating agents, partial response rates approach 50% in some series. The time to response has been slow with alkylating agents.39 The duration of treatment has not been defined, but it is generally recommended to discontinue treatment after best response has been achieved.17 The use of alkylating agents should be considered in older patients in whom rapid control of the disease is not necessary. Numerous nonrandomized studies have reported on the efficacy of nucleoside analogs (fludarabine or cladribine) and noted response rates range from 30 to 70%.40 – 42 A randomized controlled trial compared treatment with fludarabine to treatment with the combination of Cyclophosphamide, Adriamycin, and prednisone in patients who had failed alkylating agents. Fludarabine was associated with responses in 28% of patients compared to only 11% of patients treated with combination chemotherapy. Survival was similar as nonresponders were allowed to cross over.43 The faster time to response is a major advantage to treatment with nucleoside analogs and these agents are generally indicated in patients when rapid responses are needed.44 Rituximab, a monoclonal anti-CD20 antibody has been used in newly diagnosed and previously treated case series; response rates range from 20–70% in the former and around 30% in the latter group of patients. The time to response in rituximab-treated patients is in the order of 3 months.30,44,45 Accordingly, rituximab is often used as a first-line agent in young patients who may be candidates to autologous stem cell transplantation, and in whom a longer time to response is acceptable. Autologous stem cell transplantation has resulted in high rates of responses (approaching 90%), and lasting responses (progression-free survival approaching 70 months) in small series of patients.46,47 The small number of patients, the nonrandomized nature of the studies and potential for treatmentrelated morbidity makes it difficult to routinely recommend this approach in many patients. It should, however, be considered in younger patients after cytoreductive treatment with rituximab. Treatment with alkylating agents and nucleoside analog may impair the ability to collect stem cells and should be judiciously used in younger patients. In addition, novel therapies involving Thalidomide, Alemtuzumab, Bortezomib, and Sildenafil are undergoing evaluation in clinical trials.
HEAVY CHAIN DISEASE Biology Heavy chain diseases (HCD) are characterized by the production of an abnormal truncated heavy chain with no associated
572
HEMATOLOGICAL MALIGNANCIES
light chain by monoclonal plasma cells.48 The three types of HCD refer to the heavy chain produced: α, γ , and µ. The diagnosis of HCD often requires serum immunofixation since protein electrophoresis often does not detect the monoclonal protein but may occasionally detect depressed or increased immunoglobulins.48
Presentation α HCD is an enteric disease and usually involves secretory IgA. A geographic clustering of reports in the Mediterranean basin has been noted as well as an association with lower socioeconomic status and poor hygiene in immigrants to developed countries.49 Affected patients are usually males in their second and third decade of life and present with diarrhea, malabsorption, and abdominal pain. The diagnosis often requires endoscopy with biopsy. Different stages may coexist in a particular patient. Stage A involves the presence of a mature plasmacytic infiltrate in the intestinal mucosa and usually responds to longterm antimicrobial treatment. Stage B is characterized by the involvement of the submucosa, while stage C involves extensive involvement of the bowels. Patients with stage B or C may benefit from antimicrobial treatment as well as an anthracycline-based combination chemotherapy regimen.48 The presentation of γ HCD is one of a lymphoproliferative disorder with lymphadenopathy and constitutional symptoms but without a specific histopathologic pattern (often lymphoplasmacytic infiltration).50 The diagnosis requires a consistent immunofixation pattern. An association with autoimmune conditions has been described.51
Treatment and Prognosis Treatment recommendations vary according to the course of the disease, with some patients having an indolent course and occasionally spontaneous regression and others with an aggressive course requiring treatment with systemic chemotherapy (often Melphalan and Prednisone).48,51 µ HCD is a very rare disorder of adults and often coexists with a chronic lymphocytic leukemia. While lymphadenopathy is less common, hepatosplenomegaly is common. The diagnosis requires a consistent immunofixation as well. The course is of a slowly progressive disease. Treatment involves the treatment of the underlying lymphoproliferative disorder.48
OSTEOSCLEROTIC MYELOMA Presentation Osteosclerotic myeloma, also known in the medical literature as polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy, and skin changes (POEMS) syndrome, Crow-Fukase syndrome, serum protein electrophoresis (PEP) syndrome, or Takatsuki syndrome, represents a constellation of findings, including the following: polyneuropathy, organomegaly, endocrinopathy, the presence of a monoclonal protein, and a variety of dermatologic findings.52,53 Other associated findings include sclerotic bony
lesions, Castleman’s disease, papilledema, erythrocytosis, and thrombocytosis.52,54,55 Patients frequently present in the fifth decade of life, often with signs and symptoms of polyneuropathy.52 Sensory deficits in the lower extremities occur early and are followed by motor deficits, which may result in significant disabilities. While no characteristic electromyographic findings have been identified, slowing of nerve conduction velocity and decreased muscle action potentials are noted.56,57 An increase in the cerebrospinal fluid protein is noted in the overwhelming majority of patients.27,56 Hepatosplenomegaly or lymphadenopathy is notable in about half the patients. Castleman’s disease and reactive lymphadenopathy were the diagnoses rendered on biopsies of lymph in patients with POEMS syndrome.58 Endocrinopathies associated with the diagnosis of POEMS syndrome have included hypogonadism, hypothyroidism, or dysfunction of the pituitary adrenal axis.52 Evidence of a monoclonal plasma cell dyscrasia was noted in all patients of one series, approximately 90% of which have evidence of a monoclonal protein in the serum or urine, while the remainder had evidence of monoclonal plasmacytosis on biopsy specimen.52 Reported skin changes associated with the diagnosis of POEMS syndrome include hyperpigmentation, acrocyanosis, hypertrichosis, angiomas, and plethora.59 Most patients with POEMS syndrome have demonstrable bony lesions on x-rays, a mixture of sclerotic and lytic lesions or only sclerotic lesions.52 About half the patients present with a single lesion, while the other have multiple lesions. Hypercalcemia was not associated with bony lesions. Renal dysfunction occasionally requiring dialysis has been described at diagnosis or during the course of the disease.60 Arterial or venous thrombotic events (including myocardial infarction, strokes, Budd-Chiari syndrome) have been described in a number of series, which suggest a greater prevalence than expected by chance.52 Pulmonary arterial hypertension has been observed in a small series of patients.61 Erythrocytosis was more common than anemia in one series of patients, and leukocytosis and thrombocytosis have also been reported. Only about 30% of patients with osteosclerotic myeloma present with the five classical findings of the POEMS acronym).52 Dispenzieri et al. have proposed the following diagnostic criteria for the diagnosis of POEMS syndrome. Patients must have evidence of polyneuropathy, and a monoclonal plasma cell dyscrasia as well as one of the following symptoms: sclerotic bone lesions, evidence of Castleman’s disease on a lymph node biopsy, organomegaly, endocrinopathy (with the exception of diabetes or hypothyroidism), evidence of volume overload, papilledema or consistent skin changes, and thrombocytosis.52 Because of the protean nature of the findings in patients with osteosclerotic myeloma and the independent multidisciplinary evaluation, the clinician must have a high index of suspicion in patients presenting with unexplained polyneuropathy and evidence of a monoclonal plasma cell disorder.
UNCOMMON PRESENTATIONS OF PLASMA CELL DYSCRASIAS
Treatment and Prognosis The natural history of patients with POEMS syndrome is frequently one of progressive neurologic deterioration. The median survival of patients exceeds a decade in many series.52,54,55 Adverse prognostic factors include evidence of volume overload or clubbing.52 Patients with single or multiple localized osteosclerotic lesions benefit from radiation therapy to the area (often 40–50 Gy of radiation are delivered). Improvement in the neuropathy is often slow after radiation therapy. Patients with widespread lesions require systemic therapy.52 Therapeutic agents that have been used in this setting include melphalan and prednisone, combination chemotherapy such as cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) or vincristine, doxorubicin and dexamethasone (VAD), as well as corticosteroids as single agents.52,55,62 Reports of responses to autologous stem cell transplantation have been described.63 An evidence-based approach for the choice of the systemic therapy cannot be recommended due to the paucity of structured, available data.
SOLITARY PLASMACYTOMAS Presentation Plasmacytomas are not infrequent in patients with multiple myeloma and have been described in virtually every organ. Solitary plasmacytoma of bone refers to a single bony lesion composed of monoclonal plasma cells without evidence of a systemic multiple myeloma. Solitary plasmacytoma of the bone (SPB) accounts for about 5% of patients with plasma cell dyscrasias.64 Patients present in their fifth decade of life with a male predominance. Patients with solitary plasmacytoma of bone usually present with bony pain related to a skeletal lesion.65 Occasionally, signs and symptoms of spinal cord compression may be present at the time of diagnosis. Involvement of virtually every bony structure has been described although vertebral involvement is most common. Appearance on plain radiograph is that of a lytic lesion. A skeletal survey must not show involvement of other bony structures, while a serum protein electrophoresis occasionally shows a low concentration of a monoclonal protein with preservation of the uninvolved immunoglobulins.66 A bone marrow biopsy should be performed to rule out a systemic myeloma. In addition, the patient must not have evidence of anemia, hypercalcemia, or renal dysfunction related to the monoclonal protein.66 While magnetic resonance imaging has been used and reported to identify additional skeletal abnormalities in patients, it is not an integral part of the workup of patients.66
Treatment and Prognosis The median survival of patients with SPB is around 10 years, mostly related to the fact that about half of the patients eventually develop multiple myeloma after a median time of 2 years.67 – 69 The development of multiple myeloma has been described as late as 15 years after radiotherapy. Several series have attempted to identify predictors of progression
573
to multiple myeloma, but results are difficult to duplicate in view of the small number of patients included and the differences in diagnostic criteria. The following are such predictors of progression: advanced age, axial lesions, larger lesions, and the lack of resolution of the monoclonal protein following radiotherapy.68,70,71 The treatment of SPB has been definitive local radiotherapy. Local control rates exceed 90% with this modality alone.67 – 69,72,73 Recurrences tend to cluster in the first 3 years after the original diagnosis. The use of surgery followed by radiotherapy is occasionally needed for the patients who present with acute neurologic dysfunction or who require prophylactic internal fixation. Radiotherapy often results in the disappearance of the monoclonal protein from the serum of patients. At present, there is no defined role systemic therapy.65 Solitary extramedullary plasmacytoma refers to involvement of a non-marrow containing organ by monoclonal plasma cells.65 It accounts for less than 3% of patients with plasma cell dyscrasias.70 Most lesions have been reported in the head and neck, but involvement of any other organs has been reported as well.74 The presenting symptoms are mostly related to the site of involvement: submucosal nasal involvement may result in nasal discharge, epistaxis, or nasal obstruction.74 Occasionally plasmacytomas are asymptomatic and detected on imaging studies ordered for other conditions, as in plasmacytoma of the lungs. Involvement of adjacent lymph nodes and bone is still consistent with the localized nature of the disease.75 Treatment involves definitive radiotherapy and is associated with an excellent local control rate.76 Progression to multiple myeloma occurs in about 15%.75 Similar to SPB, there is no role for systemic therapy.66 It is to be stressed that an extensive and careful workup is necessary to rule out the presence of systemic disease, wherein the latter local radiation therapy is given with a noncurative intent i.e., lower dose.
SUMMARY There are several variants of myeloma and plasma cell dyscrasias. It is important to characterize these carefully at a clinical, pathological, and molecular level, as these factors as well as the natural history of these variants will define the nature of treatment and prognosis. There is overlap with some of the syndromes of presentation of rare lymphomas, and this chapter should be read in conjunction with the accompanying presentation on rare lymphomas (see Chapter 49).
REFERENCES 1. Costello R, et al. Primary plasma cell leukaemia: a report of 18 cases. Leuk Res 2001; 25(2): 103 – 7. 2. Dimopoulos MA, et al. Primary plasma cell leukaemia. Br J Haematol 1994; 88(4): 754 – 9. 3. Blade J, Kyle RA. Nonsecretory myeloma, immunoglobulin D myeloma, and plasma cell leukemia. Hematol Oncol Clin North Am 1999; 13(6): 1259 – 72. 4. Garcia-Sanz R, et al. Primary plasma cell leukemia: clinical, immunophenotypic, DNA ploidy, and cytogenetic characteristics. Blood 1999; 93(3): 1032 – 7.
574
HEMATOLOGICAL MALIGNANCIES
5. Noel P, Kyle RA. Plasma cell leukemia: an evaluation of response to therapy. Am J Med 1987; 83(6): 1062 – 8. 6. Hovenga S, et al. Consolidation therapy with autologous stem cell transplantation in plasma cell leukemia after VAD, high-dose cyclophosphamide and EDAP courses: a report of three cases and a review of the literature. Bone Marrow Transplant 1997; 20(10): 901 – 4. 7. Saccaro S, et al. Primary plasma cell leukemia: report of 17 new cases treated with autologous or allogeneic stem-cell transplantation and review of the literature. Am J Hematol 2005; 78(4): 288 – 94. 8. Bourantas K. Nonsecretory multiple myeloma. Eur J Haematol 1996; 56(1 – 2): 109 – 11. 9. Cavo M, et al. Nonsecretory multiple myeloma. Presenting findings, clinical course and prognosis. Acta Haematol 1985; 74(1): 27 – 30. 10. Dreicer R, Alexanian R. Nonsecretory multiple myeloma. Am J Hematol 1982; 13(4): 313 – 8. 11. Blade J, Lust JA, Kyle RA. Immunoglobulin D multiple myeloma: presenting features, response to therapy, and survival in a series of 53 cases. J Clin Oncol 1994; 12(11): 2398 – 404. 12. Jancelewicz Z, et al. IgD multiple myeloma. Review of 133 cases. Arch Intern Med 1975; 135(1): 87 – 93. 13. Sinclair D. IgD myeloma: clinical, biological and laboratory features. Clin Lab 2002; 48(11 – 12): 617 – 22. 14. Wechalekar A, et al. IgD multiple myeloma – a clinical profile and outcome with chemotherapy and autologous stem cell transplantation. Ann Hematol 2005; 84(2): 115 – 7. 15. Blade J, Kyle RA. IgD monoclonal gammopathy with long-term followup. Br J Haematol 1994; 88(2): 395 – 6. 16. Fibbe WE, Jansen J. Prognostic factors in IgD myeloma: a study of 21 cases. Scand J Haematol 1984; 33(5): 471 – 5. 17. Dimopoulos MA, et al. Diagnosis and management of Waldenstrom’s macroglobulinemia. J Clin Oncol 2005; 23(7): 1564 – 77. 18. Owen RG, et al. Clinicopathological definition of Waldenstrom’s macroglobulinemia: consensus panel recommendations from the Second International Workshop on Waldenstrom’s macroglobulinemia. Semin Oncol 2003; 30(2): 110 – 5. 19. Dimopoulos MA, et al. Waldenstrom’s macroglobulinemia: clinical features, complications, and management. J Clin Oncol 2000; 18(1): 214 – 26. 20. Fudenberg HH, Virella G. Multiple myeloma and Waldenstrom macroglobulinemia: unusual presentations. Semin Hematol 1980; 17(1): 63 – 79. 21. Kwaan HC, Bongu A. The hyperviscosity syndromes. Semin Thromb Hemost 1999; 25(2): 199 – 208. 22. Civit T, et al. Waldenstrom’s macroglobulinemia and cerebral lymphoplasmocytic proliferation: Bing and Neel syndrome. Apropos of a new case. Neurochirurgie 1997; 43(4): 245 – 9. 23. Dispenzieri A. Symptomatic cryoglobulinemia. Curr Treat Options Oncol 2000; 1(2): 105 – 18. 24. Farhangi M, Merlini G. The clinical implications of monoclonal immunoglobulins. Semin Oncol 1986; 13(3): 366 – 79. 25. Daoud MS, et al. Monoclonal gammopathies and associated skin disorders. J Am Acad Dermatol 1999; 40(4): 507 – 35; quiz 536 – 8. 26. Gertz MA, Kyle RA, Noel P. Primary systemic amyloidosis: a rare complication of immunoglobulin M monoclonal gammopathies and Waldenstrom’s macroglobulinemia. J Clin Oncol 1993; 11(5): 914 – 20. 27. Ropper AH, Gorson KC. Neuropathies associated with paraproteinemia. N Engl J Med 1998; 338(22): 1601 – 7. 28. Alexanian R, et al. Asymptomatic Waldenstrom’s macroglobulinemia. Semin Oncol 2003; 30(2): 206 – 10. 29. Dimopoulos M, et al. The international staging system for multiple myeloma is applicable in symptomatic Waldenstrom’s macroglobulinemia. Leuk Lymphoma 2004; 45(9): 1809 – 13. 30. Dimopoulos MA, et al. Treatment of Waldenstrom’s macroglobulinemia with rituximab: prognostic factors for response and progression. Leuk Lymphoma 2004; 45(10): 2057 – 61. 31. Garcia-Sanz R, et al. Waldenstrom macroglobulinaemia: presenting features and outcome in a series with 217 cases. Br J Haematol 2001; 115(3): 575 – 82. 32. Morel P, et al. Patients with the description of a new scoring system and its validation on 253 other patients. Blood 2000; 96(3): 852 – 8.
33. Dimopoulos MA, et al. Survival and prognostic factors after initiation of treatment in Waldenstrom’s macroglobulinemia. Ann Oncol 2003; 14(8): 1299 – 305. 34. Kyle RA, et al. Prognostic markers and criteria to initiate therapy in Waldenstrom’s macroglobulinemia: consensus panel recommendations from the Second International Workshop on Waldenstrom’s Macroglobulinemia. Semin Oncol 2003; 30(2): 116 – 20. 35. Weber D, et al. Uniform response criteria in Waldenstrom’s macroglobulinemia: consensus panel recommendations from the Second International Workshop on Waldenstrom’s Macroglobulinemia. Semin Oncol 2003; 30(2): 127 – 31. 36. Zarkovic M, Kwaan HC. Correction of hyperviscosity by apheresis. Semin Thromb Hemost 2003; 29(5): 535 – 42. 37. Drew MJ. Plasmapheresis in the dysproteinemias. Ther Apher 2002; 6(1): 45 – 52. 38. Cavanna L, et al. Advanced Waldenstrom’s macroglobulinemia: a case of possible cure after systemic chemotherapy, splenic radiation and splenectomy. Acta Haematol 2002; 108(2): 97 – 101. 39. Kyle RA, et al. Waldenstrom’s macroglobulinaemia: a prospective study comparing daily with intermittent oral chlorambucil. Br J Haematol 2000; 108(4): 737 – 42. 40. Dimopoulos MA, et al. Treatment of Waldenstrom’s macroglobulinemia with nucleoside analogues. Leuk Lymphoma 1993; 11(Suppl 2): 105 – 8. 41. Dimopoulos MA, et al. Treatment of Waldenstrom’s macroglobulinemia with the combination of fludarabine and cyclophosphamide. Leuk Lymphoma 2003; 44(6): 993 – 6. 42. Weber DM, et al. 2-Chlorodeoxyadenosine alone and in combination for previously untreated Waldenstrom’s macroglobulinemia. Semin Oncol 2003; 30(2): 243 – 7. 43. Leblond V, et al. Multicenter, randomized comparative trial of fludarabine and the combination of cyclophosphamide-doxorubicinprednisone in 92 patients with Waldenstrom macroglobulinemia in first relapse or with primary refractory disease. Blood 2001; 98(9): 2640 – 4. 44. Gertz MA, et al. Treatment recommendations in Waldenstrom’s macroglobulinemia: consensus panel recommendations from the Second International Workshop on Waldenstrom’s Macroglobulinemia. Semin Oncol 2003; 30(2): 121 – 6. 45. Dimopoulos MA, et al. Predictive factors for response to rituximab in Waldenstrom’s macroglobulinemia. Clin Lymphoma 2005; 5(4): 270 – 2. 46. Anagnostopoulos A, et al. High-dose chemotherapy followed by stem cell transplantation in patients with resistant Waldenstrom’s macroglobulinemia. Bone Marrow Transplant 2001; 27(10): 1027 – 9. 47. Tournilhac O, et al. Transplantation in Waldenstrom’s macroglobulinemia – the French experience. Semin Oncol 2003; 30(2): 291 – 6. 48. Fermand JP, Brouet JC. Heavy-chain diseases. Hematol Oncol Clin North Am 1999; 13(6): 1281 – 94. 49. Ben-Ayed F, et al. Treatment of alpha chain disease. Results of a prospective study in 21 Tunisian patients by the Tunisian-French intestinal Lymphoma Study Group. Cancer 1989; 63(7): 1251 – 6. 50. Wester SM, Banks PM, Li CY. The histopathology of gamma heavychain disease. Am J Clin Pathol 1982; 78(4): 427 – 36. 51. Kyle RA, Greipp PR, Banks PM. The diverse picture of gamma heavychain disease. Report of seven cases and review of literature. Mayo Clin Proc 1981; 56(7): 439 – 51. 52. Dispenzieri A, et al. POEMS syndrome: definitions and long-term outcome. Blood 2003; 101(7): 2496 – 506. 53. Nakanishi T, et al. The Crow-Fukase syndrome: a study of 102 cases in Japan. Neurology 1984; 34(6): 712 – 20. 54. Perniciaro C. POEMS syndrome. Semin Dermatol 1995; 14(2): 162 – 5. 55. Soubrier MJ, Dubost JJ, Sauvezie BJ, French Study Group on POEMS Syndrome. POEMS syndrome: a study of 25 cases and a review of the literature. Am J Med 1994; 97(6): 543 – 53. 56. Min JH, Hong YH, Lee KW. Electrophysiological features of patients with POEMS syndrome. Clin Neurophysiol 2005; 116(4): 965 – 8. 57. Sung JY, et al. Patterns of nerve conduction abnormalities in POEMS syndrome. Muscle Nerve 2002; 26(2): 189 – 93. 58. Papo T, et al. Human herpesvirus 8 infection, Castleman’s disease and POEMS syndrome. Br J Haematol 1999; 104(4): 932 – 3. 59. Kanitakis J, et al. Cutaneous angiomas in POEMS syndrome. An ultrastructural and immunohistochemical study. Arch Dermatol 1988; 124(5): 695 – 8.
UNCOMMON PRESENTATIONS OF PLASMA CELL DYSCRASIAS 60. Modesto-Segonds A, et al. Renal involvement in POEMS syndrome. Clin Nephrol 1995; 43(5): 342 – 5. 61. Lesprit P, et al. Pulmonary hypertension in POEMS syndrome: a new feature mediated by cytokines. Am J Respir Crit Care Med 1998; 157(3 Pt 1): 907 – 11. 62. Kuwabara S, et al. Long term melphalan-prednisolone chemotherapy for POEMS syndrome. J Neurol Neurosurg Psychiatry 1997; 63(3): 385 – 7. 63. Soubrier M, et al. Successful use of autologous bone marrow transplantation in treating a patient with POEMS syndrome. Bone Marrow Transplant 2002; 30(1): 61 – 2. 64. Hjorth M et al. The Myeloma Group of Western Sweden. Impact of active and passive exclusions on the results of a clinical trial in multiple myeloma. Br J Haematol 1992; 80(1): 55 – 61. 65. Dimopoulos MA, Kiamouris C, Moulopoulos LA. Solitary plasmacytoma of bone and extramedullary plasmacytoma. Hematol Oncol Clin North Am 1999; 13(6): 1249 – 57. 66. Soutar R, et al. Guidelines on the diagnosis and management of solitary plasmacytoma of bone and solitary extramedullary plasmacytoma. Clin Oncol (R Coll Radiol) 2004; 16(6): 405 – 13. 67. Frassica DA, et al. Solitary plasmacytoma of bone: Mayo Clinic experience. Int J Radiat Oncol Biol Phys 1989; 16(1): 43 – 8.
575
68. Holland J, et al. Plasmacytoma. Treatment results and conversion to myeloma. Cancer 1992; 69(6): 1513 – 7. 69. Galieni P, et al. Solitary plasmacytoma of bone and extramedullary plasmacytoma: two different entities? Ann Oncol 1995; 6(7): 687 – 91. 70. Knowling MA, Harwood AR, Bergsagel DE. Comparison of extramedullary plasmacytomas with solitary and multiple plasma cell tumors of bone. J Clin Oncol 1983; 1(4): 255 – 62. 71. Chak LY, et al. Solitary plasmacytoma of bone: treatment, progression, and survival. J Clin Oncol 1987; 5(11): 1811 – 5. 72. Dimopoulos MA, et al. Curability of solitary bone plasmacytoma. J Clin Oncol 1992; 10(4): 587 – 90. 73. Liebross RH, et al. Solitary bone plasmacytoma: outcome and prognostic factors following radiotherapy. Int J Radiat Oncol Biol Phys 1998; 41(5): 1063 – 7. 74. Miller FR, et al. Plasmacytomas of the head and neck. Otolaryngol Head Neck Surg 1998; 119(6): 614 – 8. 75. Galieni P, et al. Clinical outcome of extramedullary plasmacytoma. Haematologica 2000; 85(1): 47 – 51. 76. Harwood AR, Knowling MA, Bergsagel DE. Radiotherapy of extramedullary plasmacytoma of the head and neck. Clin Radiol 1981; 32(1): 31 – 6.
Section 9 : Cutaneous Malignancies
51
Unusual Cutaneous Malignancies Toni K. Choueiri, Thomas Olencki, Wolfram Samlowski, Scott Florell, Sancy Leachman, Martin Majer and Allison Vidimos
TUMORS OF THE EPIDERMIS AND EPIDERMAL APPENDAGES Hair Follicle and Hair Matrix Tumors Trichilemmal Carcinoma
This extremely rare tumor arises from the external root sheet of the hair follicle.1 It presents as a solitary ulcerating lesion on the face, although it may present as multiple tumors. It usually arises in chronically sun-exposed skin of older individuals and may be clinically misdiagnosed as a basal cell carcinoma. Histology reveals a lesion with periodic acidSchiff (PAS)-positive clear cells invading downward from the epidermis or from the external root sheath, in a multilobular pattern with a central trichilemmal keratinization.2 The potential for metastases, although uncommon, has been described.3,4 Wide surgical excision with clear margins is the current treatment of choice; however, Mohs micrographic surgery has also been used without signs of recurrence.5 Pilomatrix Carcinoma (Pilomatricarcinoma)
This follicular tumor is a rapidly growing firm nodule that usually arises within a long-standing benign pilomatricoma.6 The cancer commonly presents on the head and neck of elderly males and it has a propensity for distant metastases.6,7 Histologic examination shows a very large basaloid component with infiltration, extensive tumor necrosis, and atypical mitotic activity.8 Wide local excision is required with regular follow-up because of concerns for local and distant recurrence.9 The roles of chemotherapy and radiation therapy remain undefined.10
Sebaceous Gland Tumors Sebaceous Carcinoma
Sebaceous carcinoma (SC) or meibomian gland carcinoma represents less than 1% of all skin malignancies. It is a rare, aggressive, malignant tumor derived from the adnexal epithelium.11 The most common site of presentation is periocular, especially the meibomian glands of the tarsus or
less commonly the gland of Zeis, the lacrimal gland, or the lacrimal caruncle. The tumor is solitary and firm with a nonspecific appearance and occurs in older individuals, especially women. It is frequently mistaken for a chalazion or an inflammatory condition such as a blepharitis or keratoconjunctivitis, which delays diagnosis.12 Extraocular SC accounts for 20% of all SCs with different sites including the vulva,13 the salivary glands, and the bronchus.14 Extraocular lesions tend to be less aggressive than those found in the eyelid. SCs have been associated with prior radiation exposure to the face,15 immunosuppressive therapy after renal transplantation16 and with the Muir-Torre syndrome, an autosomal-dominant syndrome characterized by the presence of sebaceous gland tumors and visceral malignancies.17 Recently, it has been more broadly associated with the Lynch syndrome, characterized by a multitude of tumors occurring before 44 years of age.18 These include right sided colon cancer, endometrial, ovarian, breast, stomach, small bowel, pancreas, hepatobiliary, ureter, and renal pelvis cancers. Given the rarity of this cancer, the diagnosis of sebaceous gland carcinoma should prompt a consideration of these syndromes and initiate an investigation for an internal malignancy. Histologically, SC are classified as: (i) an in situ tumor; (ii) a low-grade cancer with or without pagetoid spread; (iii) an infiltrating, high-grade malignancy, with or without pagetoid spread; and (iv) a carcinoma with extraocular and extracutaneous involvement, including metastases.19 The Thomsen-Friedenreich antigen can be a helpful tool in differentiating sebaceous cancer (strong T-antigen reactivity in basaloid cells) from other mimicking neoplasms (basaloid cells are T-antigen negative).20 This cancer is characterized by frequent recurrences (40% of patients) and a tendency to metastasize (18–25% of cases).12,19 Long-term survival is found in around 30% of patients. Poor prognosis is associated with vascular and lymphatic invasion, involvement of both upper and lower eyelids, duration of symptoms greater than 6 months, tumor diameter exceeding 10 mm, a highly infiltrative pattern, and pagetoid invasion.19
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
578
CUTANEOUS MALIGNANCIES
Surgery remains the cornerstone of management. The goal of therapy should be total excision with pathologically clear margins to salvage the ocular globe and preserve vision. Mohs micrographic surgery has been reported with favorable results, but follow-up remains short.21 Radiation is not indicated in the primary management of SC but should be used when surgery is declined or when the patient is medically unfit to undergo surgery. Best responses are obtained with doses >55 Gy of radiation.22 Cryotherapy23 and topical chemotherapy with Mitomycin C24 have been used in highly select cases. Responses to chemotherapy have been documented in the metastatic setting.25
Apocrine Gland Tumors Apocrine Carcinoma
Apocrine carcinoma is a rare form of sweat gland neoplasm. Although the axilla remains the most common site for these tumors, it can develop in the anogenital region, eyelid, ear, chest, wrist, lip, foot, toe, and finger.26 There seems to be a predilection for middle-aged to older females. Classically, these slow-growing lesions present as painless, colorless or reddish, firm nodules. More than half of the reported patients with apocrine carcinoma have lymph node (LN) metastases at the time of diagnosis.26 This lesion cannot be separated from primary breast cancer on hematoxylin and eosin staining alone; therefore, it is important to rule out a primary breast cancer. Histologic examination reveals solid, ductal, and glandular tumors with decapitation secretion. Cellular and nuclear atypia with an infiltrative growth of tumor is observed. The cells are immunohistochemically positive for gross cystic disease fluid protein, lysozyme, CD15, and carcinoembryonic antigen.27 Wide surgical excision with 1 to 2 cm margins remains the standard therapy. However, lymphoscintography and sentinel lymph node (SLN) biopsy have been considered to improve survival.28 The roles of radiation and chemotherapy as adjuvant and metastatic therapy remain to be defined. Local radiation may be considered if the surgical margins are positive. A response to chemotherapy with methotrexate and bleomycin has been reported, but it should not be considered as standard therapy at this time.29
Eccrine Gland Tumors Malignant Eccrine Poroma
Malignant eccrine poroma or porocarcinoma arises from intraepidermal eccrine duct epithelium and less commonly from an underlying eccrine poroma.30 This tumor presents as a long-standing exophytic tumor, usually on the lower extremity. Metastatic spread to LNs or distant sites is possible. Histologically, porocarcinoma is an irregular tumor formed of characteristic poromatous basaloid epithelial cells, displaying ductal differentiation with significant cytologic atypia. Mature, well-formed eccrine ducts having an eosinophilic luminal cuticle may also be found in more than two-thirds of the cases, with the remaining tumors containing small, ill-formed ducts and/or intracytoplasmic lumina.31
Surgery is the mainstay of treatment and the use of Mohs micrographic surgery appears to decrease recurrence rates when compared with conventional surgical excision.32 In one report, docetaxel and 5-fluorouracil have been used to treat metastatic disease with favorable results.33 Malignant Hidradenoma
Malignant hidradenoma, also called hidradenocarcinoma or malignant acrospiroma, arises from a preexisting hidradenoma. It presents as a red, ulcerated nodule on the face or feet.34 On histologic examination, the lesions appear as large clusters of glycogen-rich cells with focal necrosis. It is most common in older adults but has been reported in children as well.35 The lesions can be aggressive and pulmonary metastases have been reported.36 Wide local excision is recommended with regular followup since recurrence rates are documented to be as high as 50%.37 Aggressive Digital Papillary Adenocarcinoma
Aggressive digital papillary adenocarcinoma is a rare variant of sweat gland carcinoma of the digits, with the potential for highly aggressive biologic behavior. Diagnosis is frequently delayed. Metastases, even from low-grade histology, have been documented.38 Both sexes, at any age, are affected. Pathology reveals a cystic lesion with papillary projections into the cystic cavities.39 Aggressive primary surgical treatment is advised.
Apocrine/Eccrine Gland Tumors These tumors may show either eccrine or apocrine differentiation, and distinction between the two is not possible unless there is a follicular differentiation. Therefore they are grouped together. Malignant Cylindroma
Malignant cylindroma is an aggressive tumor that can commonly arise from preexisting multiple cylindromas of the head and neck, particularly the scalp. The tumor can also be present in familial cylindromatosis (Brooke-Spiegler syndrome), a rare autosomal-dominant disease characterized by adnexal cylindroma tumors, trichoepitheliomas, and spiradenomas.40 Histologically, the tumors usually exhibit a transitional zone between benign cylindroma and clear malignant transformation. Mitotic activity, atypia, and tissue invasion are identified in malignant foci.41 Complete, deep surgical resection remains the treatment of choice as the tumor may invade the skull, with subsequent carcinomatous meningitis and metastases to distant viscera.41 Malignant Eccrine Spiradenoma
This tumor develops from a preexistent eccrine spiradenoma. The largest series reported 12 cases with equal sex distribution and a mean age of 62 years. Large nodular lesions, some of which had been present for months to years, were noted on the trunk and occasionally on the extremities and head and neck region.42 In all cases, continuity between benign eccrine
UNUSUAL CUTANEOUS MALIGNANCIES
spiradenoma and malignant change was observed. Malignancy was evidenced by increased mitotic rate, necrosis, and nuclear atypia. Follow-up is required, as local recurrence and distal metastases have been reported. Microcystic Adnexal Carcinoma
This indolent tumor is also called sclerosing/syringomatous sweat gland carcinoma or malignant syringoma. This cancer predominantly affects the cheek and upper lip of young and middle-aged women and rarely metastasizes. Risk factors appear to include ultraviolet (UV) light damage and previous radiation exposure to the face.43 Histology should include a deep specimen since clues to the diagnosis reside in an infiltrative growth pattern, variable ductal differentiation, and prominent perineural invasion.44,45 Aggressive treatment by Mohs micrographic surgery appears to offer the greatest likelihood of cure for this neoplasm, while providing conservation of normal tissue.44 Mucinous Carcinoma
Mucinous or adenocystic carcinoma is a slow-growing mucin-producing tumor arising as a painless nodule in the head and neck region, particularly in the eyelids.46 Men are affected more commonly than women. Histology reveals clusters of tumor cells within abundant pools of mucin, separated by fibrous septa.47 The absence of cytokeratin (CK20) may help differentiate this tumor from cutaneous metastases from gastrointestinal carcinomas.48 The clinical course is indolent, but distant metastatic disease has been rarely reported. Wide local excision or Mohs micrographic surgery is considered to be the treatment of choice.49 Eccrine Epithelioma
Also known as syringoid eccrine carcinoma, this cancer was originally considered to be a basal cell carcinoma with eccrine differentiation.50 It is an extremely rare, locally aggressive tumor found on the scalp as a large, tender, and ulcerated nodule. Less than 20 cases have been reported in the literature. Histologically, the tumor consists of numerous small cords and nests extending from the reticular dermis to the subcutaneous tissue, which forms luminal or tubular structures mimicking the nests of syringoma.51 Wide local excision with regular follow-up is required because there is chance of recurrence.52 Adenoid Cystic Carcinoma
Adenoid cystic carcinoma or primary cutaneous adenocystic carcinoma presents as a painful nodule on the scalp and neck, and frequently in the oral cavity. The pathology shows large masses of cells with cytological atypia, arranged in a distinct adenoid or cribiform pattern.53 Wide local excision is advised. Perineural invasion is common and is associated with pain and an increased recurrence rate.54 Metastases to local LNs or lungs have been recorded.55 Lymphoepithelioma-like Carcinoma
Lymphoepitheliomas are normally seen in the nasopharynx, but a subtype originates in the skin as a nodule in the head
579
and neck area of older individuals. Approximately 30 cases have been reported. These are Epstein-Barr virus (EBV) negative with less-aggressive behavior than those originating in the nasopharynx.56 Microscopic examination shows a dense lymphoplasmacytic infiltrate with large spindle-shaped cells and vesicular nuclei, prominent nucleoli, and frequent mitotic figures. Local recurrence and distant metastases are possible. Treatment consists of wide total resection and adjuvant radiotherapy (RT).57 Extramammary Paget’s Disease
Extramammary Paget’s disease was originally described in 1889 by Crocker, who reported lesions on the scrotum and penis, with histological features similar to those originally described by James Paget, in the areola.58 This disease arises anywhere on the body as a primary intraepidermal neoplasm originating from the intraepidermal apocrine gland ducts or pluripotent keratinocyte stem cells. The most common symptom is pruritis with an appearance similar to mammary Paget’s disease. Extramammary Paget’s disease is a slowgrowing tumor that can go unrecognized because of repeated excoriation or superimposed infection; frequently leading to a misdiagnosis of an inflammatory or infectious skin condition. Histopathological findings on biopsy are similar to those in mammary Paget’s disease. The cells are large with abundant, basophilic, finely granular cytoplasm and a large, central, atypical nucleus. Primary extramammary Paget’s disease has a worse prognosis in the presence of dermal invasion, and it has been suggested that the recurrence rate depends on the depth of dermal invasion.59 The most appropriate management of primary extramammary Paget’s disease at any site is wide local surgical excision.60 Mohs micrographic surgery, with a recurrence rate of 16%, may give superior results when compared with standard surgical treatment, which results in recurrences in 33 to 60% of cases.61 Imiquimod cream may be a promising therapeutic option for the treatment of limited cutaneous extramammary Paget’s disease.62 RT may be curative in select cases of in situ scalp and vulvar extramammary Paget’s disease.63
SOFT TISSUE TUMORS Fibrous and Myofibroblastic Tumors Dermatofibrosarcoma Protuberans (DFSP)
Dermatofibrosarcoma protuberans (DFSP) is a rare monoclonal cutaneous soft tissue sarcoma that accounts for approximately 1.8% of all soft tissue sarcomas. In a series of 853 patients, the trunk and the extremities were the most common sites.64 The male-to-female ratio is approximately 3 : 2 with a median age of 39 (range 12–79 years)65 Recently, 10 cases of DFSP have been reported in children. Ages ranged from 8 months to 16 years and five cases were congenital.66 DFSP arises from a chromosomal rearrangement that results in the deregulation of platelet-derived growth factor (PDGF) chain expression and leads to continuous activation of PDGF protein tyrosine kinase, which promotes DFSP cell growth.67
580
CUTANEOUS MALIGNANCIES
DFSPs arise as pink or violet-red nodules or plaques, and the surrounding skin may be telangiectatic. These lesions, typically, are fixed to the dermis and do not exhibit a nodular growth pattern except late in their course.68 Fixation to deeper structures is often observed in advanced or recurrent cases of DFSP. Histologically, DFSP is composed of monomorphic, benign-appearing spindle cells arranged in a storiform pattern, with these cells intersecting at tight, right angles around central vessels.68 Early in the course of the disease, there may be a narrow tumor-free zone (known as a grenz zone) between the lesion and the epidermis. DFSPs tend to stain positively for CD34 and negatively for factor XIIIa, whereas dermatofibromas generally stain negatively for CD34 and positively for factor XIIIa.69 Uncommon variants of DFSP include the Bednar tumor, which is characterized by the presence of melanin-containing dendritic cells, and myxoid DFSP.70 Approximately 10–15% of all DFSPs contain areas of fibrosarcoma (DFSP-FS), and these tend to exhibit more aggressive behavior, with frequent local recurrences and LN metastases.70 The optimal treatment option for patients with DFSP is Mohs micrographic surgery, which results in a recurrence rate of 2.3%.71 Wide local excision with 3-cm margins results in a 20% recurrence rate.72 – 74 The addition of adjuvant radiation therapy improved the likelihood of cure in patients with close or positive surgical margins. The 10-year local control rate after surgery and adjuvant RT is greater than 85%.75 The efficacy of chemotherapy for patients with metastatic DFSP is not well delineated; however, a limited number of clinical reports have indicated that imatinib (Gleevec), a tyrosine kinase inhibitor, may induce regression in patients with disease expressing c-KIT.76 The National Comprehensive Cancer Network (NCCN) clinical practice guidelines for DFSP summarize the current treatment recommendations.77 Giant Cell Fibroblastoma (GCF)
This locally recurrent fibroblastic tumor closely resembles DFSP and is even considered by some authorities to be the juvenile form of DFSP.78 It usually presents in male children, with rare cases in adults. The most common presentation is of an asymptomatic mass on the trunk. Pathology reveals solid fibromyxoid areas with collagen deposition, and dilated, pseudovascular spaces lined with multinucleated tumor cells.79 There is a close relationship between DFSP and giant cell fibroblastoma (GCF) clinically, histologically, and molecularly.80 Treatment is surgical. Although local recurrences are common, this tumor rarely metastasizes.81 Epithelioid Sarcoma
This tumor of young adults usually presents as a small, ulcerated, cutaneous nodule of the extremities, predominantly the upper extremities. This sarcoma is unique in that it has no normal precursor counterpart. Rarely, it may be characterized by allelic loss on chromosome 22.82 Immunohistochemistry staining is commonly positive for vimentin, keratin, epithelial membrane antigen (EMA), and CA125 and negative for S-100, actin, and CK5/CK6. It tends to grow along the aponeurosis, tendons, and fascial planes
making resection difficult. High rates of regional recurrence have been noted secondary to the predilection for LN involvement.83 SLN evaluation at presentation has been proposed to select patients who may benefit from complete LN dissection. There is optimism that this early, aggressive, surgical approach may decrease the high incidence of lung metastases seen with this tumor. Chemotherapy has been of minimal benefit for metastatic disease.
Fibrohistiocytic Tumors Malignant Fibrous Histiocytoma (MFH)
Malignant fibrous histiocytoma (MFH) encompasses pleomorphic soft tissue sarcomas presumably derived from histiocytes capable of fibroblastic transformation. Recent studies have shown that MFH is in fact composed of variable mixtures of fibroblasts, myofibroblasts, and histiofibroblasts.84 Two well-defined subtypes of MFH merit attention: myxofibrosarcoma and angiomatoid fibrous histiocytoma. Myxofibrosarcoma
This tumor is also called myxoid malignant fibrous histiocytoma. It is a neoplasm of deeper soft tissues and presents as an asymptomatic growth in middle-aged to older adults on the extremities or the trunk.85 Distinctive histologic features include a nodular growth pattern, a myxoid matrix containing elongated capillaries, and fusiform, round, or stellate tumor cells with indistinct cell margins. Cells are positive for vimentin, but rarely for actin. The myxoid change should be seen in 10% or more of the tumor before a lesion can be classified as myxofibrosarcoma.86 The grading of the tumor is particularly important since only intermediate and high-grade lesions metastasize. Myxoid tumors have a low propensity to metastasize compared to nonmyxoid tumors.87 The treatment of this tumor is wide surgical excision. Radiation therapy provides local control when margins are inadequate or to spare the patient from debilitating surgery. Doxorubicin and ifosfamide-based chemotherapy have modest activity in unresectable and/or metastatic myxoid MFH.88 The role of adjuvant chemotherapy remains undefined.87,89 Angiomatoid Fibrous Histiocytoma
This low-grade tumor presents in the limbs of children and young adults as an asymptomatic, blue or skin-colored, subcutaneous mass. Some patients present with fever, malaise, and generalized lymphadenopathy.90 On low-power examination, the tumor has hemorrhagic cystic spaces filled with red blood cells. Histology shows tumor cells arranged in sheets consisting of spindle and round cells with vesicular nuclei.79 Most cases are cured with surgical excision. Reexcision usually provides control in the 15% that recur. In one pediatric study, the use of adjuvant chemotherapy and radiation did not improve survival.91 Plexiform Fibrous Histiocytoma
This rare asymptomatic, indurated, solitary lesion occurs in the upper limbs of children and young females. Pathology reveals two components: fascicles of bland, spindle-shaped
UNUSUAL CUTANEOUS MALIGNANCIES
myofibroblastic cells and nodules of histiocyte-like cells with focal hemorrhage and hemosiderin deposition.79 The spindleshaped cells stain focally for smooth muscle actin and the nodules are focally positive for CD68.92 Mohs micrographic surgery has yielded excellent results.93 Local recurrence, LN, and pulmonary metastases have been described.94 Atypical Fibroxanthoma
Atypical fibroxanthoma occurs in sun-damaged skin of the head and neck of elderly males. Ultraviolet-induced mutations of the p53 gene, local skin trauma, and prior radiation exposure have all been implicated as possible causes.95 The lesions may appear as whitish nodules or plaques, but may be ulcerated. If the lesion does not invade the muscle or the fascia, its clinical course is relatively benign.96 Deeper, large lesions carry a worse prognosis with a 10% recurrence rate and a 1% distant metastases rate.97 A recent report concluded that the metastatic potential of atypical fibroxanthoma may be underestimated.98 Histologically, the tumor is composed of dense spindleshaped cells, histiocyte-type cells, and multinucleated giant cells. Any or all of the cells may show marked pleomorphism, atypical mitoses, and nuclear hyperchromasia. A scattered inflammatory infiltrate may be found, usually at the border of the lesion. The diagnosis can be confirmed with immunohistochemical studies that are positive for vimentin and negative for CK and S-100.99 Wide surgical excision or Mohs micrographic surgery are both adequate as initial treatment.95
Vascular Tumors Angiosarcoma
Angiosarcoma has also been referred to as hemangiosarcoma, malignant hemangioendothelioma, or lymphangiosarcoma. It is a malignant vascular tumor arising from both the vascular and lymphatic endothelia. This aggressive tumor tends to develop in three situations: as an idiopathic angiosarcoma of the head and neck in elderly individuals, as a lymphedemaassociated angiosarcoma (Stewart-Treves syndrome), and as a postirradiation angiosarcoma.100 Stewart-Treves syndrome was originally described in patients with iatrogenic lymphedema after radical mastectomy and axillary LN dissection for breast cancer.101 However, it has also been observed in cases of lymphedema not associated with mastectomy, suggesting that it is the chronic lymph stasis which in fact predisposes to the onset of angiosarcoma.102 Postirradiation angiosarcoma primarily affects women who have undergone breast-sparing surgery. Lymphedema does not seem to contribute to the pathogenesis of this subtype, which is associated with chronic radiation dermatitis.103 Skin changes usually precede the diagnosis of angiosarcoma. The tumor occurs as bluish or purple nodules with purple discoloration of the skin, or erythematous macules, frequently in combination with diffuse erythema of the breast. The most common histologic pattern is characterized by dissecting sinusoids lined by atypical endothelial cells, with
581
some cases showing a diffuse epithelioid proliferation. A mixture of the two histologic patterns is possible. The vast majority of lesions express vimentin and factor VIII-related antigen, focally. Also expressed are CD34 (74%), BNH9 (an endothelial marker, 72%), and CKs (35%). EMA is not expressed, which can be used to rule out a carcinoma. Ultrastructurally, tumor cells show intercellular and intracellular lumina with or without red cells. Tumor cells can contain occasional tonofilaments, and pinocytotic vesicles in their cytoplasm.104 Local recurrence is a major problem, with distant metastases (brain, lung) occurring in about 30% of patients. Overall 5-year survival is around 24–34%.105 Tumor of >5 cm size, depth of invasion (>3 mm), positive surgical margins, and high mitotic rate correlate with an adverse outcome. Recurrences are usually seen within 2 years of diagnosis. Treatment should include a wide surgical margin. Because of the propensity for insidious cutaneous infiltration, adjuvant RT may play an important role in treatment.106 The role of primary RT is less clear but can be attempted if the tumor is unresectable or if positive margins persist. Paclitaxel as a single agent seems to have substantial activity against angiosarcoma of the face and scalp, even in patients previously treated with other chemotherapy or radiation.107 Doxorubicin with or without ifosfamide, followed by 3 to 6 months of weekly paclitaxel has also been used. Epithelioid Angiosarcoma
Epithelioid angiosarcoma develops in the deep dermis and occasionally in the skin. It is an aggressive variant of angiosarcoma composed of endothelial cells that can mimic a carcinoma. The lesion, unlike classical angiosarcoma, is present on the extremities of young to middle-aged males.108 The typical presentation is an asymptomatic hemorrhagic papule or nodule. Rare cases associated with radiation or foreign bodies have been reported.109 Histology reveals atypical epithelioid cells with pink cytoplasm, intracytoplasmic vacuoles, vesicular nucleus, and single eosinophilic nucleoli.110 Complete surgical excision with close follow-up is advised. Epithelioid Hemangioendothelioma
Cutaneous epithelioid hemangioendothelioma is an extremely rare cutaneous malignancy. There are few cases in the literature with most occurring in association with a deep-seated soft tissue tumor, commonly a bone neoplasm. The most common presentation is a slightly raised, erythematous, and sometimes painful dermal nodule. Sex distribution seems to be equal, with an age range of 21–84 years.111 Histologically, superficial dermal nodules with nests of rounded or slightly spindled epithelioid cells with abundant eosinophilic cytoplasm are found embedded in a hyalinized or myxoid stroma.112 Usually, the lesions are circumscribed, although occasionally an infiltrating growth pattern may be apparent. Intracytoplasmic vacuolization and occasional intraluminal erythrocytes are seen. Treatment is wide excision with margin examination. Recurrence rates can be 15%, especially with deeper tumors. Aggressive adjuvant radiation and chemotherapy have been used in deep tumors with good initial results but limited follow-up.112
582
CUTANEOUS MALIGNANCIES
Retiform Hemangioendothelioma
This aggressive variant of low-grade angiosarcoma is also called hobnail hemangioendothelioma. The tumor can occur in children as well as adults and has no sex predilection.113 Human herpesvirus-8 (HHV-8) was described in one case of retiform hemangioendothelioma but it does not appear to have an etiologic role.114 Microscopically, the dermis and subcutis display a diffuse and infiltrative pattern characterized by long, arborizing blood vessels arranged in a retiform pattern lined by flattened cells with occasional hobnail appearance of endothelial cells, and with a prominent small lymphocytic infiltrate.115 The treatment of choice is surgical excision with margin examination. Recurrences are rare.113 Giant Cell Angioblastoma
This very rare pediatric tumor has been reported thrice in the literature. One was congenital and located in the hand; one appeared in the palate; and the other on the scalp of an infant. All tumors were ulcerated; the hand and palate tumors also infiltrated soft tissue and bone.116 A solid plexiform proliferation of oval-to-spindle cells, characteristic of undifferentiated mesenchymal cells, fibroblasts, and myofibroblasts is seen. Commingled with these cells are large mononuclear and multinucleate giant cells with histiocytic features.117 Treatment is complete surgical resection. In those tumors too large to resect, it appears that interferon-α may cause regression by an antiangiogenic effect.116 Kaposiform Hemangioendothelioma
Kaposiform hemangioendothelioma, also known as Kaposilike infantile hemangioendothelioma, is a locally aggressive vascular neoplasm that normally occurs in the abdominal cavity, but can affect the skin and the soft tissues.118 The presentation in the skin is a slightly raised blue-red lesion on the extremities and head and neck. More than half of the patients present with Kasabach-Merritt syndrome (KMS), a condition characterized by profound thrombocytopenia and lifethreatening hemorrhage. The tumor is not linked to HHV-8, which helps distinguish it from Kaposi sarcoma.119 Histologically, tumors consist of irregular, infiltrating nodules of compressed vessels, resembling a capillary hemangioma or Kaposi sarcoma. Lymphatic channels occur frequently and are typically seen peripheral or deep to the main tumor mass. Mortality is due to KMS and not metastatic disease, which appears limited to regional peri-nodal soft tissue.118 Complete surgical excision is the most effective treatment. Medical therapy is required when surgery is not possible or when KMS is present. In this case, treatment has proved to be challenging and may require a multimodality approach using steroids, cytotoxic agents, and interferon.120 Liposomal doxorubicin may also be considered in this setting.
Perivascular Tumors Glomangiosarcoma
Glomangiosarcomas, or malignant glomus tumors, present in older individuals as a painful mass, most frequently in
the hand.121 Histologically, it is composed of uniform, round cells, and numerous vascular components. The tumor cells are pleomorphic with large nucleoli and frequent mitotic figures. Immunohistochemical stains show strong uptake for vimentin and weak uptake for smooth muscle actin. Treatment consists of complete surgical excision.122
Muscle Tumors Superficial Leiomyosarcoma
Superficial leiomyosarcoma is a rare soft tissue neoplasm of smooth muscle derivation. It is subdivided into cutaneous and subcutaneous forms because of their different prognostic implications.123 The cutaneous form is believed to derive from the arrector pili or genital dartoic muscle, whereas the subcutaneous type is thought to arise from the smooth muscle wall of blood vessels.123 Although this tumor is most frequent in middle-aged patients, it can develop in infants and in the elderly.124 The lesions are usually solitary, but may occasionally appear as grouped nodules. Overlying skin discoloration and pain with pressure is reported in the majority of cases.125 Histologically, large, poorly circumscribed dermal proliferation composed of fascicles of pleomorphic neoplastic cells with nuclear atypia and occasional mitotic figures are seen. Positive desmin staining is also noted. The staining appears to occur with less frequency in subcutaneous and higher-grade lesions.124 The histologic differential diagnosis must take into account the possibility of a metastasis from an extracutaneous site, such as the uterus. The local recurrence rate for cutaneous leiomyosarcoma is approximately 30 to 50% and recurrent lesions tend to be larger, deeper, and with more active mitoses than the primary lesions. Metastases, mainly pulmonary, occur in 12% of the cases.126 Subcutaneous leiomyosarcoma recur locally in 70% of cases, with metastases occurring in about 30–40%. Recurrences most frequently develop in a 1- to 5-year period after the initial surgery, with positive margins being the most important risk factor for recurrence.127 Most treatment recommendations for superficial leiomyosarcomas include a surgical excision with wide margins including subcutaneous tissue and fascia. Mohs surgery has been successfully utilized with cure rates of over 85%; however, the number of treated cases is small and the follow-up is relatively short.128 There is no definitive evidence that a significant survival advantage is seen with adjuvant radiation and/or chemotherapy.129 Cutaneous Rhabdomyosarcoma
Only few cases of primary cutaneous rhabdomyosarcoma have been reported in the literature.130 The lesion is most commonly found in neonates and children and presents as a solitary skin lesion on the face.131 Histology shows diffuse dermal infiltration of poorly differentiated cells, with a focal arrangement of confluent aggregates in a vague alveolar pattern. The cells are small and round, with hyperchromatic nuclei and pale cytoplasm. Immunohistochemistry is positive for vimentin, muscle-specific actin, and myogenin, and focally positive for desmin and S-100 protein.130 The prognosis is difficult to estimate because of the rarity of
UNUSUAL CUTANEOUS MALIGNANCIES
583
cases. Treatment usually involves a combination of surgery, radiation therapy, and chemotherapy. The management of neonates with this tumor is problematic because of the immaturity of organ systems, life-threatening treatment toxicity, and the potential long-term effects of radiation/chemotherapy on normal growth.132
Fat Cell Tumors Liposarcoma
Primary liposarcoma of the skin is extremely uncommon.127 In the largest series to date involving seven patients, the tumor occurred in an age range from 39 to 95 years (median 72 years) with an equal sex distribution. The scalp and the extremities were the most common sites of involvement.133 Grossly, the tumor appears as a polypoid nodule. Histology is characterized by the presence of a relatively mature adipocytic proliferation, which showed striking variation in cell size and shape, with vacuolated lipoblasts.134 Treatment is complete surgical excision. Local recurrences can occur, but no distant metastases or disease-related deaths have been observed. Although follow-up is relatively short, it appears that cutaneous liposarcoma seems to exhibit a relatively indolent clinical behavior in this location, despite an apparent tendency to show high-grade morphologic features.133
Neuroectodermal Tumors Merkel Cell Carcinoma
Merkel cell carcinoma, a cutaneous small cell undifferentiated neuroendocrine carcinoma, is an aggressive cutaneous malignancy of neural crest origin.135,136 This uncommon tumor was initially described by Toker in 1972 as a sweat gland carcinoma. The neuroendocrine origin was identified on the basis of a finding of neurosecretory granules by electron microscopy and later by a distal deletion of 1p35-36.135 – 137 Annual age-adjusted incidence of Merkel cell carcinoma is estimated to be 0.23/100 000 for Caucasians and 0.01/100 000 for African-Americans.138 The tumor presents in middle-aged to elderly patients, but younger patients in their teens and early adulthood have been described. Men and women seem to be equally affected. Risk factors for Merkel cell carcinoma include UV radiation and immunosuppression.135,137 Recent reports from transplant patient populations and human immunodeficiency virus (HIV)-positive individuals suggest an increased incidence, earlier age of onset, and a more aggressive clinical course.139,140 The majority of Merkel cell carcinomas are located in the head and neck region, with over 20% of these cases being seen in the periorbital or eyelid area.135 Distribution appears to correlate with UV radiation exposure. Other less-frequent sites include the extremities (40%) and the trunk (10%). Clinically, the tumor often appears as a firm nodule with a red violaceous hue.135 Telangiectasia or ulceration may be seen (see Figure 1). The tumor usually measures less than 2 cm, but lesions up to 15 cm have been documented. Merkel
Figure 1 Appearance of an ulcerated 1-cm Merkel cell carcinoma on the wrist of an 83-year-old woman.
cell carcinomas are generally asymptomatic at the time of diagnosis. Histologically, Merkel cell carcinomas may resemble other small, round cell tumors including lymphoma, small cell lung carcinoma (SCLC), carcinoid primitive neuroectodermal tumors (PNET), neuroblastoma, small cell osteosarcoma, rhabdomyosarcoma, and Ewing’s sarcoma.135,136,141 On light microscopy, Merkel cell carcinoma is seen as an infiltrative dermal tumor occasionally invading underlying structures (see Figure 2). Three histological variants have been described: (i) trabecular, (ii) intermediate, and
(a)
(b)
(c)
(d)
Figure 2 Histology of Merkel cell carcinoma. An 85-year-old man presented with a papule on the scalp, with a clinical impression of basal cell carcinoma. (a) Within the dermis are infiltrative aggregates and (b) strands of neoplastic small blue cells with a high nuclear:cytoplasmic ratio, “salt and pepper” chromatin pattern, and nuclear molding. Scattered apoptotic bodies are seen. (c) Tumor cells demonstrate strong perinuclear ‘dot’ and cytoplasmic immunoreactivity with CK20 and (d) cytoplasmic immunoreactivity with CD117. The tumor cells showed no immunoreactivity with thyroid transcription factor-1 (not shown).
584
CUTANEOUS MALIGNANCIES
(iii) small cell. The small cell subtype connotes a worse prognosis and appears as sheets of diffusely infiltrating small and intermediate-sized cells. Distinction from metastatic small cell lung cancer may constitute a challenge if only hematoxylin and eosin staining are employed. On electron microscopy, Merkel cell carcinoma characteristically shows dense, core secretory granules and perinuclear intermediate filaments. Merkel cell carcinoma exhibits positive staining for CK8, CK18, CK19, and CK20 by immunohistochemistry, as well as neuroendocrine markers including neuron specific enolase, chromogranin, and synaptophysin. CK20, neurofilament protein (NFP), and thyroid transcription factor (TTF-1) immunoreactivity distinguish Merkel cell carcinoma from metastatic SCLC, which is generally CK20, NFP, and TTF-1 negative (see Figure 2). SCLC is CK7 positive while Merkel cell carcinoma is usually negative.136,142,143 Merkel cell cancer frequently expresses c-KIT CD117, in contrast to a lower incidence of expression in SCLC.144,145 Only a few samples of Merkel cell tumors have been cKIT sequenced for activating mutations, which may be rare.145 The utility of targeted therapy for Merkel cell tumor directed at c-KIT, using imatinib mesylate is currently under investigation in a Southwest Oncology Group coordinated Intergroup trial.146 Several staging systems have been proposed. The original system from Memorial Sloan-Kettering Cancer Center described stages I –III.147 Stage I was divided into IA and IB, based on diameter of the primary lesion, with a cut-off of 2 cm. Stage II described regional LN involvement, and stage III related to systemic disease including distant LN involvement. A recent study described modification of this original staging, changing IB to stage II creating a 4-tier system in the format of the American Joint Committee on Cancer (AJCC) Sixth edition 2002, Tumour-Node-Metastases (TNM) staging system for nonmelanoma skin cancer.148 Merkel cell carcinoma has a high propensity to recur locally and to have both regional and distant metastases reminiscent of the behavior of other small cell neuroendocrine cancers.141,149 Factors that have been found to predict a lower survival rate included large tumor size, histologic small cell type, and high mitotic rate.150 The overall 5-year survival rate of patients with skin primary tumors is 50–65%.135,148 Management of Merkel cell carcinoma, historically, has been multidisciplinary, including surgery and RT. There is controversy whether addition of RT to the primary site improves local control and survival, although it is widely employed. This has been brought into question following the publication of a large series of patients from Memorial SloanKettering that showed only an 8% local recurrence rate after wide excision with negative surgical margins, and a 25% nodal metastasis rate.148 Adjuvant radiation therapy was not employed to treat most patients in this retrospective study. There are few prospective trials due to the relative rarity of this condition; so treatment recommendations are based on results of retrospective analysis. The NCCN practice guidelines stress the importance of SLN biopsy to more accurately stage LNs that are not clinically involved.151,152 SLN biopsy and appropriate subsequent treatment of the
LN basin with completion LN dissection may decrease recurrences in the involved LN area, although it is not clear if this provides a survival benefit. After SLN biopsy and LN dissection, radiation treatment is frequently added with the goal of improved local and regional control. Adjuvant chemotherapy in high-risk patients (based on local tumor size and LN involvement) was evaluated by the Trans-Tasmanian Radiation Oncology Group (TROG 96 : 07).153 This study suggested efficacy of adjuvant chemoradiotherapy. The TROG investigators treated highrisk patients (primary >1 cm, gross residual disease after surgery, or LN involvement) with adjuvant RT of 50 Gy to primary and draining nodes with concomitant carboplatin AUC 4.5 and etoposide 80 mg m−2 daily for 3 days, given at weeks 1, 4, 7, and 10. It should be noted that 62% of patients enrolled in this study had LN metastases. In the 53 patients reported, there was a 3-year survival, locoregional control, and distant control in 76, 75 and 76% respectively. There were no treatment-related deaths. Prospectively randomized data are needed to determine if long-term survival is improved compared to RT alone. In contrast, patients treated at Memorial Sloan-Kettering received no benefit from adjuvant chemotherapy as it was associated with decreased survival, on univariate analysis.148 Our management strategy at the Huntsman Cancer Institute is to perform a punch biopsy or excisional biopsy of suspicious lesions. If a Merkel cell carcinoma is identified, we routinely perform an SLN biopsy and wide local excision. Completion LN dissection is performed if positive nodes are found. We have historically given adjuvant RT to the primary site, with 2–3 cm margins, as well as the involved LNs. This practice may be in question, given the recently published results of surgery alone.148 If LN involvement is detected, adjuvant chemoradiotherapy is discussed, based on the TROG 96 : 07 data.153 Patients with metastatic disease are entered in a clinical trial, if possible, or treated with SCLC regimens, such as carboplatin/etoposide or cisplatin/etoposide as first-line therapy.154 Metastatic Merkel cell carcinoma appears to have a high risk of central nervous system involvement, so any neurologic symptoms are promptly evaluated by magnetic resonance imaging (MRI) or computed tomography (CT) scan. Because of the lack of effective chemotherapy in metastatic disease, more effective agents are needed.
OTHER TUMORS Malignant Peripheral Nerve Sheath Tumor (MPNST) This tumor arises from the nerve sheath in children and young adults with neurofibromatosis type 1. Patients with this disease, which is also referred to as neurofibrosarcoma or malignant schwannoma, develop malignancy in 50% of cases.155 It grows as a nodular or polypoid lesion of the skin with the overlying epidermis demonstrating epidermal hyperpigmentation. The majority of tumors are large and located on the flexor aspects of the limbs. Few cases have been associated with RT.156 Histologically, malignant peripheral nerve sheath tumors (MPNSTs) are often hypercellular, hyperchromatic, fasciculated, and mitotically active. Low-grade tumors
UNUSUAL CUTANEOUS MALIGNANCIES
account for only about 10–15% of cases. Twenty percent of cases have unusual and potentially misleading histological features, such as epithelioid cells and divergent mesenchymal or glandular differentiation.157 Loss of NF1 tumor suppressor gene function in schwannoma cells is a key step in progression from neurofibroma to MPNST. Additional abnormalities of tumor suppressor genes, including p53 and INK4a, contribute to the malignant growth.158 Prognosis is poor with a high incidence of relapse and distant metastases, particularly in the lungs. Long-term survival is 20–30%. Aggressive surgical resection should be performed. Use of radiation postoperatively has been controversial; however, many experts agree on using adjuvant radiation therapy for all patients even if the surgical margins are negative.159 No studies have shown a consistent benefit with chemotherapy in the adjuvant setting or in the locally advanced, recurrent, or metastatic disease.160
Peripheral Primitive Neuroectodermal Tumor Also known as extraosseous Ewing’s sarcoma or peripheral neuroepithelioma, the occurrence of extraosseous disease in deep soft tissues has been well described, but primary cutaneous lesions are rare. In a recent review of cutaneous Ewing’s sarcoma, the median age at presentation was 16 years (range 7–21 years).161 The tumor presents as nonspecific solitary nodules on the trunk and extremities with a median size of 3 cm (range, 1–12 cm).161 The lesion is composed of masses of small blue cells, with perilobular fibrosis, focal hemorrhage, and ill-defined pale cytoplasm containing glycogen, and absent pericellular reticulin. An ultrastructural study shows a monotonous cell population, with focal thickening of membranes and high nuclear/cytoplasmic ratio. A reciprocal translocation t(11,22)(q24;q12) can help establish the diagnosis.162 The tumor has an indolent course and a favorable prognosis when treated with surgery followed by adjuvant radiation and chemotherapy. Initial chemotherapy can decrease the percentage of patients needing radiation therapy.163
Malignant Granular Cell Tumor (MGCT) Granular cell tumor, also called malignant granular cell myoblastoma or Abrikossoff’s tumor, is of neuroectodermal origin. In general, it appears as a singular benign lesion; however, there are rare tumors that are malignant. Most cases of malignant granular cell tumor (MGCT) develop from the mucosal (mainly the tongue), cutaneous, or subcutaneous tissue.164 The tumor appears clinically as a rapidly enlarging firm nodule, of small size, with occasional ulceration. Individuals between 30 and 60 years of age are frequently affected, with a significant female predominance. Histologically, large sheets of polygonal cells infiltrating the tissue and the subcutaneous fat form the tumor. The cytoplasm is pale, with acidophilic granules, and the nuclei are vesicular. Originally described as myoblasts, as the early specimens were isolated from the tongue, these cells are now believed to be of neuroectodermal origin.165 The prognosis of MGCT is dismal, with extensive local and distant metastases reported. Wide local excision with possible adjuvant radiation and/or chemotherapy may slow the disease progression.166
585
Clear Cell Sarcoma Primary clear cell sarcoma of the soft tissues is a very rare, distinctive malignancy having significant overlap with malignant melanoma. The tumor is usually associated with tendons and aponeuroses structures, but may regularly involve the subcutaneous tissue and dermis by direct extension.167 It usually presents as a slow-growing and frequently painful soft tissue mass, occurring primarily in young males. It has a predilection for the lower extremities, especially around the ankle and knee.168 Histology reveals broad morphologic similarities with conventional cutaneous melanoma. The lesion has a lobular growth pattern with clear cytoplasm and prominent eosinophilic nuclei. Melanin synthesis and expression of S-100 protein are usually present, as well as vimentin and occasionally minimal levels of low-molecular-weight CKs. Melanocytic differentiation and cytoplasmic premelanosomes in various stages of maturation are seen.169 A balanced translocation, t(12;22)(q13;q12), has been identified in 50–75% of clear cell sarcomas but not in melanoma.167 However a recent genomic evaluation suggests a similarity to melanoma.170 In summary, although genetically and histologically similar to melanoma, clear cell sarcoma behaves like a soft tissue sarcoma. Clear cell sarcoma is an aggressive tumor of the soft tissues with overall 5-year survival at 68%.171 These tumors present a high risk for development of distant disease and therefore warrant aggressive surgical management. Aggressive multiagent chemotherapy has minimal effect; although, a recent lone report shows excellent results with a dacarbazine-based chemotherapy in the metastatic setting.172
REFERENCES 1. Reis JP, et al. Trichilemmal carcinoma: review of 8 cases. J Cutan Pathol 1993; 20(1): 44 – 9. 2. Waibel M, et al. Tumors of the pilosebaceous unit. Skin Pharmacol 1994; 7(1 – 2): 90 – 3. 3. Knoeller SM, et al. Skeletal metastasis in trichilemmal carcinoma. Clin Orthop Relat Res 2004; 423: 213 – 6. 4. Amaral AL, Nascimento AG, Goellner JR. Proliferating pilar (trichilemmal) cyst. Report of two cases, one with carcinomatous transformation and one with distant metastases. Arch Pathol Lab Med 1984; 108(10): 808 – 10. 5. Elder D, Elenitsas R, Ragsdale BD. Tumors of the epidermal appendages. In Elder D, Elenitsas R, Jaworsky C (eds) Histopathology of the Skin. Philadelphia, Pennsylvania: Lippincott-Raven, 1997: 763. 6. Hardisson D, et al. Pilomatrix carcinoma: a clinicopathologic study of six cases and review of the literature. Am J Dermatopathol 2001; 23(5): 394 – 401. 7. Bremnes RM, et al. Pilomatrix carcinoma with multiple metastases: report of a case and review of the literature. Eur J Cancer 1999; 35(3): 433 – 7. 8. Bassarova A, et al. Pilomatrix carcinoma with lymph node metastases. J Cutan Pathol 2004; 31(4): 330 – 5. 9. Dutta R, Boadle R, Ng T. Pilomatrix carcinoma: case report and review of literature. Pathology 2001; 33(2): 248 – 51. 10. Sau P, Lupton GP, Graham JH. Pilomatrix carcinoma. Cancer 1993; 71(8): 2491 – 8. 11. Shields JA, et al. Sebaceous carcinoma of the ocular region: a review. Surv Ophthalmol 2005; 50(2): 103 – 22. 12. Shields JA, et al. Sebaceous carcinoma of the eyelids: personal experience with 60 cases. Ophthalmology 2004; 111(12): 2151 – 7. 13. Escalonilla P, et al. Sebaceous carcinoma of the vulva. Am J Dermatopathol 1999; 21(5): 468 – 72.
586
CUTANEOUS MALIGNANCIES
14. Borczuk AC, et al. Sebaceous carcinoma of the lung: histologic and immunohistochemical characterization of an unusual pulmonary neoplasm: report of a case and review of the literature. Am J Surg Pathol 2002; 26(6): 795 – 8. 15. Rumelt S, et al. Four-eyelid sebaceous cell carcinoma following irradiation. Arch Ophthalmol 1998; 116(12): 1670 – 2. 16. Harwood CA, et al. An association between sebaceous carcinoma and microsatellite instability in immunosuppressed organ transplant recipients. J Invest Dermatol 2001; 116(2): 246 – 53. 17. Ponti G, et al. Identification of Muir-Torre syndrome among patients with sebaceous tumors and keratoacanthomas: role of clinical features, microsatellite instability, and immunohistochemistry. Cancer 2005; 103(5): 1018 – 25. 18. Lynch HT, Lynch J. Lynch syndrome: genetics, natural history, genetic counselling, and prevention. J Clin Oncol 2000; 18(21 Suppl): 19S – 31S. 19. Rao NA, et al. Sebaceous carcinomas of the ocular adnexa: a clinicopathologic study of 104 cases, with five-year follow-up data. Hum Pathol 1982; 13(2): 113 – 22. 20. Hassanein AM. Sebaceous carcinoma and the T-antigen. Semin Cutan Med Surg 2004; 23(1): 62 – 72. 21. Snow SN, et al. Sebaceous carcinoma of the eyelids treated by mohs micrographic surgery: report of nine cases with review of the literature. Dermatol Surg 2002; 28(7): 623 – 31. 22. Yen MT, et al. Radiation therapy for local control of eyelid sebaceous cell carcinoma: report of two cases and review of the literature. Ophthal Plast Reconstr Surg 2000; 16(3): 211 – 5. 23. Lisman RD, Jakobiec FA, Small P. Sebaceous carcinoma of the eyelids. The role of adjunctive cryotherapy in the management of conjunctival pagetoid spread. Ophthalmology 1989; 96(7): 1021 – 6. 24. Shields CL, et al. Topical mitomycin-C for pagetoid invasion of the conjunctiva by eyelid sebaceous gland carcinoma. Ophthalmology 2002; 109(11): 2129 – 33. 25. Koyama S, et al. A case of lung metastasis from meibomian gland carcinoma of eyelid with effective chemotherapy. Gan To Kagaku Ryoho 1994; 21(16): 2809 – 12. 26. Paties C, et al. Apocrine carcinoma of the skin. A clinicopathologic, immunocytochemical, and ultrastructural study. Cancer 1993; 71(2): 375 – 81. 27. Katagiri Y, Ansai S. Two cases of cutaneous apocrine ductal carcinoma of the axilla. Case report and review of the literature. Dermatology 1999; 199(4): 332 – 7. 28. Delgado R, et al. Sentinel lymph node analysis in patients with sweat gland carcinoma. Cancer 2003; 97(9): 2279 – 84. 29. Morabito A, et al. Clinical management of a case of recurrent apocrine gland carcinoma of the scalp: efficacy of a chemotherapy schedule with methotrexate and bleomycin. Tumori 2000; 86(6): 472 – 4. 30. Robson A, et al. Eccrine porocarcinoma (malignant eccrine poroma): a clinicopathologic study of 69 cases. Am J Surg Pathol 2001; 25(6): 710 – 20. 31. Shaw M, et al. Malignant eccrine poroma: a study of twenty-seven cases. Br J Dermatol 1982; 107(6): 675 – 80. 32. Wildemore JK, Lee JB, Humphreys TR. Mohs surgery for malignant eccrine neoplasms. Dermatol Surg 2004; 30(12 Pt 2): 1574 – 9. 33. de Bree E, et al. Treatment of advanced malignant eccrine poroma with locoregional chemotherapy. Br J Dermatol 2005; 152(5): 1051 – 5. 34. Ohta M, et al. Nodular hidradenocarcinoma on the scalp of a young woman: case report and review of literature. Dermatol Surg 2004; 30(9): 1265 – 8. 35. Faulhaber D, et al. Clear cell hidradenoma in a young girl. J Am Acad Dermatol 2000; 42(4): 693 – 5. 36. Gortler I, et al. Metastatic malignant acrospiroma of the hand. Eur J Surg Oncol 2001; 27(4): 431 – 5. 37. Wilson KM, Jubert AV, Joseph JI. Sweat gland carcinoma of the hand (malignant acrospiroma). J Hand Surg [Am] 1989; 14(3): 531 – 5. 38. Duke WH, Sherrod TT, Lupton GP. Aggressive digital papillary adenocarcinoma (aggressive digital papillary adenoma and adenocarcinoma revisited). Am J Surg Pathol 2000; 24(6): 775 – 84. 39. Kao GF, Helwig EB, Graham JH. Aggressive digital papillary adenoma and adenocarcinoma. A clinicopathological study of 57
40. 41. 42.
43.
44.
45.
46. 47. 48.
49.
50.
51.
52. 53. 54.
55. 56.
57. 58. 59.
60.
61.
62.
63. 64. 65.
66.
patients, with histochemical, immunopathological, and ultrastructural observations. J Cutan Pathol 1987; 14(3): 129 – 46. De Francesco V, et al. Carcinosarcoma arising in a patient with multiple cylindromas. Am J Dermatopathol 2005; 27(1): 21 – 6. Durani BK, et al. Malignant transformation of multiple dermal cylindromas. Br J Dermatol 2001; 145(4): 653 – 6. Granter SR, et al. Malignant eccrine spiradenoma (spiradenocarcinoma): a clinicopathologic study of 12 cases. Am J Dermatopathol 2000; 22(2): 97 – 103. Abbate M, et al. Clinical course, risk factors, and treatment of microcystic adnexal carcinoma: a short series report. Dermatol Surg 2003; 29(10): 1035 – 8. Friedman PM, et al. Microcystic adnexal carcinoma: collaborative series review and update. J Am Acad Dermatol 1999; 41(2 Pt 1): 225 – 31. Nickoloff BJ, et al. Microcystic adnexal carcinoma. Immunohistologic observations suggesting dual (pilar and eccrine) differentiation. Arch Dermatol 1986; 122(3): 290 – 4. Cabell CE, et al. Primary mucinous carcinoma in a 54-year-old man. J Am Acad Dermatol 2003; 49(5): 941 – 3. Santa-Cruz DJ, et al. Primary mucinous carcinoma of the skin. Br J Dermatol 1978; 98(6): 645 – 53. Eckert F, et al. Cytokeratin expression in mucinous sweat gland carcinomas: an immunohistochemical analysis of four cases. Histopathology 1992; 21(2): 161 – 5. Marra DE, Schanbacher CF, Torres A. Mohs micrographic surgery of primary cutaneous mucinous carcinoma using immunohistochemistry for margin control. Dermatol Surg 2004; 30(5): 799 – 802. Freeman RG, Winkelmann RK. Basal cell tumor with eccrine differentiation (eccrine epithelioma). Arch Dermatol 1969; 100(2): 234 – 42. Ohnishi T, et al. Syringoid eccrine carcinoma: report of a case with immunohistochemical analysis of cytokeratin expression. Am J Dermatopathol 2002; 24(5): 409 – 13. Malmusi M, Collina G. Syringoid eccrine carcinoma: a case report. Am J Dermatopathol 1997; 19(5): 533 – 5. Urso C, et al. Adenoid cystic carcinoma of sweat glands: report of two cases. Tumori 1991; 77(3): 264 – 7. Salzman MJ, Eades E. Primary cutaneous adenoid cystic carcinoma: a case report and review of the literature. Plast Reconstr Surg 1991; 88(1): 140 – 4. Chang SE, et al. Primary adenoid cystic carcinoma of skin with lung metastasis. J Am Acad Dermatol 1999; 40(4): 640 – 2. Ferlicot S, et al. Lymphoepithelioma-like carcinoma of the skin: a report of 3 Epstein-Barr Virus (EBV)-negative additional cases. Immunohistochemical study of the stroma reaction. J Cutan Pathol 2000; 27(6): 306 – 11. Takayasu S, et al. Lymphoepithelioma-like carcinoma of the skin. J Dermatol 1996; 23(7): 472 – 5. Crocker H. Paget’s disease affecting the scrotum and penis. Trans Pathol Soc London 1888 – 1889; 40: 187 – 91. Goldblum JR, Hart WR. Vulvar Paget’s disease: a clinicopathologic and immunohistochemical study of 19 cases. Am J Surg Pathol 1997; 21(10): 1178 – 87. Lai YL, et al. Penoscrotal extramammary Paget’s disease: a review of 33 cases in a 20-year experience. Plast Reconstr Surg 2003; 112(4): 1017 – 23. Hendi A, Brodland DG, Zitelli JA. Extramammary Paget’s disease: surgical treatment with Mohs micrographic surgery. J Am Acad Dermatol 2004; 51(5): 767 – 73. Zampogna JC, et al. Treatment of primary limited cutaneous extramammary Paget’s disease with topical imiquimod monotherapy: two case reports. J Am Acad Dermatol 2002; 47(4 Suppl): S229 – 35. Moreno-Arias GA, et al. Radiotherapy for in situ extramammary Paget disease of the vulva. J Dermatolog Treat 2003; 14(2): 119 – 23. Enzinger FM, Weiss WS. Fibrohistiocytic Tumors of Intermediate Malignancy. St. Louis, Missouri: Mosby, 1988. Bowne WB, et al. Dermatofibrosarcoma protuberans: a clinicopathologic analysis of patients treated and followed at a single institution. Cancer 2000; 88(12): 2711 – 20. Thornton SL, et al. Childhood dermatofibrosarcoma protuberans: role of preoperative imaging. J Am Acad Dermatol 2005; 53(1): 76 – 83.
UNUSUAL CUTANEOUS MALIGNANCIES 67. McArthur G. Molecularly targeted treatment for dermatofibrosarcoma protuberans. Semin Oncol 2004; 31(2, Suppl 6): 30 – 6. 68. Lindner NJ, et al. Revision surgery in dermatofibrosarcoma protuberans of the trunk and extremities. Eur J Surg Oncol 1999; 25(4): 392 – 7. 69. Calikoglu E, et al. CD44 and hyaluronate in the differential diagnosis of dermatofibroma and dermatofibrosarcoma protuberans. J Cutan Pathol 2003; 30(3): 185 – 9. 70. Minter RM, Reith JD, Hochwald SN. Metastatic potential of dermatofibrosarcoma protuberans with fibrosarcomatous change. J Surg Oncol 2003; 82(3): 201 – 8. 71. Nouri K, et al. Mohs micrographic surgery for dermatofibrosarcoma protuberans: University of Miami and NYU experience. Dermatol Surg 2002; 28(11): 1060 – 4; discussion 4. 72. Roses DF, et al. Surgical treatment of dermatofibrosarcoma protuberans. Surg Gynecol Obstet 1986; 162(5): 449 – 52. 73. Rutgers EJ, et al. Dermatofibrosarcoma protuberans: treatment and prognosis. Eur J Surg Oncol 1992; 18(3): 241 – 8. 74. Mark RJ, et al. Dermatofibrosarcoma protuberans of the head and neck. A report of 16 cases. Arch Otolaryngol Head Neck Surg 1993; 119(8): 891 – 6. 75. Mendenhall WM, Zlotecki RA, Scarborough MT. Dermatofibrosarcoma protuberans. Cancer 2004; 101(11): 2503 – 8. 76. Labropoulos SV, et al. Sustained complete remission of metastatic dermatofibrosarcoma protuberans with imatinib mesylate. Anticancer Drugs 2005; 16(4): 461 – 6. 77. Miller SJ. Dermatofibrosarcoma protuberans: clinical practice guidelines in Oncology. J Natl Compr Canc Netw 2004; 2: 74 – 8. 78. Layfield LJ, Gopez EV. Fine-needle aspiration cytology of giant cell fibroblastoma: case report and review of the literature. Diagn Cytopathol 2002; 26(6): 398 – 403. 79. Billings SD, Folpe AL. Cutaneous and subcutaneous fibrohistiocytic tumors of intermediate malignancy: an update. Am J Dermatopathol 2004; 26(2): 141 – 55. 80. Sandberg AA, Bridge JA. Updates on the cytogenetics and molecular genetics of bone and soft tissue tumors. Dermatofibrosarcoma protuberans and giant cell fibroblastoma. Cancer Genet Cytogenet 2003; 140(1): 1 – 12. 81. Cook JL. Giant cell fibroblastoma: a variant of dermatofibrosarcoma protuberans treated with Mohs’ micrographic surgery. Dermatol Surg 1999; 25(6): 509 – 12. 82. Quezado MM, et al. Allelic loss on chromosome 22q in epithelioid sarcomas. Hum Pathol 1998; 29(6): 604 – 8. 83. Fong Y, et al. Lymph node metastasis from soft tissue sarcoma in adults. Analysis of data from a prospective database of 1772 sarcoma patients. Ann Surg 1993; 217(1): 72 – 7. 84. Erlandson RA, Antonescu CR. The rise and fall of malignant fibrous histiocytoma. Ultrastruct Pathol 2004; 28(5 – 6): 283 – 9. 85. Mentzel T, et al. Myxofibrosarcoma. Clinicopathologic analysis of 75 cases with emphasis on the low-grade variant. Am J Surg Pathol 1996; 20(4): 391 – 405. 86. Merck C, et al. Myxofibrosarcoma. A malignant soft tissue tumor of fibroblastic-histiocytic origin. A clinicopathologic and prognostic study of 110 cases using multivariate analysis. Acta Pathol Microbiol Immunol Scand Suppl 1983; 282: 1 – 40. 87. Zagars GK, Mullen JR, Pollack A. Malignant fibrous histiocytoma: outcome and prognostic factors following conservation surgery and radiotherapy. Int J Radiat Oncol Biol Phys 1996; 34(5): 983 – 94. 88. Patel SR, et al. Results of two consecutive trials of dose-intensive chemotherapy with doxorubicin and ifosfamide in patients with sarcomas. Am J Clin Oncol 1998; 21(3): 317 – 21. 89. Sarcoma Meta-analysis Collaboration. Adjuvant chemotherapy for localised resectable soft-tissue sarcoma of adults: meta-analysis of individual data. Lancet 1997; 350(9092): 1647 – 54. 90. Grossman LD, White RR, Arber DA. Angiomatoid fibrous histiocytoma. Ann Plast Surg 1996; 36(6): 649 – 51. 91. Corpron CA, et al. Malignant fibrous histiocytoma in children. J Pediatr Surg 1996; 31(8): 1080 – 3. 92. Mori O, Hashimoto T. Plexiform fibrohistiocytic tumor. Eur J Dermatol 2004; 14(2): 118 – 20. 93. Rahimi AD, et al. Mohs micrographic surgery of a plexiform fibrohistiocytic tumor. Dermatol Surg 2001; 27(8): 768 – 71.
587
94. Remstein ED, Arndt CA, Nascimento AG. Plexiform fibrohistiocytic tumor: clinicopathologic analysis of 22 cases. Am J Surg Pathol 1999; 23(6): 662 – 70. 95. Giuffrida TJ, Kligora CJ, Goldstein GD. Localized cutaneous metastases from an atypical fibroxanthoma. Dermatol Surg 2004; 30(12 Pt 2): 1561 – 4. 96. Fish FS. Soft tissue sarcomas in dermatology. Dermatol Surg 1996; 22(3): 268 – 73. 97. Davis JL, et al. A comparison of Mohs micrographic surgery and wide excision for the treatment of atypical fibroxanthoma. Dermatol Surg 1997; 23(2): 105 – 10. 98. Cooper JZ, et al. Metastasizing atypical fibroxanthoma (cutaneous malignant histiocytoma): report of five cases. Dermatol Surg 2005; 31(2): 221 – 5; discussion 5. 99. Maloney ME, et al. Surgical Dermatopathology. Blackwell Science, 1999. 100. Brown MD. Recognition and management of unusual cutaneous tumors. Dermatol Clin 2000; 18(3): 543 – 52. 101. Ruocco V, Schwartz RA, Ruocco E. Lymphedema: an immunologically vulnerable site for development of neoplasms. J Am Acad Dermatol 2002; 47(1): 124 – 7. 102. Komorowski AL, Wysocki WM, Mitus J. Angiosarcoma in a chronically lymphedematous leg: an unusual presentation of StewartTreves syndrome. South Med J 2003; 96(8): 807 – 8. 103. Tomasini C, Grassi M, Pippione M. Cutaneous angiosarcoma arising in an irradiated breast. Case report and review of the literature. Dermatology 2004; 209(3): 208 – 14. 104. Prieto VG, Shea CR. Selected cutaneous vascular neoplasms. A review. Dermatol Clin 1999; 17(3): 507 – 20, viii. 105. Morgan MB, et al. Cutaneous angiosarcoma: a case series with prognostic correlation. J Am Acad Dermatol 2004; 50(6): 867 – 74. 106. Mark RJ, et al. Angiosarcoma of the head and neck. The UCLA experience 1955 through 1990. Arch Otolaryngol Head Neck Surg 1993; 119(9): 973 – 8. 107. Budd GT. Management of angiosarcoma. Curr Oncol Rep 2002; 4(6): 515 – 9. 108. Farina MC, et al. Epithelioid angiosarcoma of the breast involving the skin: a highly aggressive neoplasm readily mistaken for mammary carcinoma. J Cutan Pathol 2003; 30(2): 152 – 6. 109. Enzinger FM, Weiss WS. Malignant Soft Tissue Tumors of Unknown Type. St. Louis, Missouri: CV Mosby, 1995. 110. Prescott RJ, et al. Cutaneous epithelioid angiosarcoma: a clinicopathological study of four cases. Histopathology 1994; 25(5): 421 – 9. 111. Quante M, et al. Epithelioid hemangioendothelioma presenting in the skin: a clinicopathologic study of eight cases. Am J Dermatopathol 1998; 20(6): 541 – 6. 112. Tyring S, et al. Epithelioid hemangioendothelioma of the skin and femur. J Am Acad Dermatol 1989; 20(2 Pt 2): 362 – 6. 113. Calonje E, et al. Retiform hemangioendothelioma. A distinctive form of low-grade angiosarcoma delineated in a series of 15 cases. Am J Surg Pathol 1994; 18(2): 115 – 25. 114. Gutzmer R, et al. Absence of HHV-8 DNA in hobnail hemangiomas. J Cutan Pathol 2002; 29(3): 154 – 8. 115. Fukunaga M, et al. Retiform haemangioendothelioma. Virchows Arch 1996; 428(4 – 5): 301 – 4. 116. Marler JJ, et al. Successful antiangiogenic therapy of giant cell angioblastoma with interferon alfa 2b: report of 2 cases. Pediatrics 2002; 109(2): E37. 117. Vargas SO, et al. Giant cell angioblastoma: three additional occurrences of a distinct pathologic entity. Am J Surg Pathol 2001; 25(2): 185 – 96. 118. Lyons LL. et al. Kaposiform hemangioendothelioma: a study of 33 cases emphasizing its pathologic, immunophenotypic, and biologic uniqueness from juvenile hemangioma. Am J Surg Pathol 2004; 28(5): 559 – 68. 119. Cheuk W, et al. Immunostaining for human herpesvirus 8 latent nuclear antigen-1 helps distinguish Kaposi sarcoma from its mimickers. Am J Clin Pathol 2004; 121(3): 335 – 42. 120. Hu B, et al. Kasabach-Merritt syndrome-associated kaposiform hemangioendothelioma successfully treated with cyclophosphamide, vincristine, and actinomycin D. J Pediatr Hematol Oncol 1998; 20(6): 567 – 9.
588
CUTANEOUS MALIGNANCIES
121. Kayal JD, et al. Malignant glomus tumor: a case report and review of the literature. Dermatol Surg 2001; 27(9): 837 – 40. 122. Enzinger FM, Weiss WS. Glomus Tumor, 2nd ed. Washington, District of Columbia: CV Mosby, 1988. 123. Holst VA, Junkins-Hopkins JM, Elenitsas R. Cutaneous smooth muscle neoplasms: clinical features, histologic findings, and treatment options. J Am Acad Dermatol 2002; 46(4): 477 – 90; quiz, 91-4. 124. Jensen ML, et al. Intradermal and subcutaneous leiomyosarcoma: a clinicopathological and immunohistochemical study of 41 cases. J Cutan Pathol 1996; 23(5): 458 – 63. 125. Fields JP, Helwig EB. Leiomyosarcoma of the skin and subcutaneous tissue. Cancer 1981; 47(1): 156 – 69. 126. Dahl I, Angervall L. Cutaneous and subcutaneous leiomyosarcoma. A clinicopathologic study of 47 patients. Pathol Eur 1974; 9(4): 307 – 15. 127. Dei Tos AP. Liposarcoma: new entities and evolving concepts. Ann Diagn Pathol 2000; 4(4): 252 – 66. 128. Humphreys TR, Finkelstein DH, Lee JB. Superficial leiomyosarcoma treated with Mohs micrographic surgery. Dermatol Surg 2004; 30(1): 108 – 12. 129. Bernstein SC, Roenigk RK. Leiomyosarcoma of the skin. Treatment of 34 cases. Dermatol Surg 1996; 22(7): 631 – 5. 130. Wong TY, Suster S. Primary cutaneous sarcomas showing rhabdomyoblastic differentiation. Histopathology 1995; 26(1): 25 – 32. 131. Brecher AR, et al. Congenital primary cutaneous rhabdomyosarcoma in a neonate. Pediatr Dermatol 2003; 20(4): 335 – 8. 132. Setterfield J, et al. Primary cutaneous epidermotropic alveolar rhabdomyosarcoma with t(2;13) in an elderly woman: case report and review of the literature. Am J Surg Pathol 2002; 26(7): 938 – 44. 133. Dei Tos AP, Mentzel T, Fletcher CD. Primary liposarcoma of the skin: a rare neoplasm with unusual high grade features. Am J Dermatopathol 1998; 20(4): 332 – 8. 134. Val-Bernal JF, Gonzalez-Vela MC, Cuevas J. Primary purely intradermal pleomorphic liposarcoma. J Cutan Pathol 2003; 30(8): 516 – 20. 135. Gruber S, Wilson L. Merkel cell carcinoma. In Miller SJ, Maloney M (eds) Cutaneous Oncology. Malden, Missouri: Blackwell Science, 1998: 710 – 722. 136. Reed AJ, Tumors of neural tissue. In Elder D, Elenitsas R, Jaworsky C. (eds) Histopathology of the Skin. Philadelphia, Pennsylvania: Lippincott-Raven, 1997: 1000 – 1002. 137. Vortmeyer AO, et al. Genetic changes associated with primary Merkel cell carcinoma. Am J Clin Pathol 1998; 109(5): 565 – 70. 138. Miller RW, Rabkin CS. Merkel cell carcinoma and melanoma: etiological similarities and differences. Cancer Epidemiol Biomarkers Prev 1999; 8(2): 153 – 8. 139. Penn I, First MR. Merkel’s cell carcinoma in organ recipients: report of 41 cases. Transplantation 1999; 68(11): 1717 – 21. 140. Engels EA, et al. Merkel cell carcinoma and HIV infection. Lancet 2002; 359(9305): 497 – 8. 141. Haag ML, Glass LF, Fenske NA. Merkel cell carcinoma. Diagnosis and treatment. Dermatol Surg 1995; 21(8): 669 – 83. 142. Cheuk W, et al. Immunostaining for thyroid transcription factor 1 and cytokeratin 20 aids the distinction of small cell carcinoma from Merkel cell carcinoma, but not pulmonary from extrapulmonary small cell carcinomas. Arch Pathol Lab Med 2001; 125(2): 228 – 31. 143. Hanly AJ, et al. Analysis of thyroid transcription factor-1 and cytokeratin 20 separates Merkel cell carcinoma from small cell carcinoma of lung. J Cutan Pathol 2000; 27(3): 118 – 20. 144. Feinmesser M, et al. c-kit expression in primary and metastatic Merkel cell carcinoma. Am J Dermatopathol 2004; 26(6): 458 – 62. 145. Sihto H, et al. KIT and platelet-derived growth factor receptor alpha tyrosine kinase gene mutations and KIT amplifications in human solid tumors. J Clin Oncol 2005; 23(1): 49 – 57. 146. von Mehren M. Targeted therapy with imatinib: hits and misses? J Clin Oncol 2005; 23(1): 8 – 10. 147. Yiengpruksawan A, et al. Merkel cell carcinoma. Prognosis and management. Arch Surg 1991; 126(12): 1514 – 9.
148. Allen PJ, et al. Merkel cell carcinoma: prognosis and treatment of patients from a single institution. J Clin Oncol 2005; 23(10): 2300 – 9. 149. Ratner D, et al. Merkel cell carcinoma. J Am Acad Dermatol 1993; 29(2 Pt 1): 143 – 56. 150. Skelton HG, et al. Merkel cell carcinoma: analysis of clinical, histologic, and immunohistologic features of 132 cases with relation to survival. J Am Acad Dermatol 1997; 37(5 Pt 1): 734 – 9. 151. Miller SJ, Andersen J, Beenken SW. Merkel cell carcinoma clinical practice guidelines in oncology. J Natl Compr Canc Netw 2004; 2: 80 – 7. 152. Messina JL, et al. Selective lymphadenectomy in patients with Merkel cell (cutaneous neuroendocrine) carcinoma. Ann Surg Oncol 1997; 4(5): 389 – 95. 153. Poulsen M, et al. High-risk Merkel cell carcinoma of the skin treated with synchronous carboplatin/etoposide and radiation: a Trans-Tasman Radiation Oncology Group Study – TROG 96:07. J Clin Oncol 2003; 21(23): 4371 – 6. 154. Tai PT, et al. Chemotherapy in neuroendocrine/Merkel cell carcinoma of the skin: case series and review of 204 cases. J Clin Oncol 2000; 18(12): 2493 – 9. 155. Brooks DG. The neurofibromatoses: hereditary predisposition to multiple peripheral nerve tumors. Neurosurg Clin N Am 2004; 15(2): 145 – 55. 156. Kourea HP, et al. Subdiaphragmatic and intrathoracic paraspinal malignant peripheral nerve sheath tumors: a clinicopathologic study of 25 patients and 26 tumors. Cancer 1998; 82(11): 2191 – 203. 157. Ducatman BS, et al. Malignant peripheral nerve sheath tumors. A clinicopathologic study of 120 cases. Cancer 1986; 57(10): 2006 – 21. 158. Carroll SL, Stonecypher MS. Tumor suppressor mutations and growth factor signaling in the pathogenesis of NF1-associated peripheral nerve sheath tumors: II. The role of dysregulated growth factor signaling. J Neuropathol Exp Neurol 2005; 64(1): 1 – 9. 159. Vauthey JN, Woodruff JM, Brennan MF. Extremity malignant peripheral nerve sheath tumors (neurogenic sarcomas): a 10-year experience. Ann Surg Oncol 1995; 2(2): 126 – 31. 160. Perrin RG, Guha A. Malignant peripheral nerve sheath tumors. Neurosurg Clin N Am 2004; 15(2): 203 – 16. 161. Chow E. et al. Cutaneous and subcutaneous Ewing’s sarcoma: an indolent disease. Int J Radiat Oncol Biol Phys 2000; 46(2): 433 – 8. 162. Banerjee SS, et al. Clinicopathological characteristics of peripheral primitive neuroectodermal tumour of skin and subcutaneous tissue. Histopathology 1997; 31(4): 355 – 66. 163. Shamberger RC, et al. Ewing sarcoma/primitive neuroectodermal tumor of the chest wall: impact of initial versus delayed resection on tumor margins, survival, and use of radiation therapy. Ann Surg 2003; 238(4): 563 – 7; discussion 7 – 8. 164. Kershisnik M, Batsakis JG, Mackay B. Granular cell tumors. Ann Otol Rhinol Laryngol 1994; 103(5 Pt 1): 416 – 9. 165. Becelli R, et al. Abrikossoff’s tumor. J Craniofac Surg 2001; 12(1): 78 – 81. 166. Urabe A, et al. Malignant granular cell tumor. J Dermatol 1991; 18(3): 161 – 6. 167. Langezaal SM, et al. Malignant melanoma is genetically distinct from clear cell sarcoma of tendons and aponeurosis (malignant melanoma of soft parts). Br J Cancer 2001; 84(4): 535 – 8. 168. Chung EB, Enzinger FM. Malignant melanoma of soft parts. A reassessment of clear cell sarcoma. Am J Surg Pathol 1983; 7(5): 405 – 13. 169. Lucas DR, Nascimento AG, Sim FH. Clear cell sarcoma of soft tissues. Mayo clinic experience with 35 cases. Am J Surg Pathol 1992; 16(12): 1197 – 204. 170. Segal NH, et al. Classification of clear-cell sarcoma as a subtype of melanoma by genomic profiling. J Clin Oncol 2003; 21(9): 1775 – 81. 171. Jacobs IA, et al. Clear cell sarcoma: an institutional review. Am Surg 2004; 70(4): 300 – 3. 172. Fujimoto M, et al. Complete remission of metastatic clear cell sarcoma with DAV chemotherapy. Clin Exp Dermatol 2003; 28(1): 22 – 4.
Section 9 : Cutaneous Malignancies
52
Dermatofibrosarcoma Protuberans Michael D. Alvarado, Jane L. Messina and Vernon K. Sondak
INTRODUCTION Dermatofibrosarcoma protuberans (DFSP) is an uncommon low-grade cutaneous sarcoma that typically has an indolent clinical course with the propensity for locally aggressive and infiltrative behavior. Although first described by Taylor in 1890,1 it was not until 1924 that Darier and Ferrand recorded a detailed clinical and pathological description.2 Finally, 1 year later, Hoffman gave the disease process its current name, dermatofibrosarcoma protuberans.3
BIOLOGY Dermatofibrosarcoma protuberans (DFSP) tumorigenesis is almost always attributed to rearrangements involving chromosomes 17 and 22, leading to the fusion of collagen type 1α1 and platelet-derived growth factor-β (PDGF -β) genes.4 The resulting fusion protein causes the continuous activation of the tyrosine kinase receptor PDGF-β. The PDGF-β gene is the cellular homologue of the v-sis oncogene, implicated in causing simian sarcoma.5,6 PDGF-β has been associated with several other tumors as well as DFSP, and the possibility of an autocrine loop leading to cell growth has been suggested.7,8
PATHOLOGY Microscopic examination reveals growth of spindle cells extending from the dermis into the subcutaneous tissue. The growth pattern is often referred to as “storiform” (matlike), referring to the swirling or cartwheel-like arrangement of these cells. The tumor infiltrates into and through the lobules of subcutaneous fat, giving it a “honeycomb” appearance (see Figure 1). The tumor cells, which are of uncertain histogenesis, resemble fibroblasts, and tend to be uniform with hyperchromatic, elongated nuclei, and minimal cytologic atypia.9 The tumor cells may contain melanin pigment; this uncommon pigmented variant is known as the “Bednar tumor”.10 DFSP with prominent myxoid change has also been described.
The main differential diagnosis involves distinguishing DFSP from dermatofibroma (DF). The latter is usually superficially located without significant extension into fat. DFs are characterized by a greater variation in their cellular components, frequently containing multinucleated giant cells and hemosiderin-laden macrophages. Immunohistochemical staining will identify the cells of a DFSP as being CD34- and vimentin-positive and factor XIIIa negative, while DF are positive for factor XIIIa and negative for CD34.11 A higher proliferative fraction using MIB-1 index has been found in DFSP compared with DF. Others have compared the expression of matrix metalloproteinases (MMP)-1, 2, 9 and 14 and found DF to more consistently express these factors involved in matrix remodeling, cell motility, and angiogenesis.12 Pathologists may also consider determining tumor expression of PDGF when evaluating these tumors, because therapy with imatinib has been considered. Recently, cDNA microarrays have been used to profile gene expression of these mesenchymal tumors to help with histologic diagnosis.13 Linn et al. were able to distinguish DFSP from other soft tissue tumors on the basis of their gene expression pattern. By combining morphology, immunohistochemistry, and genetic profiling, pathologists can nearly always correctly identify DFSP and allow for adequate surgical treatment. DFSP has characteristic fingerlike projections that infiltrate either laterally or deep, and can be found to extend far from the main tumor in very asymmetric fashions.14 – 16 If not completely resected, these projections lead to local recurrence, emphasizing the importance of a thorough evaluation of the peripheral and deep margins of resection specimens. Immunohistochemical staining using CD34 is often helpful in this regard. Sarcomatous transformation of DFSP is recognized in a minority of tumors, reported in some series to occur in up to 10–15% of all cases. Usually, the transformation is to a low-grade fibrosarcoma, but higher-grade and other transformants are seen occasionally. Histologically, the transformed lesion has areas with a more fascicular or herringbone pattern of growth, an increased mitotic rate, and greater nuclear
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
590
CUTANEOUS MALIGNANCIES
patient neglect and/or late physician referral for biopsy, and a history of the lesion being present for multiple years is more the rule than the exception. DFSP is found on all areas of the skin, roughly in proportion to the surface area involved. There is no apparent predilection for sun-exposed skin. Accordingly, DFSP is most commonly located on the trunk (roughly 50–60% of cases), followed by the extremities (20–30%) and head and neck (10–15%).22 Most reports state that men tend to be affected slightly more often than women, and the age at presentation is most often between the second and fifth decades of life. Although uncommon, there are cases of DFSP in infants and children. There are over 180 published childhood cases, 21 of which were described as congenital.23 – 25
TREATMENT Figure 1 DFSP showing fascicles of minimally atypical spindle cells infiltrating subcutaneous fat in a “honeycomb” pattern (100×).
pleomorphism.17,18 These features, along with increased cellularity, have been associated with an increased risk of local recurrence.18 The fibrosarcomatous areas are typically CD34negative.19 Rarely, areas of malignant fibrous histiocytoma (MFH) may be identified within a DFSP.20 There is evidence that recurrent DFSPs, particularly multiply recurrent lesions, are more prone to sarcomatous transformation, which in turn may increase the risk for the rare metastatic DFSP tumor.18,21
PRESENTATION DFSP classically presents as a slowly progressive, painless cutaneous lesion. It begins as a flat, violet or pink plaque, but usually progresses to incorporate a nodular component (see Figure 2). It is not uncommon for the lesion to ulcerate or bleed. The history is usually described as slow growing, often with long periods of time with apparent stability in size. Most often, there is a delay in diagnosis secondary to
Figure 2 DFSP of chest wall.
Wide excision with histologically negative margins is the mainstay of treatment for DFSP. Historically, high recurrence rates have been described in the literature, particularly when conservative resection is employed. The highest rate of recurrence is usually reported on the head and neck, likely due to the difficulty in achieving wide margins of resection. Adequate primary resection is important because persistent lesions tend to be more invasive at reresection with invasion into underlying fascia, muscle, and even bone, along with the possibility of sarcomatous transformation that markedly increases the risk of distant metastasis.26 An extensive review of literature published in 1996 found a mean local recurrence rate of 43% (range 26–60%) in 317 patients who underwent conservative excision.23 The same authors reported on another 489 patients with wide excision (greater than 2 cm resection margin) with an aggregate recurrence rate of 20% (range 0–60%). Mohs presented an alternative to wide excision in 1978 with use of micrographic surgery for seven patients with DFSP (6 of whom were previously treated).27 There were no local recurrences; five of the patients had >5 years of follow-up at the time of the report. Since that initial report, Mohs micrographic surgery (MMS) has been more widely utilized for DFSP. Snow et al. reported on 29 patients from their institution who underwent MMS for DFSP that had at least a 5 year follow-up.28 They also reviewed the literature for DFSP patients who underwent MMS, and had at least a 5-year follow-up. Their 29 patients had tumors located in the head and neck (45%) and the trunk and extremities (55%). Of these lesions, 21 were primary and eight were recurrent after previous non-MMS resection. There were no local recurrences or metastases during the follow-up period (mean 10.6 years, range 5–20 years). Their literature review identified 136 patients treated with MMS who had at least 5 years follow-up. Again, tumor sites were located throughout the body (25% head and neck, 62% trunk, and 13% extremities). The local recurrence rate was 6.6% (nine patients). Parker et al. used MMS in an attempt to define margins needed for complete tumor removal.29 The authors found that a margin of 2.5 cm cleared all the tumors in 20 patients
DERMATOFIBROSARCOMA PROTUBERANS
analyzed. Dividing the tumors by size revealed that tumors measuring less than 2 cm were completely cleared with a 1.5 cm margin. Ratner et al. used MMS to analyze the extent of microscopic spread in 58 patients with primary and recurrent DFSP.16 Using a concentric tumor growth model, the authors were able to estimate potential inadequate resection for various margins. For example, a 3-cm margin would have resulted in a failure rate of 15.5% in 58 patients analyzed. The authors stress that positive margins likely go undetected secondary to sampling error with standard pathological examination. Mohs surgery is a labor-intensive technique that most likely does have significant value for tumors that are situated in sites where wide excision would be difficult or particularly deforming. More importantly, MMS has helped to define the extent for wide local excision and solidify the need for more exhaustive pathological examination of the margins of resection. Recent results of standard surgical resection have shown recurrence rates that compare with MMS. Wide excision and comprehensive pathological examination of resection specimens have led to recurrence rates that approach zero. Khatri et al. reported on 24 patients (11 primary and 13 recurrent tumors) who underwent wide excision for DFSP.30 Resection margins ranged from 2.5 to 3 cm. At a median follow-up of 54 months, there were no recurrences. Stojadinovic et al. reviewed 33 patients with histologically proven DFSP of the head and neck.31 Gross surgical margin ≥2 cm (17 patients) was predictive of negative histological margin. 11 of 16 patients with less than 2 cm gross margin had positive microscopic margins. At a median follow-up of 82 months, no local recurrences occurred in the wide excision group (≥2 cm), while the less than 2 cm group had three local failures. The authors also performed an extensive literature search that revealed two other series with recurrence rates of 0%, as well as a series with a 6% recurrence rate when surgical margins were ≥2 cm. Finally, in a recent report of 62 patients with DFSP, Dubay et al. showed the value of wide excision, but also stressed the importance of comprehensive pathological analysis of surgical margins based on knowledge gained from Mohs surgery.31 Forty-two patients underwent definitive surgical therapy by wide excision alone. The authors describe a diamond-shaped surgical specimen, which allows for a more complete peripheral margin assessment. Each diamond edge is “shaved” and analyzed, along with standard horizontal sections of the deep margin. Intraoperative frozen sections were used only occasionally; reexcision was performed if the original margin was identified as positive. Using this approach, no local recurrences were detected at a median follow-up of 4.4 years. Of note, many tumors in this series – particularly smaller ones – were successfully managed with a 1-cm margin, and very few required more than 2 cm to achieve negative margins along the entire periphery of the resected area. A point of emphasis in defining margin widths is the need to accurately identify, by inspection and palpation, the edge of the tumor, which may be surprisingly subtle. Frozen sections, while they may be helpful, cannot eliminate the chance of a histologically positive final margin.
591
While primary closure is desirable whenever possible, significant undermining of adjacent tissues, skin graft, or more complicated reconstructions should be delayed until final pathology reveals negative margins. If the initial surgical defect cannot be closed primarily, a homograft or skin substitute is advocated, with definitive coverage delayed until the final pathology results are available. If any margin is found to be involved in the tumor, a wider resection of that area is carried out, and the patient is thus spared the more extensive procedure that would be required to salvage an involved margin after extensive dissection, grafting, or flap reconstruction. In particular, rotation flaps or other complex reconstructions that would alter the relationship of the surgical specimen margin to the residual tissue should not be carried out until and unless permanent pathologic analysis has confirmed a tumor-free margin.
Adjuvant Therapy Radiation in combination with surgical resection has been employed when wide surgical margins are not achievable, due to either anatomic constraints or widespread local extension of tumor. Previously, it had been thought that DFSP was resistant to radiotherapy.32 – 34 However, there have now been multiple reports of using radiation as the only therapy in isolated cases, with adequate local control.35 – 37 Based on that experience, it has been recognized that radiation could serve as an adjunct to surgical resection when close or positive margins were left behind.38 A study of 18 patients who either received radiation alone (3 patients) or radiation in combination with surgery (15 patients), revealed a 10-year actuarial local control rate of 88%. There were three local failures, each of whom underwent successful salvage procedures. The three patients who were treated with radiation alone (one primary, two recurrent) were followed up for 85–108 months with no evidence of disease recurrence. Ballo et al. reviewed 19 consecutive patients treated with radiation as an adjuvant to surgical therapy.39 Their observations revealed only one recurrence in the 19 patients, 6 of whom had positive margins. Sun et al. evaluated the treatment results of 35 cases undergoing either surgery alone or in conjunction with radiation.40 Thirtyfour patients qualified for analysis. Ten patients received postoperative radiation. Of the 24 patients who underwent surgery alone, nine had local recurrence. Only one of the 10 patients who received both radiation and surgery had recurrence of the disease. With current surgical techniques, local recurrence rates should be low after complete excision to histologically negative margins, and adjuvant radiation therapy is not indicated. If, however, a maximal effort at resecting a DFSP results in persistently positive margins, radiation therapy should be considered and is associated with good long-term control rates. In patients with sarcomatous transformation of their DFSP, particularly if there are areas of high-grade sarcoma apparent, adjuvant radiation should be considered on a routine basis.
Systemic Therapy Systemic therapy for DFSP has historically had a limited role in the treatment of the same, in view of the extremely
592
CUTANEOUS MALIGNANCIES
low risk for metastasis. However, because large lesions in “precarious” locations can be difficult to excise with adequate margins without causing significant deformity, recent investigations have examined the potential role of systemic therapy in large primary tumors, as well as in the rare case of metastatic disease. The recognition of the central role of deregulation of the platelet-derived growth factor receptor-β (PDGFR-β) and its activation by autocrine and paracrine pathways has led to investigations of receptor inhibition using imatinib mesylate (Gleevec, Novartis).41,42 Imatinib is known to be a selective inhibitor of PDGFRα, PDGFRβ, and BCR-ABL and KIT protein-tyrosine kinases. Promising preclinical research has led investigators to try treating patients with DFSP using imatinib.41,42 One of the first clinical experiences with imatinib for DFSP was in two patients with metastatic DFSP.43 The first patient responded after 4 weeks of treatment, but died soon after from aggressive pulmonary metastases. The second patient experienced a partial response following 2 months of treatment. Evidence of successful treatment was seen in the reduced volume of lung metastases, as well as the disappearance of a paratracheal mass. In another study, a patient with unresectable metastatic DFSP was treated for 4 months with 400 mg of imatinib twice daily.44 The authors used fluorodeoxyglucose-positron emission tomography (FDG-PET), magnetic resonance imaging (MRI), and histopathological and immunohistochemical analysis to document a response to therapy. Within 2 weeks of initiating therapy, the FDG-PET uptake decreased to background levels. Tumor volume reduced by 75% by the end of 4 months. This dramatic response to treatment allowed for the surgical resection of the mass. Final pathology revealed no viable tumor in the resected specimen. Labropoulos et al. treated a patient with locally recurrent and metastatic DFSP with 400 mg of imatinib once daily.45 Pretreatment physical examination demonstrated a recurrent nodule on the right upper back, as well as three lung nodules seen on computed tomography (CT) scan. Following 1 month of treatment, the patient experienced a dramatic response to therapy with disappearance of the upper back recurrent nodule. Three months after starting therapy, CT scan showed resolution of the pulmonary nodules. The largest experience using imatinib for DFSP comes from the Imatinib Target Exploration Consortium Study B2225.46 Ten patients with primary or metastatic DFSP were treated with 400 mg of imatinib twice daily. Molecular analysis was done on the tumors to identify the classic translocation, t(17;22), leading to the continuous activation of PDGFR-β. Eight of the patients had locally advanced disease and the t(17;22) translocation. The two patients with metastatic disease had complex karyotypes, lacking the typical t(17;22) translocation, and did not respond to therapy. Of the eight patients with locally advanced DFSP, 4 patients had complete clinical responses. The remaining four patients had partial responses and went on to surgery for complete excision, rendering them diseasefree.
RECOMMENDATIONS For localized DFSP, complete surgical resection with histologically confirmed negative margins should be the goal of treatment. No specific recommendation for a minimum margin of excision can be made on the basis of available data, but we begin with a margin width of 1 cm – carefully defined from the visible and palpable edge of the tumor – for smaller lesions or in tight anatomical confines. We design a diamond-shaped incision that encompasses at least the minimum measured margin, and selectively take two or four specimens for frozen section analysis from the points closest to the tumor at the outset of the procedure. Any questionable or obviously positive margins are further excised with an additional centimeter of normal-appearing tissue. The same margin taken at the skin surface is taken full-thickness down to and including the underlying muscular fascia. If the wound cannot be closed primarily or would require extensive undermining to achieve a primary closure, a staged approach is taken. The surgical wound is covered with a homograft, biological dressing, or other skin substitute, and definitive closure deferred until the final pathology has been returned: the options then being primary closure with significant undermining, split thickness skin graft, or myocutaneous flap reconstruction. When margins are found to be positive at permanent pathologic analysis, re-excision of the positive margins is performed and reconstruction is deferred again, if necessary. In the rare case where anatomical constraints do not allow for further surgical treatment despite a positive margin, radiation should be employed as adjuvant therapy. In cosmetically sensitive situations or where tissue conservation is critical, such as the face or ears, Mohs surgery may be recommended. Locally advanced cases of DFSP, or those which are identified as metastatic and cannot be treated surgically, should be considered for clinical trials involving imatinib treatment. Patients with sarcomatous transformation of the tumor should be treated in a manner consistent with the treatment of the highest-grade sarcoma found in the specimen.
REFERENCES 1. Suit H, Spiro IJ. Radiation in management of patients with Dermatofibrosarcoma protuberans. J Clin Oncol 1996; 14: 2365 – 9. 2. Darier J, Ferrand M. Dermatofibromas progressifs et recidivants ou fibrosarcomas de la peau. Ann Dermatol Syphiligr 1924; 5: 545 – 62. 3. Hoffman E. Uber das krollentreibende fibrosarkom der haut (Dermatofibrosarkoma protuberans). Dermat Zeitschr 1925; 43: 1 – 28. 4. Shimizu A, et al. Cancer Res. 1999; 110: 14 – 8. 5. Doolittle RF, et al. Simian sarcoma virus onc gene, v-sis, derived from the gene (or genes) encoding platelet-derived growth factor. Science 1983; 221: 275 – 7. 6. Waterfield MD, et al. Platelet-derived growth factor is structurally related to the putative transforming protein p28sis of simian sarcoma virus. Nature 1983; 304: 35 – 9. 7. Smits A, et al. Expression of platelet-derived growth factor and its receptors in proliferative disorders of fibroblastic origin. Am J Pathol 1992; 140: 639 – 48. 8. Kikuchi K, et al. Dermatofibrosarcoma protuberans: increased growth response to platelet-derived growth factor BB in cell culture. Biochem Biophys Res Commun 1993; 196: 409 – 15.
DERMATOFIBROSARCOMA PROTUBERANS 9. Szollosi Z, Nemes Z. Transformed Dermatofibrosarcoma protuberans: a clinicopathological study of eight cases. J Clin Pathol 2005; 58: 751 – 6. 10. Dupress WB, et al. Pigmented dermatofibrosarcoma protuberans (Bednar tumor). A pathologic, ultrastructural, and immunohistochemical study. Am J Surg Pathol 1985; 9: 630 – 9. 11. Mendenhall WM, et al. Dermatofibrosarcoma protuberans. Cancer 2004; 101: 2503 – 8. 12. Weinrach DM, et al. Metalloproteinases 1, 2, 9, and 14 in Dermatofibrosarcoma protuberans and common fibrous histiocytoma (Dermatofibroma) play a role in tumor angiogenesis. Arch Pathol Lab Med 2004; 128: 1136 – 41. 13. Linn SC, et al. Gene expression patterns and gene copy number changes in Dermatofibrosarcoma protuberans. Am J Pathol 2003; 163: 2383 – 95. 14. Taylor HB, Helwig EB. Dermatofibrosarcoma protuberans: a study of 115 cases. Cancer 1962; 15: 717 – 25. 15. Hobbs ER, Wheeland RG. Treatment of dermatofibrosarcoma protuberans with Mohs micrographic surgery. Ann Surg Oncol 1988; 207: 102 – 7. 16. Ratner D, et al. Mohs micrographic surgery for the treatment of dermatofibrosarcoma protuberans. Results of a multiinstitutional series with an analysis of the extent of microscopic spread. J Am Acad Dermatol 1997; 37: 600 – 13. 17. Connelly JH, Evans HL. Dermatofibrosarcoma protuberans. A clinicopathologic review with emphasis on fibrosarcomatous areas. Am J Surg Pathol 1992; 16: 921 – 5. 18. Bowne WB, et al. Dermatofibrosarcoma protuberans: a clinicopathologic analysis of patients treated and followed at a single institution. Cancer 2000; 88: 2711 – 20. 19. Goldblum JR. CD34 positivity in fibrosarcomas which arise in Dermatofibrosarcoma protuberans. Arch Pathol Lab Med 1995; 119: 238 – 41. 20. O’Dowd J, Laidler P. Progression of dermatofibrosarcoma protuberans to malignant fibrous histiocytoma: report of a case with implications for tumor histogenesis. Hum Pathol 1988; 19: 368 – 70. 21. Bennabeau RC Jr, et al. Dermatofibrosarcoma protuberans. Report of a case with pulmonary metastasis and multiple intrathoracic recurrences. Oncology 1974; 29: 1 – 12. 22. Gloster HM. Dermatofibrosarcoma protuberans. J Am Acad Dermatol 1996; 35: 335 – 74. 23. Checketts SR. Congenital and childhood dermatofibrosarcoma protuberans: a case report and review of the literature. J Am Acad Dermatol 2000; 42: 907 – 13. 24. Thornton SL, et al. Childhood dermatofibrosarcoma protuberans: role of preoperative imaging. J Am Acad Dermatol 2005; 53: 76 – 83. 25. Weinstein JM, et al. Congenital dermatofibrosarcoma protuberans: variability in presentation. Arch Dermatol 2003; 139: 207 – 11. 26. Rutgers EJ, et al. Dermatofibrosarcoma protuberans: treatment and prognosis. Eur J Surg Oncol 1992; 18: 241 – 8. 27. Mohs FE. Chemosurgery: Microscopically Controlled Surgery for Skin Cancer. Springfield, IL: Charles C Thomas, 1978: 249 – 255. 28. Snow SN, et al. Dermatofibrosarcoma protuberans: a report on 29 patients treated by Mohs micrographic surgery with long-term followup and review of the literature. Cancer 2004; 101: 28 – 38. 29. Parker TL, et al. Surgical margins for excision of dermatofibrosarcoma protuberans. J Am Acad Dermatol 1995; 32: 233 – 6.
593
30. Khatri VP, et al. Dermatofibrosarcoma protuberans: reappraisal of wide local excision and impact of inadequate initial treatment. Ann Surg Oncol 2003; 10: 1118 – 22. 31. Stojadinovic A, et al. Dermatofibrosarcoma protuberans of the head and neck. Ann Surg Oncol 2000; 7: 696 – 704. 32. Burkhardt BR, Soule EH. Dermatofibrosarcoma protuberans. Study of fifty-six cases. Am J Surg 1966; 111: 638 – 44. 33. Taylor HB, Helwig EB. Dermatofibrosarcoma protuberans. A study of 115 cases. Cancer 1962; 15: 717 – 25. 34. Bendix-Hansen K, Myhre-Jensen O. Dermatofibrosarcoma protuberans. A clinico-pathological study of nineteen cases and review of world literature. Scand J Plast Reconstr Surg 1983; 17: 247 – 52. 35. Pack GT, Tabah EJ. Dermatofibrosarcoma protuberans. A report of thirty-nine cases. Arch Surg 1951; 62: 391 – 411. 36. Mark RJ, et al. Dermatofibrosarcoma protuberans of the head and neck. A report of 16 cases. Arch Otolaryngol Head Neck Surg 1993; 119: 891 – 6. 37. Marks LB, et al. Dermatofibrosarcoma protuberans treated with radiation therapy. Int J Radiat Oncol Biol Phys 1989; 17: 379 – 84. 38. Suit HD, et al. Radiation in management of patients with dermatofibrosarcoma protuberans. J Clin Oncol 1996; 14: 2365 – 9. 39. Ballo MT, et al. The role of radiation therapy in the management of dermatofibrosarcoma protuberans. Int J Radiat Oncol Biol Phys 1998; 40: 823 – 7. 40. Sun LM, et al. Dermatofibrosarcoma protuberans: treatment results of 35 cases. Radiother Oncol 2000; 57: 175 – 81. 41. Shimizu A, et al. The dermatofibrosarcoma protuberans-associated collagen type I alpha1/platelet-derived growth factor (PDGF) B-chain fusion gene generates a transforming protein that is processed to functional PDGF-BB. Cancer Res 1999; 59: 3719 – 23. 42. Sjoblom T, et al. Growth inhibition of dermatofibrosarcoma protuberans tumors by the platelet-derived growth factor receptor antagonist STI571 through induction of apoptosis. Cancer Res 2001; 61: 5778 – 83. 43. Maki RG, et al. Differential sensitivity to imatinib of 2 patients with metastatic sarcoma arising from dermatofibrosarcoma protuberans. Int J Cancer 2002; 100: 623 – 6. 44. Rubin BP, et al. Molecular targeting of platelet-derived growth factor B by imatinib mesylate in a patient with metastatic dermatofibrosarcoma protuberans. J Clin Oncol 2002; 20: 3586 – 91. 45. Labropoulos S, et al. Sustained complete remission of metastatic dermatofibrosarcoma protuberans with imatinib mesylate. Anticancer Drugs 2005; 16: 461 – 6. 46. McArthur GA, et al. Molecular and clinical analysis of locally advanced dermatofibrosarcoma protuberans treated with imatinib: Imatinib Target Exploration Consortium Study B2225. J Clin Oncol 2005; 23: 866 – 73.
FURTHER READING Dubay D, et al. Low recurrence rate after surgery for dermatofibrosarcoma protuberans. A multidisciplinary approach from a single institution. Cancer 2004; 100: 1008 – 16.
Section 9 : Cutaneous Malignancies
53
Merkel Cell Carcinoma Wolfram Goessling and Robert J. Mayer
In 1875, Friedrich Sigmund Merkel (1845–1919) described a unique epidermal nondendritic, nonkeratinocyte cell, which he called a tactile cell (Tastzelle). This cell, now bearing his name, was thought to be a primary receptor for tactile stimuli.1 Merkel cells are now generally believed to be primary neural cells, found as single cells within the basal layer of the epidermis or grouped together as a component of the “tactile hair disc of Pinkus” in the hair-bearing skin of mammals, functioning as slowly adapting type I mechanoreceptors.2,3 Their origin from the neural crest is supported by recent interspecies embryonic transplantation models in birds,4 but other authors invoke an epidermal origin from keratinocytes.5,6 In 1972, Toker described five cases of a trabecular cell carcinoma of the skin, which was initially thought to be derived from sweat glands.7 However, in 1978 Tang and Toker found dense-core granules typical of Merkel cells and other neuroendocrine cells on electron microscopy in these trabecular tumors,8 suggesting an origin from the Merkel cell. Also called (neuro)endocrine cancer of the skin or small cell carcinoma of the skin, the name Merkel cell carcinoma (MCC) was suggested by De Wolf-Peeters in 1980.9 More than 2000 cases have thus far been reported in the literature.
BIOLOGICAL FEATURES Epidemiology MCC is a rare cutaneous neoplasm of the elderly population, with an estimated 470 new cases in the United States each year; this incidence figure compares to 31 000 new cases of melanoma annually.10 Miller and Rabkin calculated the annual incidence on the basis of Surveillance Epidemiology and End Results (SEER) data as 0.23 per 100 000 for whites,11 which is similar to an estimate using the regional patient population served by the Mayo Clinic.12 However, a recent analysis of the SEER data from 1986 to 2001 revealed a significant increase in incidence over time, tripling from 0.15 cases per 100 000 in 1986 to 0.44 cases per 100 000 in 2001.13 Whether this apparent increase is due to a true rise in the number of new cases or the development of more precise diagnostic techniques remains uncertain.
The incidence among blacks appears lower (0.01 cases per 100 000), only a few cases having been reported in this population group.14 – 16 While most studies show a male predominance with SEER data suggesting a ratio of 2.3 : 1,11 other case series demonstrate a slightly higher incidence in women.17,18 MCC is a disease of the elderly, with the patients averaging 69 years at the time of diagnosis;13,17 – 21 only 5% of all reported patients have been diagnosed below the age of 50 years,11 with a reported age range of 7 to 104 years.22,23
Clinical Presentation At the time of diagnosis, patients typically present with a flesh-colored, red or blue, firm, nontender intracutaneous mass that has grown rapidly over a few weeks to months and may ulcerate.24,25 Most commonly, the tumor is nodular but may also have a plaquelike appearance (see Figure 1). Tumor size ranges from 2 to 200 mm, but is most commonly less than 20 mm. It occurs primarily in the sun-exposed areas of the skin, with approximately 50% of all tumors arising in the face and neck; 40% appear on the extremities and 10% on the trunk and genitals.17,26 – 32 In the face, the eyelids are involved frequently.33 – 35 Rare occurrences in sun-protected areas, such as oral mucosa,36 vulva,37 and penis38 have been described. The tumor spreads frequently; common secondary sites include the skin (28%), lymph nodes (27%) liver (13%), lung (10%), bones (10%), and brain (6%).16 Symptoms reflect local infiltration or involvement of lymph nodes or distant sites due to tumor growth. Superior vena cava syndrome secondary to obstruction by tumor mass39 as well as paraneoplastic neurologic complications40,41 with the presence of anti-Hu antibodies42 have been reported.
Staging and Prognosis Most patients with MCC present with localized disease (70–80%); 10–30% of patients have regional lymph node involvement, and 1–4% have distant metastases at the time of initial presentation.17,24,25,28 There is no accepted staging system for MCC, but several investigators have adopted a simple system proposed by Yiengpruksawan:24 Stage I: localized skin disease (stage IA ≤ 2 cm, IB >2 cm); stage II: regional lymph node disease; stage III: metastatic disease.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
MERKEL CELL CARCINOMA
Figure 1 Merkel cell tumor – this 2.0 cm pigmented tumor on the thigh was clinically misdiagnosed as a melanocytic lesion (Reproduced from Pathology of the Skin, 2nd ed, Phillip H. McKee, 1996, with permission from Elsevier).
A more recent staging system used at Memorial SloanKettering Cancer Center (MSKCC)43 has greater consistency with the four-stage methodology employed by the American Joint Committee on Cancer. In the MSKCC system, the prior stage IB disease is termed stage II, thereby designating stages I and II for low and high-risk local disease, with stages III and IV now representing metastatic disease to lymph nodes (stage III) and other organs (stage IV). Purported favorable prognostic factors include primary tumor location in the head and neck region without involvement of regional lymph nodes,18,24 tumors <2 cm in size (stage I),44 and female gender.18,25,28,45 Initially, MCC was thought to be a cancer with a favorable prognosis, as only one out of the first five patients described by Toker succumbed to the disease,7 and only three deaths occurred among the initial 24 reported cases.29 However, MCC has subsequently been shown to be a highly aggressive and lethal tumor, comparable to small cell lung cancer (SCLC) and melanoma in its behavior in regard to recurrence, metastatic spread, and mortality. The overall recurrence rate ranges from 55 to 79%, relapsing most often locally or in regional lymph nodes,17,20,24,25,43 – 46 with the majority of recurrences developing within the first 6–12 months after the initial diagnosis.
Pathology MCCs arise in the dermis and frequently extend into the subcutaneous fat. The epidermis is often intact but may
595
be ulcerated, and in a minority of cases there is epidermal involvement either as neuroendocrine carcinoma in situ or as single cells randomly distributed throughout the epithelium.27,47 Squamous cell carcinoma in situ may additionally be present in the overlying epidermis.48 The tumor is composed of “small blue cells” with hyperchromatic nuclei and minimal cytoplasm. Mitoses are frequently abundant, and apoptosis is often widespread. Lymphovascular invasion is an almost invariable feature.27 Three main histologic subtypes are recognized, termed by one group of pathologists as intermediate, small cell, and trabecular (the least common),49 and as solid, diffuse, and trabecular by another50 There appears to be no clinical significance to these subtypes, and in the majority of cases, an admixture is present.51 The intermediate (or “solid”) variant is encountered in approximately two-thirds of all cases. It consists of nodules and diffuse sheets of basophilic cells with imperceptible cytoplasm and round or oval vesicular nuclei, the dispersed chromatin giving a pathognomonic “watery” appearance (see Figure 2a). The small cell (or diffuse) variant is histologically identical to other small cell carcinomas and consists of irregular, hyperchromatic cells, often showing crush artifact and frequently displaying nuclear molding (see
(a)
(b)
(c)
(d)
Figure 2 (a) Intermediate variant of MCC showing vesicular, basophilic nuclei with prominent nucleoli and multiple mitoses. (b) Small cell variant of MCC. The histological features are indistinguishable from bronchial small cell carcinoma. (c) Trabecular variant of MCC. This pattern is rare and normally only seen as a small component of a mixed variant. (d) Merkel cell carcinoma. Cytochemical stain demonstrating cytokeratin expression (CK 20) with conspicuous paranuclear dots (Reproduced from Pathology of the Skin, 2nd ed, Phillip H. McKee, 1996, with permission from Elsevier).
Table 1 Immunocytochemical differential diagnosis of Merkel cell tumor.
Tumor Merkel cell tumor Small cell carcinoma of lung Lymphoma Peripheral primitive neuroectodermal tumor Small cell melanoma
CK20
CK7
NSE
NFP
S-100
LCA
CD99
TTF-1
+ −
− +
+ +
+ +/−
− −
− −
+ (cytoplasmic) in up to 50% Rarely + (cytoplasmic)
− +
− −
− −
− +
− Rarely +
− −
+ −
− + (membranous)
− −
−
−
+
−
+
−
−
−
CK 20, cytokeratin 20; CK 7, cytokeratin 7; NSE, neuron-specific enolase; NFP, neurofilament protein; LCA, leukocyte common antigen; TTF-1, thyroid transcription factor 1.
596
CUTANEOUS MALIGNANCIES
Figure 2b). This variant, in particular, must be distinguished from metastatic small cell carcinoma of the lung. The trabecular variant consists of delicate ribbons of small basophilic cells typically displaying nuclear molding (see Figure 2c). Spindle cell forms are also encountered, and occasionally keratinization and ductal differentiation are seen. Diagnosing MCC and distinguishing it from metastatic small cell carcinoma is achieved through immunohistochemical staining (see Table 1). The tumor cells express CAM 5.2 and more specifically cytokeratin 20 (CK 20), often as a paranuclear dot (see Figure 2d).50,52 – 59 CK 7, which characterizes bronchial small cell carcinoma, is typically negative. Similarly, thyroid transcription factor (TTF-1), a homeodomain-containing nuclear transcription factor that is present in small cell carcinoma of bronchial derivation is not found in MCC.60 – 63 In contrast, neurofilament protein, again presenting as a paranuclear dot, is commonly identified in MCC but is usually absent in bronchial small cell carcinoma.59,64 – 66 MCC frequently reacts to neuron-specific enolase, synaptophysin, epithelial membrane antigen, Fli-1 and chromogranin; BER-EP4 may also be expressed, and about half of the tumors express CD99.50,67,68 S-100 protein and leukocyte common antigen invariably are absent from MCC, thereby distinguishing such tissue from small cell melanoma and cutaneous lymphomatous deposits. The reactivity with Fli-1 and CD99 can make distinction from Ewing’s sarcoma challenging. Electron microscopy discloses electron dense granules (80–200 nm) and paranuclear globular aggregates of keratin and neurofilament protein. The histologic differential diagnosis of MCC includes peripheral primitive neuroectodermal tumor in addition to metastatic small cell carcinoma, small cell melanoma, and cutaneous lymphoma.
Etiology Little is known about specific etiologic factors in the pathogenesis of MCC. However, the condition has been linked, similar to melanoma, to increased sun exposure, both in its anatomical and geographical distribution. Miller and Rabkin describe a correlation between the solar ultraviolet B (UVB) index and regional differences in the incidence of MCC.11 One study demonstrates a typical UVB-induced p53 mutation in a case of MCC.69 A recent report describes a 100-fold increased incidence of MCC in patients treated with methoxsalen and UVA for psoriasis,70 and infrared light damage has also been described as a risk factor.71 In addition, several reports describe the appearance of MCC in association with synchronous or metachronous squamous cell cancer48,72 – 76 or basal cell cancer of the skin,72 either indicating sun exposure as a risk factor for developing MCC or a common precursor cell for both. Arsenic exposure has also been implicated in the pathogenesis.77 Another possible cause of MCC is an impaired immune status, either from iatrogenic immunosuppression, human immunodeficiency virus (HIV) infection or neoplasia. MCC has been reported to occur after organ transplantation, with 41 cases published from the Cincinnati Tumor Registry.78 The incidence of MCC in this patient population was 0.9% of
all de novo malignancies post-transplantation, with a younger mean age at onset (46 years) than that seen in the general population. Other studies report MCC after solid organ transplant and in patients receiving immunosuppressive therapy for rheumatoid arthritis.11,79 – 86 A recent report describes the occurrence of MCC 7 years after bone marrow transplant for non-Hodgkin’s lymphoma.87 Several cases of MCC arising in HIV-infected patients have been reported,88 – 95 and another association has been noted between MCC and chronic lymphocytic leukemia.96 – 100 These reports suggest that immunosuppression from various etiologies may play a role in the pathogenesis of MCC, especially in younger patients. An immunologic etiology has also been proposed to explain the increased frequency of secondary malignancies in patients with MCC (25%), as compared to those with melanoma (5.8%).76
Molecular Aspects A number of chromosomal abnormalities have been described in MCC, the most intriguing possibly relating the pathogenesis of the disease to a deletion on the short arm of chromosome 1 (1p36).101 – 104 This chromosomal region has also been implicated in the pathogenesis of neuroblastoma105 and melanoma,106 suggesting the presence of a tumor suppressor gene. The p73 gene has been localized to 1p36.33,107 but a recent analysis found mutations of the p73 gene in only one out of 10 patients with MCC,69 similar to the infrequent mutations of p73 in melanoma.108 The tumor suppressor gene p53 is thought to be transiently expressed in MCC; a mutant p53 protein has been identified in six of nine patients who had poor clinical outcome as compared with none of 10 patients with a more favorable outcome.109 Another important abnormality may be the reported loss of heterozygosity in chromosome 3p21,110 the same region that is also affected in more than 90% of patients with SCLC and for which a candidate tumor suppressor gene, Ras association domain family 1 (RASSF1A) gene, has recently been postulated.111,112 This may indicate a common oncogenic pathway for MCC and SCLC, both of which exhibit neuroendocrine origin and share the histologic features of “small blue cells” and the tendency to metastasize early. Several other chromosomal abnormalities have been described in MCC: trisomy 1,113,114 trisomy 6,114 – 116 trisomy 11,113,116 and trisomy 18117 as well as deletion of chromosome 7.116 One study reported two to three copies of chromosome X in 71% of tumor cells, compared to only one copy in almost all normal Merkel cells.117 Loss of heterozygosity has also been reported in chromosome 10q,118 and chromosome 13 where one report shows a frequent deletion of the retinoblastoma gene RB1.113,119,120 While some of these findings suggest the involvement of tumor suppressor genes or oncogenes in the pathogenesis of MCC, no conclusive candidates have been identified, and little is known about the prognostic value of the reported abnormalities. Most recently, several types of cancers have been characterized by the gene expression profile on microarray analysis.
MERKEL CELL CARCINOMA
This has also been accomplished in MCC, revealing a distinct expression pattern that may aide the diagnosis and its differentiation from SCLC.121
597
localized disease at the time of presentation will develop local recurrence or lymph node involvement following such an excision. Therefore, 2–3-cm wide and 2-cm deep margins have been generally recommended.24,25,28,140 Yiengpruksawan et al. detected local recurrence in 4 out of 27 patients with margins >3 cm, but in none of 11 patients with margins >3 cm,24 and O’Connor et al. reported a similar reduction in local recurrence when margins of 3 cm were compared to 2 cm.141 However, more recently, Ott et al. reported no difference in survival in 33 patients when resection margins were greater or less than 2 cm,18 and Gillenwater et al. demonstrated no difference in outcome in 18 patients based on margins <1 cm, 1–2 cm, or >2 cm.142 Allen et al. stressed the importance of excision with the margins free of disease, even though in their sample population, it was not a statistically significant predictor of disease-free survival.43 Mohs micrographic surgery143 has been proposed as being more successful in controlling local disease than traditional wide excision, especially in such cosmetically sensitive anatomic areas as the face. Uncontrolled clinical experience has provided promising results in terms of local control,27,144,145 but definitive clinical studies have not yet been conducted, and the impact of this technology on survival remains unknown.
Diagnostic Evaluation After histologic diagnosis, patients should undergo further imaging studies for staging to exclude other sites as the primary source of a small cell carcinoma. Computed tomography (CT) of the chest should be performed to exclude the presence of a lung mass suspicious for SCLC as well as evidence of metastatic disease. Abdominal and pelvic CT scans are also useful to assess for metastases.122 Octreotide scans have been used to detect various neuroendocrine neoplasms, especially in the gastrointestinal tract, since the early 1990s,123 and several case reports evaluating this technique in the diagnosis of primary MCC and metastatic disease have reported a greater sensitivity than CT.124 – 130 Other authors, however, have been more cautious about its utility.131 The presence of somatostatin receptors as documented in these scans has also been utilized therapeutically in MCC and other neuroendocrine tumors.124,129,132,133 Positron emission tomography (PET) using fluorodeoxyglucose has become an important diagnostic tool in many cancers. It has been examined in the staging of melanoma, especially in the detection of metastatic disease.134 – 136 Several reports describe the usefulness of PET in the initial staging and follow-up after chemotherapy in patients with MCC.137 – 139
Sentinel Lymph Node Biopsy
TREATMENT
Pathologic involvement of regional lymph nodes is present in 10–30% of all patients presenting with MCC who undergo lymph node dissection.24,25,43 The detection of lymph node involvement is another important prognostic factor. Consequently, elective lymph node dissection has been recommended for younger patients with large lesions or tumors arising in the head and neck region.24,28,146 Elective regional lymphadenectomy or sentinel node biopsy was first introduced by Cabanas for penile carcinoma in 1977,147 but has recently been applied predominantly to patients with
Because of the rare occurrence of MCC, no prospective clinical studies and no multicenter studies assessing initial surgical therapy or subsequent radiation therapy and chemotherapy have been performed. The general treatment approach to MCC according to stage is summarized in Table 2.
Surgical Treatment Surgical excision with tumor-free margins is the primary therapy for stage I disease. Up to two-thirds of patients with
Table 2 Proposed staging system for Merkel cell carcinoma, relationship to overall survival (data from reference 24) and recommended treatment.
Median survival (months)
Stage I IA IB
Localized disease 2 cm >2 cm
5-year survival (%) 64
30 26
II
Lymph node involvement
18
47
III
Distant metastases
5
0
CAV, cyclophosphamide, doxorubicin, and vincristine; EP, etoposide plus cisplatin.
Treatment recommendations Surgery: Local excision with >2 cm margin, sentinel lymph node biopsy Radiation therapy: Adjuvant treatment after resection with 45 – 50 Gy Chemotherapy: Little experience for adjuvant chemotherapy Surgery: Local excision with >2 cm margin, lymph node dissection Radiation therapy: Adjuvant therapy to both primary site and lymph node region Chemotherapy: One multicenter trial for adjuvant radiochemotherapy Radiation therapy: Palliative use of radiation Chemotherapy: CAV or EP most commonly used
598
CUTANEOUS MALIGNANCIES
melanoma148,149 and breast cancer,150,151 as sentinel nodes have been shown to reflect the histology of the remaining lymph nodes in the specific lymph node basin in both cancers. Several recent reports have shown the emerging role of lymphoscintigraphy and sentinel lymph node biopsy in MCC.152 – 159 The published experience so far is limited to a few dozen patients reported with relatively brief median follow-up times. Currently, little can be concluded about the therapeutic value of lymph node dissection in preventing regional recurrence. However, the assessment for lymph node involvement is recommended, sparing patients without sentinel lymph node involvement the morbidity of full lymph node dissection.43
Radiation Therapy Because of the aggressive nature of the disease and the high local and regional failure rates after surgery alone, radiation therapy has been used in the adjuvant setting as well as after resection for local recurrence. MCC cell lines have been shown to be radiosensitive in vitro,160 and the similarities to small cell lung carcinoma on light microscopy led to the use of radiation therapy early on.161 Cotlar et al. in 1986 reported their own experience of 10 patients and reviewed another 139 reported in the literature and strongly argued for radiation therapy after initial surgery.162 Subsequently, several studies from Australia,163 – 165 the United Kingdom,161 Israel,166 France,167 Germany,168 – 170 and the United States18,142,171 – 174 have argued for the benefits of both adjuvant radiation treatment after initial surgery with curative intent as well as after resection for recurrent MCC and palliation. Optimal loco–regional control was achieved with resection followed by radiation, and a higher rate of recurrence was observed in those patients treated with initial surgery alone. However, these nonrandomized retrospective analyses are very heterogeneous with regard to tumor size and location as well as radiation techniques, and only two studies18,164 suggest a survival advantage with adjuvant radiation. The total radiation doses varied from 30 to 70 Gy, in 20–25 daily fractions, with 45–50 Gy total being most commonly used. Ott et al. reported a radiation dose of >45 Gy as having a significant impact on loco–regional control and prolonged survival in nine patients, whereas a subset of seven patients who received <45 Gy had a poorer outcome.18 Because of the limited number of patients with MCC, extensive data from prospective studies are not available, and other reports suggest no significant benefit from adjuvant radiation.24,31 A recent multicenter trial from the TransTasman Radiation Oncology Group in Australia studying the concurrent administration of carboplatin and etoposide chemotherapy with radiation therapy at 50 Gy revealed high rates of loco–regional control and survival.175,176 In addition to radiation therapy, concomitant hyperthermia has been shown to successfully treat a few patients with MCC.177,178 Given the current experience, routine adjuvant nodal radiation therapy at doses of 45–50 Gy is recommended by the National Comprehensive Cancer Network for those patients who had no sentinel lymph node biopsy or who have lymph node involvement, as this may prevent local recurrence.179
Table 3 Chemotherapy regimen commonly used for the treatment of MCC.
Regimen Cyclophosphamide, doxorubicin, vincristine Etoposide, cisplatin Cyclophosphamide, epirubicin, vincristine Cyclophosphamide, doxorubicin, vincristine alternating with etoposide, cisplatin Cyclophosphamide, doxorubicin, vincristine + prednisone Doxorubicin, cisplatin ± bleomycin Doxorubicin Doxorubicin/ifosfamide Cisplatin ± oxorubicin Mitoxantrone
Chemotherapy Chemotherapy is the least studied therapeutic component of therapy for MCC. While MCC was initially considered resistant to chemotherapy, subsequent reports have shown encouraging results with chemotherapeutic approaches similar to the therapeutic regimens utilized for SCLC and neuroendocrine tumors in other locations.180,181 Over the last 15 years, many different cytotoxic combinations have been used both in the adjuvant setting as well as for recurrent or metastatic disease and as primary therapy in patients with inoperable tumors.27,166,182 – 188 The different regimens commonly used in this disease are listed in Table 3. However, because of the low incidence of MCC, no prospective randomized trials have been conducted. While chemotherapy in recurrent or metastatic MCC is frequently utilized, its value as adjuvant therapy remains to be further determined. Another problem is the advanced age of most patients with MCC who may be intolerant to the high doses of chemotherapy some of these regimens require. Most cases of toxicities have been related to bone marrow suppression and neutropenia.16,189 Tumor lysis syndrome with renal failure has been described after chemotherapy with doxorubicin and ifosfamide in a patient with extensive metastatic MCC.190 Voog et al. recently reviewed the use of chemotherapy in locally advanced or metastatic MCC in 107 patients from 37 reports.16 Among the chemotherapeutic agents, cyclophosphamide (56%), anthracyclines (49%), and cisplatin (25%) have been the most commonly used, usually as part of polychemotherapy. The overall objective response rate has been 60%, slightly higher in the setting of locally advanced disease (69%) than in metastatic disease (57%). Regimens combining doxorubicin and cisplatin as well as those containing 5-fluorouracil had significantly higher response rates when compared with other therapies, and those patients who responded with complete remission (CR) to initial chemotherapy had a significantly increased rate of survival after 1 and 5 years compared to those with only partial response (PR) or no responses. However, toxic deaths related to chemotherapy occurred in nine of 107 patients (8.4%), especially in those patients >65 years of age. Tai et al. summarized the outcome of 204 patients with MCC who were treated with chemotherapy, including those managed in the adjuvant as well as metastatic disease settings.189 In this review, 68% of patients without distant metastases responded
MERKEL CELL CARCINOMA
to the various treatments with CR, PR, or minor responses. The overall response rate for patients with distant metastases was 59%. The combinations of cyclophosphamide, doxorubicin, and vincristine (CAV), or cyclophosphamide in combination with epirubicin and vincristine ± prednisone and etoposide plus cisplatin (EP), were the most commonly used drug regimens with an overall combined response rate of 69%. Taxanes, although used in the treatment of SCLC, have not yet been examined in the treatment of MCC. So far, two cases of MCC treated with high-dose chemotherapy followed by either autologous bone marrow191 or stem cell192 transplantation have been reported; the first achieved a partial remission, while the second patient was in CR for 6 months prior to recurrence in the lung. Currently, chemotherapy has no established role in the adjuvant treatment of MCC. For the treatment of metastatic disease, however, CAV and EP are the most commonly used regimens.
Other Therapeutic Agents Biologic agents, especially interferon193 – 196 and tumor necrosis factor,137,193,197,198 have occasionally been used in the therapy for MCC. Two groups from Japan have reported complete regression of local tumors without recurrence as a result of direct intratumoral injection of tumor necrosis factor.197,198 On the basis of the observation that MCC cell lines require several growth stimuli to survive in culture199 and one of them may be nerve growth factor, farnesylthiosalicylic acid, an inhibitor of ras signal transduction, has been studied and shown to inhibit the growth of human MCC in SCID mice.200 On the basis of the report that Bcl-2 is frequently overexpressed in MCC cells compared to normal Merkel cells,199,201,202 Bcl-2 antisense oligonucleotides have been used in SCID mice with MCC; inhibition of human MCC tumor growth has been observed when compared with controls or cisplatin-treated animals.203 These preliminary studies have led to a multicenter phase II trial using the Bcl-2 antisense nucleotide oblimersen in patients with metastatic or recurrent MCC that is currently enrolling patients.204 These recent developments give hope for new treatment approaches in the future, based on identifiable molecular targets.
SUMMARY MCC is a rare skin cancer, probably of neuroendocrine origin, with approximately 470 new cases diagnosed in the United States annually. It affects mainly Caucasian individuals, in whom the tumor development is likely related to sun exposure and chronic immunosuppression. It is an aggressive tumor, and survival is dependent on tumor stage at presentation. Histologically, MCC is a “small blue cell” tumor, differentiated from SCLC and the small cell variant of melanoma by immunohistochemistry, using S-100, CK20, CK7, and TTF-1. While several genetic abnormalities have been reported, no strong pathogenetic correlation has been identified. The diagnostic evaluation includes CT imaging; octreotide and PET scans may be helpful in diagnosis and staging. Wide surgical excision with tumor-free margins is
599
the primary mode of treatment for nonmetastatic disease, and sentinel lymph node biopsy is recommended for the staging of the disease. Adjuvant radiation therapy at 45–50 Gy to the primary site and involved lymph nodes can prevent local recurrences and may improve survival rates. Chemotherapy, patterned after regimens for SCLC, leads to tumor regression in up to 70% of cases with metastatic disease, but has no established role in the adjuvant setting.
REFERENCES 1. Merkel F. Tastzellen und Tastk¨orperchen bei den Haustieren und beim Menschen. Arch Mikrosk Anat Entwmech 1875; 11: 636. 2. Winkelmann RK, Breathnach AS. The Merkel cell. J Invest Dermatol 1973; 60: 2. ¨ 3. Pinkus F. Uber einen bisher unbekannten Nebenapparat am Haarsystem des Menschen: Haarscheiben. Dermatol Z 1902; 9: 465. 4. Grim M, Halata Z. Developmental origin of avian Merkel cells. Anat Embryol (Berl) 2000; 202: 401. 5. Moll I, Zieger W, Schmelz M. Proliferative Merkel cells were not detected in human skin. Arch Dermatol Res 1996; 288: 184. 6. Frigerio B, et al. Merkel cell carcinoma of the skin: the structure and origin of normal Merkel cells. Histopathology 1983; 7: 229. 7. Toker C. Trabecular carcinoma of the skin. Arch Dermatol 1972; 105: 107. 8. Tang CK, Toker C. Trabecular carcinoma of the skin: an ultrastructural study. Cancer 1978; 42: 2311. 9. De Wolf-Peeters C, et al. A cutaneous APUDoma or Merkel cell tumor? A morphologically recognizable tumor with a biological and histological malignant aspect in contrast with its clinical behavior. Cancer 1980; 46: 1810. 10. Goessling W, McKee PH, Mayer RJ. Merkel cell carcinoma. J Clin Oncol 2002; 20: 588. 11. Miller RW, Rabkin CS. Merkel cell carcinoma and melanoma: etiological similarities and differences. Cancer Epidemiol Biomarkers Prev 1999; 8: 153. 12. Chuang TY, Su WP, Muller SA. Incidence of cutaneous T cell lymphoma and other rare skin cancers in a defined population. J Am Acad Dermatol 1990; 23: 254. 13. Hodgson NC. Merkel cell carcinoma: changing incidence trends. J Surg Oncol 2005; 89: 1. 14. Anderson LL, Phipps TJ, McCollough ML. Neuroendocrine carcinoma of the skin (Merkel cell carcinoma) in a black. J Dermatol Surg Oncol 1992; 18: 375. 15. Morrison WH, et al. The essential role of radiation therapy in securing locoregional control of Merkel cell carcinoma. Int J Radiat Oncol Biol Phys 1990; 19: 583. 16. Voog E, et al. Chemotherapy for patients with locally advanced or metastatic Merkel cell carcinoma. Cancer 1999; 85: 2589. 17. Meyer-Pannwitt U, et al. Merkel cell tumor or neuroendocrine skin carcinoma. Langenbecks Arch Chir 1997; 382: 349. 18. Ott MJ, et al. Multimodality management of Merkel cell carcinoma. Arch Surg 1999; 134: 388; discussion 392 – 3. 19. Smith DF, et al. Clinical approach to neuroendocrine carcinoma of the skin (Merkel cell carcinoma). Cancer Control J 2000; 7: 72. 20. Boyle F, Pendlebury S, Bell D. Further insights into the natural history and management of primary cutaneous neuroendocrine (Merkel cell) carcinoma. Int J Radiat Oncol Biol Phys 1995; 31: 315. 21. Kroll MH, Toker C. Trabecular carcinoma of the skin: further clinicopathologic and morphologic study. Arch Pathol Lab Med 1982; 106: 404. 22. Schmid C, et al. Recurrent and subsequently metastasizing Merkel cell carcinoma in a 7-year-old girl. Histopathology 1992; 20: 437. 23. Sonak RA, Trede K, Gerharz CD. Merkel cell tumor of the hand in a 104-year-old patient. Case report with review of the literature. Handchir Mikrochir Plast Chir 1996; 28: 43. 24. Yiengpruksawan A, et al. Merkel cell carcinoma. Prognosis and management. Arch Surg 1991; 126: 1514.
600
CUTANEOUS MALIGNANCIES
25. Hitchcock CL, et al. Neuroendocrine (Merkel cell) carcinoma of the skin. Its natural history, diagnosis, and treatment. Ann Surg 1988; 207: 201. 26. Ratner D, et al. Merkel cell carcinoma. J Am Acad Dermatol 1993; 29: 143. 27. Gollard R, et al. Merkel cell carcinoma: review of 22 cases with surgical, pathologic, and therapeutic considerations. Cancer 2000; 88: 1842. 28. Shaw JH, Rumball E. Merkel cell tumour: clinical behaviour and treatment. Br J Surg 1991; 78: 138. 29. Raaf JH, et al. Trabecular (Merkel cell) carcinoma of the skin. Treatment of primary, recurrent, and metastatic disease. Cancer 1986; 57: 178. 30. Haag ML, Glass LF, Fenske NA. Merkel cell carcinoma. Diagnosis and treatment. Dermatol Surg 1995; 21: 669. 31. Savage P, et al. The natural history and management of Merkel cell carcinoma of the skin: a review of 22 patients treated at the Royal Marsden Hospital. Clin Oncol (R Coll Radiol) 1997; 9: 164. 32. Skelton HG, et al. Merkel cell carcinoma: analysis of clinical, histologic, and immunohistologic features of 132 cases with relation to survival. J Am Acad Dermatol 1997; 37: 734. 33. Gackle HC, et al. Merkel cell tumor of the eyelids: review of the literature and report of 2 patients. Klin Monatsbl Augenheilkd 2000; 216: 10. 34. Metz KA, et al. Merkel cell carcinoma of the eyelid: histological and immunohistochemical features with special respect to differential diagnosis. Graefes Arch Clin Exp Ophthalmol 1998; 236: 561. 35. Soltau JB, Smith ME, Custer PL. Merkel cell carcinoma of the eyelid. Am J Ophthalmol 1996; 121: 331. 36. Hauschild A, et al. Merkel cell carcinoma: follow-up of 10 patients. Current diagnosis and therapy. Langenbecks Arch Chir 1997; 382: 185. 37. Chen KT. Merkel’s cell (neuroendocrine) carcinoma of the vulva. Cancer 1994; 73: 2186. 38. Tomic S, et al. Penile Merkel cell carcinoma. Urology 1995; 45: 1062. 39. Routh A, Hickman BT, Johnson WW. Superior vena cava obstruction from Merkel cell carcinoma. Arch Dermatol 1987; 123: 714. 40. Eggers SD, et al. Paraneoplastic and metastatic neurologic complications of Merkel cell carcinoma. Mayo Clin Proc 2001; 76: 327. 41. Lopez MC, et al. Merkel cell carcinoma associated with a paraneoplastic neurological syndrome. Histopathology 2004; 44: 628. 42. Greenlee JE, et al. Anti-Hu antibodies in Merkel cell carcinoma. Ann Neurol 2002; 52: 111. 43. Allen PJ, et al. Merkel cell carcinoma: prognosis and treatment of patients from a single institution. J Clin Oncol 2005; 23: 2300. 44. Allen PJ, Zhang ZF, Coit DG. Surgical management of Merkel cell carcinoma. Ann Surg 1999; 229: 97. 45. Pitale M, Sessions RB, Husain S. An analysis of prognostic factors in cutaneous neuroendocrine carcinoma. Laryngoscope 1992; 102: 244. 46. Goepfert H, et al. Merkel cell carcinoma (endocrine carcinoma of the skin) of the head and neck. Arch Otolaryngol 1984; 110: 707. 47. LeBoit PE, Crutcher WA, Shapiro PE. Pagetoid intraepidermal spread in Merkel cell (primary neuroendocrine) carcinoma of the skin. Am J Surg Pathol 1992; 16: 584. 48. Iacocca MV, et al. Mixed Merkel cell carcinoma and squamous cell carcinoma of the skin. J Am Acad Dermatol 1998; 39: 882. 49. Gould VE, et al. Biology of disease: neuroendocrine (Merkel) cells of the skin: hyperplasias, dysplasias, and neoplasms. Lab Invest 1985; 52: 334. 50. Llombart B, et al. Clinicopathological and immunohistochemical analysis of 20 cases of Merkel cell carcinoma in search of prognostic markers. Histopathology 2005; 46: 622. 51. Pilotti S, et al. Clinicopathologic correlations of cutaneous neuroendocrine Merkel cell carcinoma. J Clin Oncol 1988; 6: 1863. 52. Shin HJ, Caraway NP. Fine-needle aspiration biopsy of metastatic small cell carcinoma from extrapulmonary sites. Diagn Cytopathol 1998; 19: 177. 53. Collins BT, et al. Fine-needle aspiration of Merkel cell carcinoma of the skin with cytomorphology and immunocytochemical correlation. Diagn Cytopathol 1998; 18: 251.
54. Layfield LJ, Glasgow BJ. Aspiration biopsy cytology of primary cutaneous tumors. Acta Cytol 1993; 37: 679. 55. Gottschalk-Sabag S, Ne’eman Z, Glick T. Merkel cell carcinoma diagnosed by fine-needle aspiration. Am J Dermatopathol 1996; 18: 269. 56. Skoog L, Schmitt FC, Tani E. Neuroendocrine (Merkel-cell) carcinoma of the skin: immunocytochemical and cytomorphologic analysis on fine-needle aspirates. Diagn Cytopathol 1990; 6: 53. 57. Pettinato G, De Chiara A, Insabato L. Diagnostic significance of intermediate filament buttons in fine needle aspirates of neuroendocrine (Merkel cell) carcinoma of the skin. Acta Cytol 1989; 33: 420. 58. Domagala W, et al. Neuroendocrine (Merkel-cell) carcinoma of the skin. Cytology, intermediate filament typing and ultrastructure of tumor cells in fine needle aspirates. Acta Cytol 1987; 31: 267. 59. Alvarez-Gago T, et al. Intermediate filament aggregates in mitoses of primary cutaneous neuroendocrine (Merkel cell) carcinoma. Histopathology 1996; 28: 349. 60. Collaco L, et al. Merkel cell carcinoma of the eyelid: a case report. Eur J Ophthalmol 2000; 10: 173. 61. Leland JY, Shah RP, Adelman HM. A skin lesion found by serendipity. Hosp Pract 2000; 35: 32. 62. Agoff SN, et al. Thyroid transcription factor-1 is expressed in extrapulmonary small cell carcinomas but not in other extrapulmonary neuroendocrine tumors. Mod Pathol 2000; 13: 238. 63. Cheuk W, et al. Immunostaining for thyroid transcription factor 1 and Cytokeratin 20 aids the distinction of small cell carcinoma from Merkel cell carcinoma, but not pulmonary from extrapulmonary small cell carcinomas. Arch Pathol Lab Med 2001; 125: 228. 64. Schmidt U, et al. Cytokeratin and neurofilament protein staining in Merkel cell carcinoma of the small cell type and small cell carcinoma of the lung. Am J Dermatopathol 1998; 20: 346. 65. Narisawa Y, Hashimoto K, Kohda H. Immunohistochemical demonstration of the expression of neurofilament proteins in Merkel cells. Acta Derm Venereol 1994; 74: 441. 66. Shah IA, et al. Neurofilament immunoreactivity in Merkel-cell tumors: a differentiating feature from small-cell carcinoma. Mod Pathol 1993; 6: 3. 67. Jimenez FJ, et al. Ber-EP4 immunoreactivity in normal skin and cutaneous neoplasms. Mod Pathol 1995; 8: 854. 68. Kontochristopoulos GJ, et al. Differentiation between Merkel cell carcinoma and malignant melanoma: An immunohistochemical study. Dermatology 2000; 201: 123. 69. Van Gele M, et al. Mutation analysis of P73 and TP53 in Merkel cell carcinoma. Br J Cancer 2000; 82: 823. 70. Lunder EJ, Stern RS. Merkel-cell carcinomas in patients treated with methoxsalen and ultraviolet A radiation. N Engl J Med 1998; 339: 1247. 71. Hewitt JB, et al. Merkel cell and squamous cell carcinomas arising in erythema ab igne. Br J Dermatol 1993; 128: 591. 72. Cerroni L, Kerl H. Primary cutaneous neuroendocrine (Merkel cell) carcinoma in association with squamous- and basal-cell carcinoma. Am J Dermatopathol 1997; 19: 610. 73. Jones CS, et al. Development of neuroendocrine (Merkel cell) carcinoma mixed with squamous cell carcinoma in erythema ab igne. Arch Dermatol 1988; 124: 110. 74. Silva EG, et al. Endocrine carcinoma of the skin (Merkel cell carcinoma). Pathol Annu 1984; 19(Pt 2): 1. 75. Gomez LG, et al. Association between neuroendocrine (Merkel cell) carcinoma and squamous carcinoma of the skin. Am J Surg Pathol 1983; 7: 171. 76. Brenner B, et al. Second neoplasms in patients with Merkel cell carcinoma. Cancer 2001; 91: 1358. 77. Tsuruta D, et al. Merkel cell carcinoma, Bowen’s disease and chronic occupational arsenic poisoning. Br J Dermatol 1998; 139: 291. 78. Penn I, First MR. Merkel’s cell carcinoma in organ recipients: report of 41 cases. Transplantation 1999; 68: 1717. 79. Urbatsch A, et al. Merkel cell carcinoma occurring in renal transplant patients. J Am Acad Dermatol 1999; 41: 289. 80. Williams RH, et al. Merkel cell carcinoma in a renal transplant patient: increased incidence? Transplantation 1998; 65: 1396.
MERKEL CELL CARCINOMA 81. Veness MJ. Aggressive skin cancers in a cardiac transplant recipient. Australas Radiol 1997; 41: 363. 82. Gooptu C, et al. Merkel cell carcinoma arising after therapeutic immunosuppression. Br J Dermatol 1997; 137: 637. 83. Vazquez-Mazariego Y, et al. Cytogenetic study of neuroendocrine carcinoma of Merkel cells. Cancer Genet Cytogenet 1996; 92: 79. 84. Douds AC, Mellotte GJ, Morgan SH. Fatal Merkel-cell tumour (cutaneous neuroendocrine carcinoma) complicating renal transplantation. Nephrol Dial Transplant 1995; 10: 2346. 85. Formica M, et al. Merkel cell carcinoma in renal transplant recipient. Nephron 1994; 68: 399. 86. Stempfle HU, et al. Rapid growth of cutaneous neuroendocrine (Merkel cell) carcinoma during treatment of refractory cardiac allograft rejection with OKT3 monoclonal antibody. J Heart Lung Transplant 1993; 12: 501. 87. Miller J, et al. Merkel cell carcinoma in a stem cell transplant patient. Dermatol Surg 1998; 24: 913. 88. Catlett JP, Todd WM, Carr ME Jr. Merkel cell tumor in an HIVpositive patient. Va Med Q 1992; 119: 256. 89. Samarendra P, et al. Primary nodal neuroendocrine (Merkel cell) tumor in a patient with HIV infection. South Med J 2000; 93: 920. 90. Colebunders R, et al. Merkel cell carcinoma and multiple basal cell carcinoma in an African albino woman with HIV infection. HIV Med 2004; 5: 452. 91. Burack J, Altschuler EL. Sustained remission of metastatic Merkel cell carcinoma with treatment of HIV infection. J R Soc Med 2003; 96: 238. 92. Calza L, et al. Merkel cell carcinoma in a human immunodeficiency virus-infected patient. Br J Dermatol 2002; 146: 895. 93. Matichard E, et al. Merkel cell carcinoma in a black human immunodeficiency virus-infected patient. Br J Dermatol 2002; 146: 671. 94. Engels EA, et al. Merkel cell carcinoma and HIV infection. Lancet 2002; 359: 497. 95. An KP, Ratner D. Merkel cell carcinoma in the setting of HIV infection. J Am Acad Dermatol 2001; 45: 309. 96. Ziprin P, et al. Two cases of Merkel cell tumour arising in patients with chronic lymphocytic leukaemia. Br J Dermatol 2000; 142: 525. 97. Quaglino D, et al. Association between chronic lymphocytic leukaemia and secondary tumours: unusual occurrence of a neuroendocrine (Merkel cell) carcinoma. Eur Rev Med Pharmacol Sci 1997; 1: 11. 98. Safadi R, et al. Merkel cell tumor in a woman with chronic lymphocytic leukemia. Leuk Lymphoma 1996; 20: 509. 99. Agnew KL, et al. Cutaneous findings in chronic lymphocytic leukaemia. Br J Dermatol 2004; 150: 1129. 100. Vlad R, Woodlock TJ. Merkel cell carcinoma after chronic lymphocytic leukemia: case report and literature review. Am J Clin Oncol 2003; 26: 531. 101. Van Gele M, et al. Molecular analysis of 1p36 breakpoints in two Merkel cell carcinomas. Genes Chromosomes Cancer 1998; 23: 67. 102. Judson H, et al. Structure and mutation analysis of the gene encoding DNA fragmentation factor 40 (caspase-activated nuclease), a candidate neuroblastoma tumour suppressor gene. Hum Genet 2000; 106: 406. 103. Harnett PR, et al. Loss of allelic heterozygosity on distal chromosome 1p in Merkel cell carcinoma. A marker of neural crest origins? Cancer Genet Cytogenet 1991; 54: 109. 104. Leonard JH, et al. Deletion mapping on the short arm of chromosome 1 in Merkel cell carcinoma. Cancer Detect Prev 2000; 24: 620. 105. Maris JM, Matthay KK. Molecular biology of neuroblastoma. J Clin Oncol 1999; 17: 2264. 106. Smedley D, et al. Characterization of chromosome 1 abnormalities in malignant melanomas. Genes Chromosomes Cancer 2000; 28: 121. 107. Kaghad M, et al. Monoallelically expressed gene related to p53 at 1p36, a region frequently deleted in neuroblastoma and other human cancers. Cell 1997; 90: 809. 108. Tsao H, et al. Mutational and expression analysis of the p73 gene in melanoma cell lines. Cancer Res 1999; 59: 172. 109. Carson HJ, Lueck NE, Horten BC. Comparison of mutant and wildtype p53 proteins in Merkel cell carcinoma. Clin Diagn Lab Immunol 2000; 7: 326.
601
110. Leonard JH, et al. Deletion mapping of the short arm of chromosome 3 in Merkel cell carcinoma. Genes Chromosomes Cancer 1996; 15: 102. 111. Dammann R, et al. Epigenetic inactivation of a RAS association domain family protein from the lung tumour suppressor locus 3p21.3. Nat Genet 2000; 25: 315. 112. Burbee DG, et al. Epigenetic inactivation of RASSF1A in lung and breast cancers and malignant phenotype suppression. J Natl Cancer Inst 2001; 93: 691. 113. Leonard JH, Leonard P, Kearsley JH. Chromosomes 1, 11, and 13 are frequently involved in karyotypic abnormalities in metastatic Merkel cell carcinoma. Cancer Genet Cytogenet 1993; 67: 65. 114. Harle M, et al. Comparative genomic hybridization (CGH) discloses chromosomal and subchromosomal copy number changes in Merkel cell carcinomas. J Cutan Pathol 1996; 23: 391. 115. Larsimont D, Verhest A. Chromosome 6 trisomy as sole anomaly in a primary Merkel cell carcinoma. Virchows Arch 1996; 428: 305. 116. Sandbrink F, et al. Short communication: deletion 7q, trisomy 6 and 11 in a case of Merkel-cell carcinoma. Cancer Genet Cytogenet 1988; 33: 305. 117. Amo-Takyi BK, et al. Diagnostic relevance of chromosomal insitu hybridization in Merkel cell carcinoma: targeted interphase cytogenetic tumour analyses. Histopathology 1999; 34: 163. 118. Van Gele M, et al. Frequent allelic loss at 10q23 but low incidence of PTEN mutations in Merkel cell carcinoma. Int J Cancer 2001; 92: 409. 119. Leonard JH, Hayard N. Loss of heterozygosity of chromosome 13 in Merkel cell carcinoma. Genes Chromosomes Cancer 1997; 20: 93. 120. Van Gele M, et al. Characteristic pattern of chromosomal gains and losses in Merkel cell carcinoma detected by comparative genomic hybridization. Cancer Res 1998; 58: 1503. 121. Van Gele M, et al. Gene-expression profiling reveals distinct expression patterns for classic versus variant Merkel cell phenotypes and new classifier genes to distinguish Merkel cell from small-cell lung carcinoma. Oncogene 2004; 23: 2732. 122. Gollub MJ, Gruen DR, Dershaw DD. Merkel cell carcinoma: CT findings in 12 patients. AJR Am J Roentgenol 1996; 167: 617. 123. Carnaille B, et al. Scintiscans and carcinoid tumors. Surgery 1994; 116: 1118. 124. Cirillo F, et al. Merkel cell tumor. Report of case and treatment with octreotide. Minerva Chir 1997; 52: 1359. 125. Lobrano MB, et al. Metastatic carcinoid tumor imaged with CT and a radiolabeled somatostatin analog: a case report. Am J Gastroenterol 1997; 92: 513. 126. Straka JA, Straka MB. A review of Merkel cell carcinoma with emphasis on lymph node disease in the absence of a primary site. Am J Otolaryngol 1997; 18: 55. 127. Kau R, Arnold W. Somatostatin receptor scintigraphy and therapy of neuroendocrine (APUD) tumors of the head and neck. Acta Otolaryngol 1996; 116: 345. 128. Lastoria S, et al. Comparison of labeled MIBG and somatostatin analogs in imaging neuroendocrine tumors. Q J Nucl Med 1995; 39: 145. 129. Kau RJ, et al. Detection of somatostatin receptors in tumors in the area of the head and neck and their clinical importance. Laryngorhinootologie 1994; 73: 21. 130. Kwekkeboom DJ, et al. Somatostatin analogue scintigraphy. A simple and sensitive method for the in vivo visualization of Merkel cell tumors and their metastases. Arch Dermatol 1992; 128: 818. 131. Durani BK, et al. Somatostatin analogue scintigraphy in Merkel cell tumours. Br J Dermatol 2003; 148: 1135. 132. Di Bartolomeo M, et al. Clinical efficacy of octreotide in the treatment of metastatic neuroendocrine tumors. A study by the Italian Trials in Medical Oncology Group. Cancer 1996; 77: 402. 133. Meier G, et al. Successful targeted radiotherapy with 90Y-DOTATOC in a patient with Merkel cell carcinoma. A case report. Oncology 2004; 66: 160. 134. Klein M, et al. Contribution of whole body F-18-FDG-PET and lymphoscintigraphy to the assessment of regional and distant metastases in cutaneous malignant melanoma. A pilot study. Nuklearmedizin 2000; 39: 56.
602
CUTANEOUS MALIGNANCIES
135. Paquet P, et al. An appraisal of 18-fluorodeoxyglucose positron emission tomography for melanoma staging. Dermatology 2000; 200: 167. 136. Acland KM, O’Doherty MJ, Russell-Jones R. The value of positron emission tomography scanning in the detection of subclinical metastatic melanoma. J Am Acad Dermatol 2000; 42: 606. 137. Lampreave JL, et al. PET evaluation of therapeutic limb perfusion in Merkel’s cell carcinoma. J Nucl Med 1998; 39: 2087. 138. Wong CO, Pham AN, Dworkin HJ. F-18 FDG accumulation in an octreotide negative Merkel cell tumor. Clin Positron Imaging 2001; 3: 71. 139. Yao M, et al. Merkel cell carcinoma: two case reports focusing on the role of fluorodeoxyglucose positron emission tomography imaging in staging and surveillance. Am J Clin Oncol 2005; 28: 205. 140. Kokoska ER, et al. Early aggressive treatment for Merkel cell carcinoma improves outcome. Am J Surg 1997; 174: 688. 141. O’Connor WJ, Brodland DG. Merkel cell carcinoma. Dermatol Surg 1996; 22: 262. 142. Gillenwater AM, et al. Merkel cell carcinoma of the head and neck. Effect of surgical excision and radiation on recurrence and survival. Arch Otolaryngol Head Neck Surg 2001; 127: 149. 143. Mohs FE. Chemosurgery, a microscopically controlled method of cancer excision. Arch Surg 1941; 42: 279. 144. O’Connor WJ, Roenigk RK, Brodland DG. Merkel cell carcinoma. Comparison of Mohs micrographic surgery and wide excision in eighty-six patients. Dermatol Surg 1997; 23: 929. 145. Boyer JD, et al. Local control of primary Merkel cell carcinoma: review of 45 cases treated with Mohs micrographic surgery with and without adjuvant radiation. J Am Acad Dermatol 2002; 47: 885. 146. Smith DE, et al. Cutaneous neuroendocrine (Merkel cell) carcinoma. A report of 35 cases. Am J Clin Oncol 1995; 18: 199. 147. Cabanas RM. An approach for the treatment of penile carcinoma. Cancer 1977; 39: 456. 148. Balch CM, Ross MI. Sentinel lymphadenectomy for melanoma – is it a substitute for elective lymphadenectomy? Ann Surg Oncol 1999; 6: 416. 149. Balch CM, et al. 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 1996; 224: 255; discussion 263 – 6. 150. Krag D, et al. The sentinel node in breast cancer – a multicenter validation study. N Engl J Med 1998; 339: 941. 151. McMasters KM, et al. Sentinel lymph node biopsy for breast cancer: a suitable alternative to routine axillary dissection in multi-institutional practice when optimal technique is used. J Clin Oncol 2000; 18: 2560. 152. Pfeifer T, Weinberg H, Brady MS. Lymphatic mapping for Merkel cell carcinoma. J Am Acad Dermatol 1997; 37: 650. 153. Messina JL, et al. Selective lymphadenectomy in patients with Merkel cell (cutaneous neuroendocrine) carcinoma. Ann Surg Oncol 1997; 4: 389. 154. Bilchik AJ, et al. Universal application of intraoperative lymphatic mapping and sentinel lymphadenectomy in solid neoplasms. Cancer J Sci Am 1998; 4: 351. 155. Ames SE, Krag DN, Brady MS. Radiolocalization of the sentinel lymph node in Merkel cell carcinoma: a clinical analysis of seven cases. J Surg Oncol 1998; 67: 251. 156. Hill AD, Brady MS, Coit DG. Intraoperative lymphatic mapping and sentinel lymph node biopsy for Merkel cell carcinoma. Br J Surg 1999; 86: 518. 157. Wasserberg N, et al. Applicability of the sentinel node technique to Merkel cell carcinoma. Dermatol Surg 2000; 26: 138. 158. Zeitouni NC, Cheney RT, Delacure MD. Lymphoscintigraphy, sentinel lymph node biopsy, and Mohs micrographic surgery in the treatment of Merkel cell carcinoma. Dermatol Surg 2000; 26: 12. 159. Wasserberg N, et al. Sentinel-node guided lymph-node dissection for Merkel cell carcinoma. Eur J Surg Oncol 1999; 25: 444. 160. Leonard JH, et al. Radiation sensitivity of Merkel cell carcinoma cell lines. Int J Radiat Oncol Biol Phys 1995; 32: 1401. 161. Ashby MA, et al. Primary cutaneous neuroendocrine (Merkel cell or trabecular carcinoma) tumour of the skin: a radioresponsive tumour. Clin Radiol 1989; 40: 85. 162. Cotlar AM, Gates JO, Gibbs FA Jr. Merkel cell carcinoma: combined surgery and radiation therapy. Am Surg 1986; 52: 159.
163. Pacella J, et al. The role of radiotherapy in the management of primary cutaneous neuroendocrine tumors (Merkel cell or trabecular carcinoma): experience at the Peter MacCallum Cancer Institute (Melbourne, Australia). Int J Radiat Oncol Biol Phys 1988; 14: 1077. 164. Meeuwissen JA, Bourne RG, Kearsley JH. The importance of postoperative radiation therapy in the treatment of Merkel cell carcinoma. Int J Radiat Oncol Biol Phys 1995; 31: 325. 165. Veness MJ, et al. Merkel cell carcinoma: improved outcome with adjuvant radiotherapy. ANZ J Surg 2005; 75: 275. 166. Fenig E, et al. The role of radiation therapy and chemotherapy in the treatment of Merkel cell carcinoma. Cancer 1997; 80: 881. 167. Bedane C, et al. Neuroendocrine primary cutaneous carcinoma. Therapeutic aspects in 13 patients. Ann Dermatol Venereol 1996; 123: 443. 168. Bischof M, et al. Merkel cell carcinoma: the role of radiation therapy in general management. Strahlenther Onkol 1999; 175: 611. 169. Hohaus K, et al. Merkel cell carcinoma – a retrospective analysis of 17 cases. J Eur Acad Dermatol Venereol 2003; 17: 20. 170. Eich HT, et al. Role of postoperative radiotherapy in the management of Merkel cell carcinoma. Am J Clin Oncol 2002; 25: 50. 171. Marks ME, Kim RY, Salter MM. Radiotherapy as an adjunct in the management of Merkel cell carcinoma. Cancer 1990; 65: 60. 172. Wilder RB, et al. Merkel cell carcinoma. Improved locoregional control with postoperative radiation therapy. Cancer 1991; 68: 1004. 173. Nathu RM, Mendenhall WM, Parsons JT. Merkel cell carcinoma of the skin. Radiat Oncol Investig 1998; 6: 233. 174. McAfee WJ, et al. Merkel cell carcinoma. Cancer 2005; 104: 1761 – 4. 175. Poulsen M, et al. Analysis of toxicity of Merkel cell carcinoma of the skin treated with synchronous carboplatin/etoposide and radiation: a Trans-Tasman Radiation Oncology Group study. Int J Radiat Oncol Biol Phys 2001; 51: 156. 176. Poulsen M, et al. High-risk Merkel cell carcinoma of the skin treated with synchronous carboplatin/etoposide and radiation: a Trans-Tasman Radiation Oncology Group Study – TROG 96:07. J Clin Oncol 2003; 21: 4371. 177. Muggianu M, et al. Radiotherapy and hyperthermia in the treatment of primary Merkel cell carcinoma of the skin: a case report. Bull Cancer Radiother 1994; 81: 237. 178. Knox SJ, Kapp DS. Hyperthermia and radiation therapy in the treatment of recurrent Merkel cell tumors. Cancer 1988; 62: 1479. 179. Merkel cell carcinoma. Clinical practice guidelines in oncology. J Natl Compr Cancer Netw 2005; 2: 80. 180. George TK, di Sant’agnese PA, Bennett JM. Chemotherapy for metastatic Merkel cell carcinoma. Cancer 1985; 56: 1034. 181. Wynne CJ, Kearsley JH. Merkel cell tumor. A chemosensitive skin cancer. Cancer 1988; 62: 28. 182. Pectasides D, et al. Chemotherapy for Merkel cell carcinoma with carboplatin and etoposide. Am J Clin Oncol 1995; 18: 418. 183. Feun LG, et al. Chemotherapy for metastatic Merkel cell carcinoma. Review of the M.D. Anderson Hospital’s experience. Cancer 1988; 62: 683. 184. Ferrau F, Micali G, Guitart J. Merkel cell carcinoma of the scalp: dramatic resolution with primary chemotherapy. J Am Acad Dermatol 1994; 31: 271. 185. Bajetta E, et al. 5-Fluorouracil, dacarbazine, and epirubicin in the treatment of patients with neuroendocrine tumors. Cancer 1998; 83: 372. 186. Fenig E, et al. Oral etoposide for Merkel cell carcinoma in patients previously treated with intravenous etoposide. Am J Clin Oncol 2000; 23: 65. 187. Eng TY, et al. Treatment of recurrent Merkel cell carcinoma: an analysis of 46 cases. Am J Clin Oncol 2004; 27: 576. 188. Eng TY, et al. Treatment of Merkel cell carcinoma. Am J Clin Oncol 2004; 27: 510. 189. Tai PT, et al. Chemotherapy in neuroendocrine/Merkel cell carcinoma of the skin: case series and review of 204 cases. J Clin Oncol 2000; 18: 2493. 190. Dirix LY, et al. Tumor lysis syndrome in a patient with metastatic Merkel cell carcinoma. Cancer 1991; 67: 2207. 191. Slichenmyer WJ, LeMaistre CF, Von Hoff DD. Response of metastatic adenoid cystic carcinoma and Merkel cell tumor to high-dose
MERKEL CELL CARCINOMA
192.
193.
194.
195. 196. 197.
melphalan with autologous bone marrow transplantation. Invest New Drugs 1992; 10: 45. Waldmann V, et al. Transient complete remission of metastasized Merkel cell carcinoma by high-dose polychemotherapy and autologous peripheral blood stem cell transplantation. Br J Dermatol 2000; 143: 837. Olieman AF, et al. Hyperthermic isolated limb perfusion with tumor necrosis factor alpha, interferon gamma, and melphalan for locally advanced nonmelanoma skin tumors of the extremities: a multicenter study. Arch Surg 1999; 134: 303. Bajetta E, et al. Treatment of metastatic carcinoids and other neuroendocrine tumors with recombinant interferon-alpha-2a. A study by the Italian Trials in Medical Oncology Group. Cancer 1993; 72: 3099. Zilembo N, et al. Salvage treatment after r-interferon alpha-2a in advanced neuroendocrine tumors. Acta Oncol 1993; 32: 245. Durand JM, et al. Treatment of Merkel cell tumor with interferonalpha-2b. Br J Dermatol 1991; 124: 509. Ito Y, et al. Merkel cell carcinoma. A successful treatment with tumor necrosis factor. Arch Dermatol 1989; 125: 1093.
603
198. Hata Y, et al. Two cases of Merkel cell carcinoma cured by intratumor injection of natural human tumor necrosis factor. Plast Reconstr Surg 1997; 99: 547. 199. Moll I, et al. Differences of bcl-2 protein expression between Merkel cells and Merkel cell carcinomas. J Cutan Pathol 1996; 23: 109. 200. Jansen B, et al. Farnesylthiosalicylic acid inhibits the growth of human Merkel cell carcinoma in SCID mice. J Mol Med 1999; 77: 792. 201. Feinmesser M, et al. Expression of the apoptosis-related oncogenes bcl-2, bax, and p53 in Merkel cell carcinoma: can they predict treatment response and clinical outcome? Hum Pathol 1999; 30: 1367. 202. Kennedy MM, et al. Expression of bcl-2 and p53 in Merkel cell carcinoma. An immunohistochemical study. Am J Dermatopathol 1996; 18: 273. 203. Schlagbauer-Wadl H, et al. Bcl-2 antisense oligonucleotides (G3139) inhibit Merkel cell carcinoma growth in SCID mice. J Invest Dermatol 2000; 114: 725. 204. http://www.cancer.gov/clinicaltrials/MSKCC-03105, accessed Aug 10, 2005.
Section 10 : Neurological Malignancies
54
Melanotic Lesions of the Meninges Paul L. Moots and Michael L. Edgeworth
INTRODUCTION AND HISTORICAL BACKGROUND Visual inspection of the meninges at autopsy often reveals a slight pigmentation along the ventral surface of the brain stem. Since the observations of Virchow, it has been known that this pigmentation is from melanin contained in melanocytes, and has the capacity to develop into primary meningeal melanocytic tumors. There are a number of ways in which melanotic tumors can arise in the meninges. They include: (i) derivation from dysplastic melanosis, typically in conjunction with cutaneous pigmented lesions, (ii) as solitary histologically benign melanocytoma, and (iii) as primary meningeal melanoma in a form similar to leptomeningeal involvement by metastatic melanoma. In the latter two situations, the neoplasm arises from meningeal melanocytes and there may or may not be any associated developmental dysplasia. Each of these forms is rare. The first two forms have the capacity to evolve into frankly malignant melanoma. The advent of magnetic resonance imaging (MRI) has greatly aided in the evaluation of meningeal diseases in general, and is particularly good at identifying melanin-containing lesions. Yet, these forms of melanosis/melanoma are so rare that confirmation by cerebrospinal fluid (CSF) cytology or biopsy remains important. All of them tend to lead to progressive neurological problems, and like systemic melanoma, can be indolent for a while but when active are often refractory to therapy.
NEUROCUTANEOUS MELANOSIS Biology Melanin-containing cells, along with the rest of the piaarachnoid, are neural crest derivatives that begin to coalesce around the developing neural tube during the early – mid-portion of fetal development. Melanocytes are found in normal fetal meninges at or about 5 months gestation. Melanocytes also constitute the ocular choroids and uvea, and other nondermal structures. At any of these sites, there exists a potential for occurrence of melanoma. Given the important interactions between mesodermal elements, neural crest–derived structures, and the developing
neural tube, it is not surprising that perturbations of the normal developmental sequence can give rise to anomalies, as well as to precancerous, dysplastic, or frankly malignant lesions. Such is the case in neurocutaneous melanosis (NCM). Fox believed this disorder to be a congenital dysplasia of embryonal neuroectoderm-derived neural crest cells.1 This disorder is largely confined to young children, and is associated with congenital anomalies of the central nervous system (CNS), particularly Dandy–Walker cyst of the fourth ventricle. Other anomalies such as lipoma and arachnoid cysts occur less often. NCM is classified as a phakomatosis; however, it is not a familial disorder. A mouse model of the disease has been developed, which provides some insight into the biology of this illness. Hepatocyte growth factor/scatter factor (HGF/SF) is a cytokine involved in the inductive interaction between mesenchymal and epithelial cells. The inductive signaling pathways stimulated by HGF/SC involves c-met, a receptor with tyrosine kinase activity found on epithelial cells. Transgenic mice that inappropriately express HGF/SC have multiple developmental abnormalities including abnormal melanocyte development. These mice demonstrate meningeal melanosis and pigmented nevi with a marked resemblance to the human condition.2 These mice sometimes also develop rhabdomyosarcoma, a feature found in few patients with NCM. The C-met oncogene appears to be important in the aggressiveness, metastatic potential, and distribution of metastases in melanoma.3,4 Immunohistochemical studies of an intracranial melanotic lesion in a 4-year old child with NCM revealed staining for c-met in a pattern indicating altered expression.
Pathology NCM is generally apparent in infancy, presenting as congenital nevi. These are often solitary, large, pigmented nevus posteriorly located on the trunk, or head and shoulders. Multiple small nevi sometimes present without a large nevus. Histologically, these are benign melanotic nevi. However, there are case reports of conversion to frank melanoma. The coexisting melanotic lesions of the meninges also vary considerably. The gross appearance may be limited
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
606
NEUROLOGICAL MALIGNANCIES
to few small, flat grayish plaques scattered throughout the cerebral and spinal meninges, with some predilection for the posterior fossa and basal meninges. A more profound case may demonstrate multiple melanotic nodules, sometimes confluent or aggregate, that are large enough to produce compression of the underlying brain, spinal cord, or cranial or spinal nerve roots. An obvious tumor mass is found in about half of the patients, usually in the frontal or temporal region. In the most profound cases, the entire meningeal surface is covered with a dark firm material that fills and obscures the underlying sulci. Gross inspection of cut brain sections at autopsy often reveals hydrocephalus and distortion or compression of the underlying brain. Furthermore, some areas of the brain parenchyma adjacent to the meninges appear brown or coffee-colored and are expanded. This appearance results from the tracking of melanotic cells along the pial sheath investing the penetrating vessels, and expansion of the Virchow –Robbins spaces. A few reported cases describe solitary intraparenchymal melanocytic mass lesions that presumably arose within the Virchow –Robin space and grew without extension to the surface subarachnoid space.5,6 Other maldevelopmental lesions of the nervous system may sometimes arise in association with NCM. The most important of these is the Dandy–Walker malformation of the posterior fossa. Arachnoid cysts and intraspinal lipomas may also be present. The microscopic appearance of the meningeal lesions is often histologically benign, and commonly termed melanosis. This can be nodular or diffuse. Yet the appearance of invasion along vessels accompanied by areas of higher cellularity, varying in appearance from spindle-shaped to epithelioid, with some atypia suggests a neoplastic character. In about 60% of cases, one or multiple foci on a widely affected surface show features of malignant melanoma including mitoses, necrosis, hemorrhage, and invasion through the basal laminae. Although the finding of metastatic melanoma outside the nervous system is usually taken as evidence against the diagnosis of NCM, there are a few well-documented cases of melanoma arising intracranially metastasized to other organs. Most cases involve seeding of the abdominal cavity through a ventriculoperitoneal shunt.
Clinical Features Criteria for the diagnosis of NCM were first defined by Fox.1 These included unduly large or numerous pigmented nevi, an absence of malignant changes in the skin lesions as that would raise suspicion that the CNS lesions were metastatic rather than primary, and no evidence of melanoma in other organs, a condition predicated on the belief that melanoma arising in the meninges would not metastasize elsewhere. Kadonaga and Frieden published revised diagnostic criteria in 1991 that include: (i) large (>20 cm for adults;>9 cm on the head or >6 cm on the trunk for infants) or multiple (>3) congenital melanotic nevi associated with meningeal melanosis or melanoma, (ii) no evidence of cutaneous melanoma unless the meningeal lesions are pathologically
benign, and (iii) no evidence of meningeal melanoma except in patients whose cutaneous lesions are histologically benign.7 The nevi are most commonly present on the trunk, often posteriorly and covering a large area. A posterior lumbosacral “bathing trunk” distribution is most common followed by the occipital and upper back “cape” distribution. Giant nevi that are not posterior midline lesions are rarely associated with meningeal melanosis. The nevi are often apparent at birth, and the vast majority of cases are diagnosed in the first 2 years, but a few reported cases were diagnosed in adulthood.8 The female : male ratio is 1 : 1, and no specific racial group is known to be at higher risk.7 The neurological manifestations also present most often in the first 2 years, but a few individuals do not become symptomatic until their teens or early- to mid-adulthood. Mass effect and hydrocephalus combine to make symptoms of elevated intracranial pressure the most common presenting feature with headache, lethargy, vomiting, and bulging fontanelles or rapid increase in head circumference. Seizures, cranial nerve palsies, ataxia due to cerebellar anomalies, myelopathy, and a long list of less-common neurologic symptoms because of localized compressive changes are reported.
Diagnosis MRI rapidly replaced all other neuro-imaging techniques for the assessment of all types of meningeal pathology. The case is even stronger for melanotic lesions. These often have signal characteristics that are unusual for CNS mass lesions. Normal melanin in the meninges can sometimes be seen as hypointensity on the T2-weighted MRI sequence along the anterior surface of the medulla. Melanocytic lesions are usually hyperintense on T1 and hypointense on T2 sequences. These characteristics are because of the presence of melanin, but are similar to the appearance of recent hemorrhage. Since hemorrhage is a relatively common observation in intracranial metastatic melanoma, the interpretation of these scans can be very difficult. The meningeal lesions often enhance intensely with intravenous contrast. The adjacent superficial brain parenchyma often appears abnormal also because of the perivascular infiltration of melanocytes. In NCM, the MRI findings are most conspicuous around the cerebellum/posterior fossa and the anterior temporal lobes.9 CSF studies are sometimes diagnostic of NCM. Melanin itself can produce a very dark, xanthochromic appearance. However, xanthochromia occurs in many conditions because of recent hemorrhage or very high protein levels. Measurement of 5-S-cysteinyl dopa, a melanin precursor, in CSF is a very specific way to establish the diagnosis.10 The CSF protein level is often very high, and occasionally CSF glucose level is low as with other forms of neoplastic meningitis. Cytology is diagnostic in some cases. In situations where CSF studies are uninformative or cannot be obtained, an open biopsy of the meninges is the best diagnostic procedure. There is a tendency to overdiagnose malignant melanoma, particularly in limited biopsies.7 By comparison, when one reviews autopsy material, the wide spectrum of findings typical of NCM, ranging from benignappearing melanosis to invasive melanoma, can be put into
MELANOTIC LESIONS OF THE MENINGES
context more clearly. This difference is important when considering the prognosis for NCM, which is more indolent than that of leptomeningeal melanoma.
Treatment Neurosurgical management of NCM is largely that of diagnostic procedures and treatment of hydrocephalus. There are instances where mass lesions in the supratentorial compartment, posterior fossa, or spinal canal require urgent decompression. Dispersion of the lesions along the meninges and infiltration into the parenchyma make the therapeutic value of resection limited. Yet, the limited efficacy of radiation and chemotherapy for these lesions also influence the decision on surgery, particularly in the face of progressive neurological deficits that are partially attributed to compression. Other surgical issues revolve around management of the skin lesion. Both for cosmetic purposes and for reduction in cancer risk because of malignant degeneration, excision of the congenital nevus is often considered. The lifetime risk of malignant degeneration has been estimated at about 5% in prospective trials. The risk of developing symptomatic neurological symptoms because of NCM in the setting of a large congenital melanocytic nevus has been estimated at between 2 and 15%.11 The life expectancy of patients with NCM is sufficiently poor that the decision making regarding a major surgical procedure to reduce a lifetime cancer risk for these infants is very difficult. MRI scanning to exclude patients with meningeal involvement, and/or deferring a surgical decision until after the peak period of risk for developing neurologic symptoms in early- to mid-childhood are ways to work through these difficult options.12 Radiation therapy, including craniospinal radiation, has been utilized for progressive intracranial lesions of NCM, but it is not very effective. Since the majority of patients are very young children, the delayed sequelae of CNS radiation can be anticipated to be severe. Thus, focal radiotherapy of lesions that have progressed through other modalities in an older child is the most rational use of this modality. The few case reports of radiation and chemotherapy for childhood meningeal melanoma include treatment with DTIC and cisplatin, and another with cisplatin and VP-16 have reported survival of 15 and 54 months.13 Other combinations have been reported in the rare occurrence of childhood malignant melanoma such as vincristine, actinomycin, and cyclophosphamide.14 A case report of treatment with temozolomide plus cisplatin for metastatic meningeal melanoma suggests that to be a reasonable consideration.15 In reports of treatment of brain metastases from melanoma, temozolomide produces a response in approximately 7% and stable disease in 30% of patients.16 – 18 Nitrosoureas are also active against melanoma and penetrate the CNS well. Fotemustine demonstrates an equivalent response rate (7%).19 Other approaches that have been applied to systemic melanoma, such as thalidomide, might also be considered. Intrathecal therapies with methotrexate and with IL-2 have been reported.20 Intrathecal therapy for adults with leptomeningeal melanoma produces occasional responses. A randomized trial of solid tumor patients including three with leptomeningeal melanoma, methotrexate produced one response in three
607
patients. DepoCyt (sustained release ARA-C) did not (0/2).21 Thiotepa has also been used.
Prognosis The prognosis for NCM is poor; however, a few patients are diagnosed during adolescence and early adulthood and thus have relatively long asymptomatic periods. Additionally, some patients have a relatively indolent, albeit progressive course after symptoms appear. Almost all patients with progressive CNS disease succumb. This is the case whether the meningeal lesion is largely melanosis or is high-grade melanoma. Essentially, all patients are symptomatic with neurological abnormalities. As reported by Kadonaga and Frieden 70% of the patients die before 10 years of age and only 4 of the 36 they reviewed lived beyond the age of 25. Fifty percent of the patients died within 3 years of developing neurologic symptoms.7
OTHER MELANOTIC LESIONS OF THE MENINGES Melanotic lesions unrelated to NCM have been recognized for many years, and most commonly take the form of isolated nodular masses attached to the meninges. These are melanocytomas. For many years, they were categorized as melanotic meningiomas. Some overlap between these lesions and NCM exists in that melanocytomas can be associated with melanotic skin lesions, a classic example being the association with the nevus of Ota, a congenital nevus involving a portion of the forehead, eyelid, and sclera.22 This fosters the concept that some meningeal tumors not associated with NCM also arise as the result of abnormal neural crest development. Also, like NCM, they sometimes evolve into a malignant process. Finally there is the rare occurrence of meningeal melanoma, in essence a primary form of “leptomeningeal carcinomatosis”, or more properly “melanomatosis”. This highly aggressive process presumably arises from meningeal melanocytes, although distinction from the more common problem of metastatic meningeal involvement by malignant melanoma without a defined site of origin can be difficult. It is also important to note that many different types of meningeal tumors are sometimes pigmented. Some tumors that are correctly classified as meningiomas are pigmented. A variant of schwannoma, the melanotic schwannoma, is pigmented and often demonstrates spindle-shaped cells that become indistinguishable from melanoma and shares some of the immunohistochemical staining characteristics, such as S-100 positivity. Melanin pigmentation is also sometimes apparent in neurofibromas, choroid plexus tumors, ependymomas, and primitive neuroectodermal tumors, as well as in metastatic melanoma.
Meningeal Melanocytoma The clinical and pathological features of 33 cases of melanotic meningeal neoplasms were presented by Brat et al.23 This series did not include any patients with known cutaneous manifestation or any case associated with diffuse leptomeningeal melanosis. Seventeen of the 33 lesions were
608
NEUROLOGICAL MALIGNANCIES
melanocytomas most often composed of spindle-shaped cells. Two were amelanotic, a feature that adds variation to the usual MRI appearance. A small percentage had cellular atypia and rare mitotic figures. Immunohistochemical staining was uniformly positive for S-100 protein and HMB45, and EMA staining was negative. MIB-1 labeling index was less than 2%. The mean age was 51 years (range 16–73 years), which is lower than that of meningiomas. They most often were located in the spinal canal (n = 11). Three arose in the region of Meckel’s cave, a fold of dura along the skull base that transmits a portion of the trigeminal nerve. A gross total resection was achieved for most of the melanocytomas. Three patients underwent radiation after a subtotal resection. Long-term outcome was excellent, with no patients having recurrence at a median of 3 years. Other series confirm the benefits of resection for patients with these lesions.24 Three of the cases reported by Brat et al. were classified as intermediate grade. These were more cellular, shared the same immunohistochemical profile as the melanocytomas, and had MIB-1 labeling indices of 1 to 4%. All three lesions showed indications of invasion of adjacent neural tissue. The patients were in the 50 to 70 year age range. One tumor recurred after total resection. None underwent radiotherapy. Further follow-up data was very limited. Thirteen cases were classified as malignant melanoma. These were more cellular than the melanocytomas, with considerable atypia, pleomorphism, and necrosis. The mean MIB-1 labeling index was approximately 8% (range 2 to 15%). Immunohistochemical staining was again positive for S-100 and HMB-45, while negative for EMA. The mean age for these patients was 43 years (range 15 to 71), which is lower than in patients with meningiomas. The neoplasms were equally divided among the spinal, supratentorial, and infratentorial compartments. Total resection was accomplished in five patients. The remaining patients were subtotally resected. Four patients received radiation postoperatively, and two others were irradiated for local recurrence. The outcomes observed in this group of patients with malignant melanoma were remarkable. Four of the totally resected patients were recurrence free at 14 to 36 months follow-up, and only one of these had received radiation. Six of the eight subtotally resected patients had recurrence ranging from 8 to 76 months. One patient died postoperatively, and three others died of tumor progression. Two of these patients had recurrence with “multiple lesions in the vicinity of the recurrence”. None developed systemic evidence of melanoma. Thus, it appears that no more than 2 of 13 adult patients with primary malignant melanoma of the meninges developed evidence of meningeal seeding, and none developed subarachnoid dissemination with resultant leptomeningeal melanomatosis (i.e. neoplastic meningitis). Also, it is remarkable that four patients remained progression-free over long intervals after total resection. An accumulated series of 89 reported cases of meningeal melanocytoma was analyzed by Rades et al.25 The mean age was 45 years with slightly more females than males. The tumors were equally divided between the intracranial and spinal compartments, with the most common sites being the thoracic and cervical spinal canal. The most common
intracranial sites were the region of Meckel’s cave and the posterior fossa. Complete resection was achieved in 49 patients, 3 of whom also received radiation. Incomplete resection was achieved in 40 patients of whom 17 were irradiated. Local control and survival were significantly worse in the group treated with subtotal resection and no radiation than in any of the other three treatment groups. That group had a 5-year local control rate of 18% and a 5-year survival rate of 46%. By comparison, the 5-year outcomes for the other treatment groups ranged from 72 to 100% for local control, and 100% survival for all three groups. A trend toward improved local control was seen with a radiation dose of 45 to 55 Gy when compared with lower doses. These series are extremely informative, and yet the nature of these surgical series includes an important, inherent bias. Patients with multifocal meningeal involvement by cancer would frequently not be considered candidates for resection. Thus, the remarkably low incidence of subarachnoid dissemination in the Brat et al. series may reflect this bias. Cases such as that of Bydon et al. demonstrate the malignant potential of these initially localized tumors.26 They describe a 79-year old patient who presented with an intradural, extramedullary spinal canal mass at the T8 level. A CT scan of the brain, obtained preoperatively, was unremarkable. The mass was resected and histologically was a melanocytoma of low histologic grade. It demonstrated the characteristic immunophenotype, being positive for S-100, vimentin, and HMB-45. EMA staining was negative. Immunostaining for collagen type IV did not show staining enveloping individual tumor cells, as would be seen in schwannomas. A careful search for cutaneous, ocular, and systemic melanoma was negative. Four months postoperatively additional enhancing lesions were identified on MRI in the conus medullaris and in the fourth ventricle. CSF cytology was negative. The patient was treated with intrathecal methotrexate and radiation. The abnormalities identified on MRI progressed leading to death 2 years after diagnosis. It appears that cases of primary leptomeningeal melanoma with subarachnoid dissemination can arise from preexisting localized melanocytomas of any grade. As illustrated by the case study of Bydon et al., some have a relatively indolent course when compared with that of neoplastic meningitis from metastatic melanoma. In those patients, the median survival is 4 months. Few adults present with parenchymal brain lesion that prove to be metastatic; however, a primary source cannot be identified in them after careful investigation. Approximately 2% of patients presenting with neoplastic meningitis lack an identifiable primary carcinoma. Few such patients have primary leptomeningeal melanoma. By comparison, metastatic melanoma accounts for approximately 12% of patients with neoplastic meningitis. The approaches that have found some application in the treatment of neoplastic meningitis–primarily focal radiation or infrequently craniospinal radiation, and intrathecal or systemic chemotherapies–should be considered as potential therapies for each of the various forms of meningeal involvement by melanotic lesions.27 Those related to an underlying dysplastic process seem to
MELANOTIC LESIONS OF THE MENINGES
evolve in an indolent fashion more often than the more common neoplastic meningitis related to melanoma.
REFERENCES 1. Fox H. Neurocutaneous melanosis. In Vinken PJ, Ang LC, Bryun GW (eds) Handbook of Clinical Neurology. The Phakomatoses. New York: American Elsevier Publishing Company, 1972: Vol. 14, Chap. 14: 414 – 428. 2. Takayam H, et al. Immunohistochemical detection of the c-met protooncogene product in the congenital melanocytic nevus of an infant with neurocutaneous melanosis. J Am Acad Dermatol 2001; 44: 538 – 40. 3. Cruz J, et al. Expression of c-met tyrosine kinase receptor is biologically and prognostically relevant for primary cutaneous malignant lemanoma. Oncology 2003; 65: 72 – 82. 4. Economou MA, et al. Receptors for the liver synthesized growth factors IGF-1 and HGF/SF in uveal melanoma: intercorrelation and prognostic implications. Invest Ophthalmol Vis Sci 2005; 46: 4372 – 5. 5. Rubinstein LJ. Tumors and tumor-like lesions of maldevelopmental origin. Tumors of the Central Nervous System. Washington, DC: Atlas of Tumor Pathology: Fascicle 6 Armed Forces Institute of Pathology, 1972: 285 – 311. 6. Harkin JC, Reed RJ. Tumors and lesions of neurofibromatosis and other types of neurocutaneous phakomatosis. Tumors of the Peripheral Nervous System. Washington, DC: Atlas of Tumor Pathology: Fascicle 3 Armed Forces Institute of Pathology, 1968: 67 – 106. 7. Kadonaga JN, Frieden IJ. Neurocutaneous melanosis: definition and review of the literature. J Am Acad Dermatol 1991; 24: 747 – 55. 8. Oka H, et al. Leptomeningeal melanomatosis with multiple cutaneous pigmented nevi: tumor cell proliferation and malignant transformation in an autopsy case. J Neuro-oncol 1999; 44: 41 – 5. 9. Byrd Se, et al. MR of leptomeningeal melanosis in children. Eur J Radiol 1995; 20: 93 – 9. 10. Kamei S, et al. Measurement and cytologic demonstration of 5-Scysteinyl dopa for the clinical diagnosis of primary leptomeningeal melanoma. Neurology 1994; 44: 175 – 6. 11. D Rocco F, et al. Neurocutaneous melanosis. Child Dis Nerv 2004; 20: 23 – 8. 12. Plikaitis CM, David LR, Argenta LC. Neurocutaneous melanosis: clinical presentations. J Craniofac Surg 2005; 16: 921 – 5.
609
13. Allcutt D, et al. Primary leptomeningeal melanoma: an unusually aggressive tumor in childhood. Neurosurgery 1993; 32: 721 – 9. 14. Makin GW, et al. Leptomeningeal melanoma in childhood. Cancer 1999; 86: 878 – 86. 15. Samaggi A, et al. Temozolomide and cisplatin in the treatment of leptomeningeal metastatic involvement from melanoma: a case report. Neurol Sci 2002; 23: 257 – 8. 16. Hofmann M, et al. Temozolomide with or without radiotherapy in melanoma with unresectable brain metastases. J Neurooncol 2006; 76(1): 59 – 64. 17. Agarwala SS, et al. Temozolomide for the treatment of brain metastases associated with metastatic melanoma. J Clin Oncol 2004; 22: 2101 – 7. 18. Bafaloukos D, Gogas H. The treatment of brain metastases in melanoma patients. Cancer Treat Rev 2004; 30: 515 – 20. 19. Mornex F, et al. A prospective randomized multicentre phase III trial of fotemustine plus whole brain radiation versus fotemustine alone in cerebral metastases of malignant melanoma. Melanoma Res 2003; 13: 97 – 103. 20. Fathallah HM, et al. Response of primary leptomeningeal melanoma to intrathecal recombinant interleukin-2. A case report. Cancer 1996; 77: 1544 – 50. 21. Glantz MJ, et al. A randomized trial comparing intrathecal sustainedrelease Cytarabine (DepoCyt) to intrathecal methotrexate in patients with neoplastic meningitis form solid tumors. Clin Cancer Res 1999; 5: 3394 – 402. 22. Balmaceda CM, et al. Nevus of Ota and leptomeningeal melanocytic lesions. Neurology 1993; 43: 381 – 6. 23. Brat DJ, et al. Primary melanocytic neoplasms of the central nervous system. Am J Surg Pathol 1999; 23: 745 – 54. 24. Fagundes-Pereyra WJ, et al. Meningeal melanocytoma of the posterior fossa: case report and review of the literature. Surg Neurol 2005; 63: 269 – 73. 25. Rades D, et al. Therapy of meningeal melanocytomas. Cancer 2004; 100: 2442 – 7. 26. Bydon A, Guutierrez JA, Mahmood A. Meningeal melanocytoma: an aggressive course for a benign tumor. J Neuro-oncol 2003; 64: 259 – 63. 27. Chamberlain MC. Neoplastic meningitis. J Clin Oncol 2005; 23: 3605 – 13.
Section 10 : Neurological Malignancies
55
Langerhans’ Cell Histiocytosis of the Central Nervous System Rima F. Jubran and Jonathan Finlay
INTRODUCTION
PATHOLOGY
Over a century has passed since Langerhans’ cell histiocytosis (LCH) or histiocytosis X was first described by Hand in 1893, yet central nervous system (CNS) involvement remains a rare and poorly understood entity. This chapter attempts to assemble the available information on CNS involvement in LCH in a comprehensive fashion, based up a literature review.
On light microscopy of typical LCH granulomas, Langerhans’ cells are characterized as large mononucleated cells with abundant eosinophilic cytoplasm and “coffee beans” nuclei.10 Electron microscopy reveals the diagnostic Birbeck bodies or granules, which are rigid tubular structures with an average diameter of 34 nm and have a tennis racket shape. They are present in the cytoplasm and are believed to arise from invagination of the cell membrane. The presence of Birbeck granules or immunohistochemical staining for CD1a positivity11 are required for the definitive diagnosis of LCH (Histiocyte Society). A recent report by Grois et al.12 represents a comprehensive study of the histopathology of CNS lesions in LCH. They presented material from 12 CNS biopsy specimens and one autopsy. These investigators identified three discrete pathologic lesions of LCH in the CNS. The first are well-circumscribed granulomas within the brain connective tissue space (meninges and choroids plexus). The second are granulomas within the brain connective tissues with partial infiltration of the surrounding parenchyma. The third are neurodegenerative lesions in the cerebellum and the brain stem. Histopathology of granulomas in the brain connective tissue showed them to be discretely separate from the CNS tissue. They consisted of histiocytes, foamy macrophages, lymphocytes, plasma cells, multinucleated giant cells and eosinophils. This is very similar to the composition of LCH granulomas in other organs, with the exception of the CD8+ T cells in the CNS lesions outnumbering the CD3+ /CD8− cells. The presence of CD1a+ cells was variable, being more abundant in newly diagnosed lesions when compared with material from a patient diagnosed 5 years earlier. Granulomas with partial infiltration into CNS tissue were found in patients with infundibular lesions infiltrating into the hypothalamus. This is the most common site of CNS involvement with LCH. Histopathology of the main lesion was the same as described above, however, the adjacent tissue was infiltrated by CD1a+ cells, CD68+ macrophages,
BIOLOGY CNS involvement has been recognized since the earliest descriptions of LCH.1,2 However, the prevalence of such involvement among patients with LCH is not known and the pathogenesis of CNS LCH remains a matter of speculation.3 The writing committee of the Histiocyte Society at its annual meeting in 19964 tried to explain the pathogenesis. They described a histochemical entity of neuronal injury and astrocytosis mediated by neuroexcitatory transmitters,5 glutamate, and N -methyl-D-aspartate, by a similar mechanism as described by Lipton and Gendelman6 to explain the mechanism of neuronal death in dementia associated with acquired immunodeficiency syndrome (AIDS). They reported that the human immunodeficiency virus (HIV) glycoprotein (gp) 120 coat stimulates brain macrophages and microglial cells causing them to secrete neurotoxins and cytokines. This leads to neuronal death through cytotoxic calcium influx. As LCH cells exhibit characteristics of activated macrophages in other organs, it is likely that similar mechanisms may play a pivotal role in the CNS as well. The local production of cytokines may recruit inflammatory cells into the brain, creating the inflammatory lesions seen within and surrounding LCH granulomas.7 On the other hand, some authors have also described autoantibodies to neuronal tissue in LCH.8,9 It is possible that LCH granulomas could harvest neuronal antigens and stimulate an autoimmune destruction of brain tissue. This would explain the neurodegenerative lesions that lack LCH cells.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
LANGERHANS’ CELL HISTIOCYTOSIS OF THE CENTRAL NERVOUS SYSTEM
and lymphocytes with nearly complete loss of neurons and axons. In addition, a much larger area of inflammation dominated by CD8+ lymphocytes surrounded these lesions. Finally, autopsy of a patient with neurodegenerative lesions demonstrated the full spectrum of this disease. There was diffuse inflammation in the brain dominated by CD8+ lymphocytes, as well as microglial activation and tissue degeneration. No focal granulomas or CD1a+ cells were found. Tissue degeneration was most profound in the cerebellum, brain stem, infundibulum and optic chiasm, and the basal ganglia. In the cerebellum, the deep white matter was atrophic with loss of neurons and myelin and reactive glial scar formation. The Virchow Robin spaces were enlarged and contained macrophages and T lymphocytes. Atrophy was also found in the cerebellar cortex. Other lesions throughout the brain showed similar histopathologic changes, but were less extensive.
CLINICAL FEATURES The clinical spectrum shows that all regions of the CNS can be involved in this disease. CNS disease, in order of frequency, involves the hypothalamic –pituitary axis, cerebellum, pons, and cerebral hemispheres. However, a small number of other cases involving the basal ganglia, spinal cord, and optic nerves and tracts have also been reported. Involvement of the choroid plexus has been reported in exceedingly rare cases. Signs and symptoms of CNS lesions can occur years before presentation of any other manifestations.13 Some patients have presented with acute signs of intracranial hypertension or with focal neurologic symptoms. Various signs and symptoms include, but are not limited to, ataxia, dysarthria, tremors, dysdiadochokinesis, and hyperreflexia.14 Eventually, the disease can progress, leading to marked neurologic disability ranging from spastic diplegia and tetraplegia, to discrete intellectual and behavioral changes. The hallmark of neuroendocrine manifestations of CNS LCH is diabetes insipidus.3 Panhypopituitarism, behavioral changes, and developmental delay have also been observed.15
611
• Type Ib: White matter lesions with enhancement. As with type Ia, there was a striking infratentorial location. The lesions presented as poorly to well-defined areas of prolonged T1 and T2 relaxation times, that is, low-signal intensity on T1-weighted images and high-signal intensity on T2-weighted and proton density weighted images, without mass effect and with strong Gd-DTPA enhancement. The differential diagnoses to be considered include multiple sclerosis, ADEM, metastases, and sarcoidosis. • Type IIa: Gray matter lesions without enhancement. Changes of this type were mostly seen in the dentate nucleus with bilateral involvement, as areas of low-signal intensity on T1-weighted images and high-signal intensity on T2-weighted and proton density weighted images. These lesions showed no mass effect. However, calcifications were occasionally present. These lesions must be distinguished from Fahr disease (if calcified), ADEM, glioma, and infarction. • Type IIb: Gray matter lesions with enhancement. These lesions were predominantly located in the cerebellar gray matter and basal ganglia (see Figure 1). Well-defined areas of low-signal intensity on T1-weighted images and high-signal intensity on T2-weighted and proton density weighted images showed mass effect and strong Gd-DTPA enhancement, and were sometimes surrounded by edema. Such lesions can be mistaken for low-grade glial tumors. • Type III : Extraparenchymal lesions. These masses were
RADIOGRAPHIC FEATURES Magnetic resonance imaging (MRI) remains the technique of choice for demonstrating CNS lesions. The writing committee of the Histiocyte Society, in 1996, proposed the following classification based on MRI to illustrate the spectrum of lesions. MRI and the pattern of gadolinium contrast enhancement suggest the correct diagnosis in any given clinical setting. • Type Ia: White matter lesions without enhancement. These lesions were predominantly located in the pons, cerebellar peduncles, and the cerebellar white matter. They presented as poorly defined areas of low-signal intensity on T1-weighted images and high-signal intensity on T2weighted and proton density weighted images. Multiple sclerosis, acute disseminated encephalomyelitis (ADEM), leucodystrophies, and infections have to be considered among alternate possible diagnoses.
Figure 1 Axial Contrast T1-weighted section through the basal ganglia shows several bilateral enhancing brain lesions, the largest at the heads of the caudate nuclei.
612
NEUROLOGICAL MALIGNANCIES
Figure 2 Coronal contrast T1-weighted image shows enhancing dural mass arising from edge of tentorium.
(i) dural based (see Figure 2), (ii) arachnoid based, or (iii) choroid plexus based, and appeared iso- to hypointense to brain on T1-weighted images, hypointense on T2weighted images, and showed uniform contrast enhancement. Alternate diagnoses for meningeal-based lesions include leukemia, lymphoma, and carcinomatous masses, and, for type IIIc lesions, choroid plexus tumors have to be considered. • Type IV : Lesions of the hypothalamic –pituitary axis. These lesions include (i) type IVa, infundibular thickening, (ii) type IVb, a partially or completely empty sella, and (iii) type IVc, lack of the posterior pituitary bright signals on T1-weighted MR images and hypothalamic mass lesions (see Figure 3). Alternate diagnoses of infundibular thickening include sarcoidosis, infundibuloma, posttraumatic states, and rare neoplasm; those for hypothalamic mass lesions include glioma, lymphoma, hamartoma, and sarcoidosis. • Type V : Atrophy. Local or diffuse atrophy was the nonspecific finding with a multifactorial etiology. • Type VI : Therapy-related white matter changes with/ without contrast enhancement and specific high-signal intensity changes. These can be seen bilaterally within the deep gray matter as foci of calcification. These changes are not entirely specific, but are characteristic sequelae of irradiation and chemotherapy damage to the CNS. They have to be considered in patients who are treated with these modalities for LCH at an early age.
Figure 3 Sagittal contrast T1-weighted images of the brain at midline (a) pretreatment and (b) posttreatment show enhancing hypothalamic lesion (a) that resolves on follow-up treatment (b).
TREATMENT There are no standard recommendations concerning the choice of treatment – the individual strategy is based on type, site, and extra-CNS involvement. Therapy in patients with types I and II MRI changes remains controversial. These patients may or may not have clinical signs of ataxia and neurologic dysfunction. Several agents have been tried in an effort to reverse or stop the progression of these changes without success. Therapy for mass lesions includes surgical resection or systemic chemotherapy. Endoscopic surgical techniques may be used in choroid plexus lesions. Chemotherapeutic agents with documented efficacy in systemic LCH have been employed to treat LCH in the CNS.16 Etoposide, cyclosporine A,17 vinblastine, methotrexate, chlorambucil, corticosteroids, and 2-chlorodeoxyadenosine (2CDA) have produced responses in individual cases. 2-CDA
LANGERHANS’ CELL HISTIOCYTOSIS OF THE CENTRAL NERVOUS SYSTEM
is a purine analog with potent toxicity to monocytes in vitro. As tissue histiocytes are derived from the same marrow precursors as monocytes, it has been employed to treat patients with LCH. This drug has shown encouraging results, and is especially useful in treating refractory LCH in patients without liver, spleen, lung, or hematologic involvement.18 – 20 Other therapeutic agents in combination are recommended if treatment with a single agent fails. Systemic chemotherapy is the treatment of choice for patients with hypothalamic/pituitary involvement. Ideally, the patient is diagnosed with LCH because of another lesion elsewhere that has been biopsied. However, if pituitary stalk thickening is the only manifestation, a biopsy is necessary to establish the diagnosis prior to the initiation of chemotherapy. External beam irradiation is not recommended for the treatment of any CNS lesions since the presence of CNS involvement predisposes patients to the development of neurodegenerative lesions that may be prevented if systemic chemotherapy is administered.21 In addition, irradiation will affect the more normal brain tissue and cause additional deficits.
PROGNOSIS The prognosis for patients with LCH depends on the degree of “risk organ” involvement. The Histiocyte Society defines “risk organs” as liver, spleen, lung, or hematologic involvement. In addition, early response to chemotherapy is a good prognostic indicator in patients with systemic disease. Patients who have a poor response have a 65% mortality rate when compared with 20% in responders. Recent reports show that patients with head and neck lesions are at increased risk of developing CNS involvement −30% will develop pituitary involvement.22 Once patients develop complete diabetes insipidus, it is not reversible despite systemic chemotherapy. In addition, 1 to 2% of patients with multiple organ involvement or maxillofacial lesions will develop irreversible neurodegenerative changes. To date, there is no known therapy to arrest or reverse the progression of this devastating disorder.
REFERENCES 1. Schuller A. Uber eigenartige Shandeldefekte im Jugendalter. Fortschr Rontgenstr 1915; 23: 12. 2. Christian HA. Defects in membranous bones, exophthalmos and diabetes insipidus. Med Clin North Am 1920; 3: 849.
613
3. Hamre M, et al. Langerhan’s cell histiocytosis: an explanatory epidemiology study of 177 cases. Med Pediatr Oncol 1997; 28: 92 – 7. 4. Report of Histiocyte Society workshops on ‘central nervous system disease in Langerhans cell histiocytosis’. Med Pediatr Oncol 1997; 29(2): 73 – 8. 5. Lipton SA, Rosenberg PA. Mechanism of disease: excitatory amino acids as a final common pathway for neurologic disorders. N Engl J Med 1994; 330: 613 – 22. 6. Lipton SA, Gendelman HE. Seminars in medicine of the Beth Israel Hospital, Boston: dementia associated with the acquired immunodeficiency syndrome. N Engl J Med 1995; 332: 934 – 40. 7. Egeler RM, et al. Differential in situ cytokine profiles of Langerhanslike cells and T cells in Langerhans cell histiocytosis: abundant expression of cytokines relevant to disease and treatment. Blood 1999; 94: 4195 – 201. 8. Greenlee JE, Brashear HR. Antibodies to cerebellar purkinje cells in patients with paraneoplastic cerebellar degeneration and ovarian carcinoma. Ann Neurol 1983; 14: 609 – 13. 9. Brashear HR, et al. Localization of antibody in the central nervous system of a patient with paraneoplastic encephalomyelonemitis. Neurology 1991; 41: 1583 – 7. 10. Favara BE, Jaffe R. Pathology of Langerhans cell histiocytosis. Hematol Oncol Clin North Am 1987; 1: 75 – 97. 11. Emile J, et al. Langerhans cell histiocytosis: definitive diagnosis in the use of monoclonal antibody 010 on routinely paraffin embedded samples. Am J Surg Pathol 1995; 19: 636 – 41. 12. Grois N, et al. Neuropathology of CNS disease in Langerhans cell histiocytosis. Brain 2005; 128(Pt 4): 829 – 38. 13. Grois N, et al. Central nervous system disease associated with Langerhan’s cell histiocytosis. Am J Pediatr Hematol Oncol 1993; 15: 245 – 54. 14. Vaquero J, et al. Posterior fossa xanthogranuloma. Case report. J Neurosurg 1979; 51: 718 – 22. 15. Hayward J, Packer R, Finlay J. Central nervous system and Langerhans’ cell histiocytosis. Med Pediatr Oncol 1990; 18: 325 – 8. 16. Ladisch S, Gadner H. Treatment of Langerhans cell histiocytosisevolution and current approaches. Br J Cancer Suppl 1994; 23(5): 541 – 6. 17. Colella R, et al. Cyclosporine therapy in recurrent Langerhans cell histiocytosis. Preliminary results from the A IEOP ICL-R93 study. Histiocyte Society, 10th Annual Meeting, Paris, France, September 17 – 19, 1994. 18. Wayne S, 2-Chlorodeoxyadenosine (2CDA) for refractory Histiocytosis Syndromes (HS): preliminary experience in pediatrics. Histiocyte Society, 10th Annual Meeting, Paris, France, September 17 – 19, 1994. 19. Saven A, Foon KS, Piro LD. 2 Chlorodeoxyadenosine induced complete remissions in Langerhans cell histiocytosis. Histiocyte Society, 10th Annual Meeting, Paris, France, September 17 – 19, 1994. 20. Dhall G, et al. Treatment of patients with Langerhans cell histiocytosis with central nervous system involvement with 2-chlorodeoxyadenosine. Ped Blood Cancer 2005; 45: 487. 21. Grois N, et al. Central nervous system disease in Langerhans cell histiocytosis. Br J Cancer 1994; 70: S24 – 8. 22. Donadieu J, et al. Endocrine involvement in pediatric-onset Langerhans cell histiocytosis: a population based study. JPEDS 2004; 144: 344 – 9.
Section 10 : Neurological Malignancies
56
Chordomas Herbert B. Newton
INTRODUCTION Chordomas are rare, histologically benign, but clinically aggressive tumors of the axial skeleton first described in 1856 by both Virchow and Luschka.1,2 The tumor discovered by Virchow was found within the clivus during a routine autopsy; prompting the theory that the tumor had arisen from cartilage. In 1858, Muller was the first to suggest that chordomas may originate from embryonic rests of the primitive notochord, the “chorda dorsalis”.3 The first description of a symptomatic chordoma was made in 1864 by Klebs, in a patient with a tumor of the spheno-occipital region.4 In 1894, Ribbert was the first to use the term “chordoma”, and further characterized Muller’s theory by producing experimental chordomas after releasing tissue of notochordal origin from the nucleus pulposus of rabbits.5,6 The tumors produced in these experiments were histologically similar to de novo chordomas. The experiments of Ribbert were replicated by Congdon in 1952 using a similar rabbit model.7 The modern theory of the origin of chordomas proposes that the tumors derive from embryonic rests of the primitive notochord that persist within the axial skeleton.8 – 11 The notochord forms from ectodermal cells during the third or fourth week of development and is believed to act as an embryonic organizer.12 During the fourth to sixth week of development, mesenchymal cells from adjacent sclerotomes envelop the notochord as they merge to form the spinal vertebral bodies.8,12 The notochord degenerates during this process and by the seventh week remains only between the vertebral bodies as the nucleus pulposus of the intervertebral discs. Pathological studies of tissue using both light and electron microscopic techniques have demonstrated similarities between chordoma and human intervertebral disc.8,12 It is postulated that incomplete degeneration of residual notochord may occur within the vertebral body at the junction of the adjacent sclerotomal regions. These incompletely degenerated rests can potentially undergo malignant transformation and develop into a chordoma. Investigations of the persistence and regression of the human notochord in fetuses of 4 to 18 weeks gestation suggest that there is great variation in
this process and that the presence of aberrant notochordal tissue is not uncommon.13 Furthermore, the topographical distribution of these heterotopic notochordal rests corresponds closely to the common sites of chordoma in the adult (i.e., sacrococcygeal, clivus).12,14 Autopsy studies reveal rests of presumed notochordal tissue anterior to the clivus and around the sacrum in up to 2% of cases.10,11 Finally, a shared immunophenotype is noted between notochordal and chordoma cells, with both types of cells containing S-100 protein, cytokeratins, and human epithelial polymorphic mucin.14
TUMOR EPIDEMIOLOGY Chordomas are rare neoplasms, representing only 0.1 to 0.2% of all intracranial tumors, 6.15% of all primitive skull base tumors, and 1 to 4% of primary malignant bone tumors.8 – 11,15 They can arise anywhere within the midline axial skeleton where the notochord existed (i.e., clivus, sellar and parasellar region, nasopharynx, foramen magnum, vertebrae, and sacrococcygeal region), but have a predilection for the sacrum and clivus. In adults, approximately 50% of chordomas arise in the sacrum, 35 to 40% within the base of skull and clivus, and 10 to 15% throughout the true vertebrae.8 – 11,15 – 17 When chordomas affect the vertebral column, more than half will occur in the lumbar region, 25 to 30% in the cervical vertebrae, and 10 to 15% in the thoracic spine.16 In children, chordomas most often involve the skull base.18,19 On rare occasions, chordomas can arise in extraosseous or off-the-midline sites such as the transverse process of a vertebra, skin, paranasal sinuses, sella turcica, hypothalamus, or foramen magnum.8 – 11,15,20 – 23 Chordomas can occur at any age, but are most common between the fourth and sixth decades of life.8 – 11,15 Although these tumors can arise in children, less than 5% of all cases develop before 20 years of age.18,19 There is a male predominance in some series, especially for tumors of the sacrum, with a ratio ranging from 2 : 1 to 3 : 1.8 – 11,15 In other series, especially chordomas of the skull base, the male to female frequency is equal.8
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
CHORDOMAS
615
PATHOLOGY Chordomas are generally slow growing, unencapsulated neoplasms that are locally invasive within bone and soft tissues.8 – 11,24 – 26 A pseudocapsule may be noted around tumors that grow into soft tissues or the dura mater. As the tumors enlarge, they often stretch cranial nerves and displace structures such as blood vessels and the brainstem. Grossly, the tumors are usually reddish or purple in color, with a nodular appearance to the surface. Internally, the mass is frequently gelatinous and soft; regions that contain cartilage or calcium are more firm. Foci of hemorrhage may be present and can be small or extensive. The size of the lesion can be quite variable, with sacral tumors often becoming very large. In one series of cranial base chordomas, average tumor volume was 58 cm3 .8 On microscopic examination, chordomas can be grouped into several different histological categories, including a typical pattern, a chondroid pattern, and tumors with features of malignant degeneration.8 – 11,24 – 30 The typical or classic pattern of chordoma (65–80% of all cases) is distinguished by a lobular arrangement, with the neoplastic cells disposed in solid sheets or irregular intersecting cords (see Figure 1). The sheets and cords of cells are set in a stroma that contains an abundant mucinous matrix. The individual cells are large, often with vacuolated eosinophilic cytoplasm, and contain variable amounts of mucin. The cell type considered diagnostic for chordomas is called physaliphorous (i.e., bubble-bearing). These cells are distinctively large and vacuolated, with eccentric nuclei (see Figure 1). Nuclei tend to be hyperchromatic, with prominent nucleoli, and rarely demonstrate atypia. Potentially aggressive features such as mitoses, necrosis, hypervascularity, and spindle cells (i.e., sarcomatous degeneration) are typically absent or rare.24 – 31 DNA ploidy analysis of typical chordomas demonstrates aneuploidy in 15 to 40% of cases.8 – 11,25 There is a trend for tumors with aneuploid DNA content to behave more aggressively and for patients with these tumors to have shorter survival.25 Other authors have not found a correlation between DNA ploidy status and tumor behavior, in terms of overall survival and tendency for local recurrence or distant metastases.32 Many authors contend that the chondroid pattern (15–30% of all cases) is a separate histological variant of chordoma, although this is controversial.8 – 11,24 – 31,33 The chondroid pattern has been associated with a more favorable prognosis, as originally described by Heffelfinger and colleagues.31 However, other authors contend that chondroid chordomas are a subgroup of low-grade chondrosarcoma and are not related to chordomas.34,35 By definition, chondroid chordomas contain regions of typical chordoma with physaliphorous cells, against a background of areas characterized by cartilaginous matrix that have stellate tumor cells occupying lacunar spaces (resembling chondrocytes; see Figure 2).31,33 As in typical chordoma, anaplastic or aggressive features such as mitoses, necrosis, hypervascularity, and spindle cells are typically absent or rare.24 – 31,33 Electron microscopic studies support
Figure 1 High-power view (600×) of classic or typical chordoma, demonstrating physaliphorous (bubble-bearing) cells. Note the large size, vacuolization, and eccentric nuclei. Several cells have hyperchromatic nuclei and prominent nucleoli. Mitoses, spindle cells, and regions of necrosis are absent.
Figure 2 Low-power view (100×) of chondroid chordoma, demonstrating an area of classic chordoma with physaliphorous cells in the upper portion of the field, with a chondroid region that demonstrates cartilaginous metaplasia in the lower portions of the field. No anaplastic or malignant features are present.
the dual nature (i.e., epithelial–mesenchymal) of these neoplasms by identifying cells with epithelial features and other cells consistent with chondrocytes.36 Chordomas with malignant degeneration (less than 5% of all nonirradiated cases) typically demonstrate sarcomatous features (i.e., spindle cells).8 – 11,37 – 40 These tumors will contain areas of classic chordoma admixed with regions characterized by the presence of atypical spindle cells. The spindle cell component demonstrates high cellularity, marked nuclear pleomorphism, and a high mitotic rate. Within the malignant spindle cell zones, regions of cartilaginous or osseous differentiation may be noted. In many tumors, a transitional zone may be present between the regions of typical chordoma and regions containing the malignant spindle cells. This transitional zone may contain an intermediate, stellate type of cell.
616
NEUROLOGICAL MALIGNANCIES
DNA ploidy analysis of chordomas with sarcomatous degeneration usually demonstrates aneuploid cell populations with a high proliferating fraction.37,40 The mean proliferating fraction (%S + G2 M) of a series of spindle cell chordomas was 34.1%.37 In comparison to the mean proliferating fraction of a series of typical chordomas (20.2%), the growth fraction of spindle cell chordomas was significantly larger (p < 0.01). In addition to the diagnostic information obtained from the histological evaluation of chordomas, immunohistochemical analysis may also be helpful in clarifying the pathologic differential diagnosis.8 – 11,28,33,34,37,40 – 45 Other tumor types to be considered are ependymoma, schwannoma, neurofibroma, metastasis (e.g., clear cell type), chondrosarcoma, and fibrous histiocytoma. The immunohistochemical profile of typical chordomas illustrates the dual epithelial–mesenchymal nature of these tumors and consists of frequent positivity to cytokeratin and epithelial membrane antigen (EMA) and less consistent staining for S100 protein and vimentin.8,12,33,37,42 Variable staining has also been noted with α1 -antichymotrypsin and tissue polypeptide antigen.41,42 Chondroid chordomas with a small cartilaginous component may have variable staining of S100, with preserved positivity to cytokeratin and EMA.33 When the cartilaginous component is more robust (20–50%), the staining within the chondroid regions for cytokeratin and EMA may become variable, with persistent positivity for S100.33 Chondrosarcomas stain consistently negative for cytokeratin and EMA since there is no epithelial component to these tumors. Chordomas with sarcomatous degeneration have an alteration of the immunohistochemical profile in the malignant regions containing spindle cells.37,38,40 The staining for vimentin becomes more prominent, while staining for cytokeratin and EMA is markedly decreased. In some tumors, staining for S100 may also be reduced.40 Other immunohistochemical studies of chordomas have evaluated the expression of cell adhesion molecules (CAMs), including E-, P-, and Ncadherin, CD44, β-catenin, intercellular CAM, neural CAM, and vascular CAM.46,47 Overall, chordomas were shown to frequently demonstrate expression of NCAM, VCAM-1, CD44, N-cadherin, and E-cadherin. The expression of Ecadherin, in particular, was very prominent and ubiquitous, and was noted in all histologic variants of chordoma. Preliminary data from studies using cytogenetic and molecular techniques are beginning to elucidate the mechanisms of transformation of chordomas.48 – 57 No specific or characteristic chromosomal abnormalities have been described thus far. Many of the cases have shown hypodiploidy or near-diploidy, which is in contrast to the DNA flow cytometry data. Several tumors have shown structural anomalies of chromosomes 1 and 21, while others have alterations (usually elongation) of the telomere.52,53 A recent loss of heterozygosity (LOH) analysis by Riva et al. suggests that the 1p36.13 region is abnormal in up to 85% of chordomas.55 Candidate tumor suppressor genes in this region with a potential role in oncogenesis include CASP9, EPH2A, and DVL1. LOH at 1p36.13 appeared to be an early event in the transformation process, since it was present in tumors of all grades and locations. Using immunohistochemical techniques, Matsuno et al. found that
p53 protein was present in chordomas that had a high proliferative index (as determined by MIB-1 staining) and in recurrent tumors.56 Furthermore, a significant correlation was noted between cyclin D1 staining and MIB-1 proliferative index or tumor recurrence. Interestingly, none of the tumors evaluated for bcl-2 were immunopositive, suggesting that apoptosis may not contribute to the recurrence of chordomas. Eisenberg et al. studied seven skull base chordomas and found that, in two very aggressive cases, there was loss of heterozygosity for the Rb tumor suppressor gene on chromosome 13, suggesting that alterations or loss of Rb may play a role in the transformation of chordomas.57 A recent study of skull base chordomas evaluated the expression of growth factors and structural proteins in good prognosis versus poor prognosis patient cohorts.58 The mean expression of transforming growth factor-α and basic fibroblast growth factor were elevated in the patients with a poor prognosis and more rapid tumor progression. A similar, but not as pronounced, difference was noted for fibronectin. Several authors have attempted to correlate pathological features of chordomas with prognosis.24,28,29 In a series of 48 mixed chordomas, Rich et al. were unable to detect a correlation between cellular pleomorphism, mitotic figures, or hyperchromatic nuclei with survival.24 The only histologic variable to correlate with survival was the presence of chondroid elements. Chondroid chordomas had a more indolent course, longer duration of symptoms, and increased survival. Forsyth et al. evaluated 51 intracranial chordomas and were unable to detect a correlation between mitosis, chondroid elements, and survival.28 In a series of 62 skull base chordomas, O’Connell et al. found that tumors with greater than 10% necrosis were associated with shorter patient survival.29 The presence of chondroid elements, mitoses, pleomorphism, nucleolar prominence, and vascular invasion were not correlated with overall survival. A molecular evaluation of a similar series of skull base tumors revealed that the expression of human telomerase reverse transcriptase (hTERT) messenger RNA was frequently associated with faster rates of tumor growth and an increased risk of recurrence.59 The expression of hTERT was also associated with the presence of mutated p53 protein and an increased doubling time for residual tumor following surgical resection. Naka et al. performed a clinicopathological comparison of skull base and non–skull base chordomas in a series of 122 patients.60 Skull base tumors were noted to have a higher MIB-1 labeling index than non–skull base tumors. The higher MIB-1 labeling index was often associated with older age, greater risk of recurrence, and nuclear pleomorphism. In contrast, for patients with non–skull base chordomas, only nuclear pleomorphism was noted to be a significant negative prognostic factor.
CLINICAL FEATURES AND PRESENTATION Approximately 50% of chordomas arise in the sacrum, 35 to 40% within the skull base and clivus, and 10 to 15% throughout the vertebral column.8 – 11,15 – 17 In general, chordomas are relatively slow growing and often have a prolonged duration of symptoms before diagnosis. The specific symptoms and
CHORDOMAS
neurological findings noted at presentation will vary according to the location of the tumor. Although these tumors are often benign histologically, systemic metastases have been noted in 10 to 40% of cases.8 – 11,61,62 The most frequent sites for metastases are the lungs, regional lymph nodes, liver, bone, and skin. Chordomas of the Sacrum
Chordomas represent the most common primary neoplasm of the sacrum.10,11,63 – 66 They often reach substantial size prior to diagnosis because of the ample room for tumor growth before critical structures are disturbed. The median age of patients in the majority of series is approximately 60 years; males are affected more often than females. The most common symptom (60–70% of patients) consists of persistent low back pain, which is slowly progressive and is often present for 12 to 18 months before diagnosis (see Table 1).10,11,63 – 66 Patients occasionally complain of more specific locations of the pain, such as the coccygeal, buttock, or anal regions. The pain may have a radicular component to it, with radiation down one of the legs. This presentation often leads to the erroneous diagnosis of nonspecific “sciatica”, delaying discovery of the tumor by many months. Rectal dysfunction consisting of alteration of bowel habits (i.e., constipation), tenesmus, or bleeding is common (approximately 40% of patients). As the tumor continues to enlarge, it usually grows ventrally and may encroach on the sacral foramina and nerve roots, causing neurologic dysfunction. Symptoms from sacral nerve root compression are variable and include perianal numbness, urinary hesitancy or retention, urinary incontinence, impotence, and rectal incontinence. The general physical is typically benign, except for the rectal examination, which often demonstrates a presacral mass.10,63 The neurological examination may be normal or show evidence of sacral root dysfunction (e.g., perianal numbness, loss of anal sphincter tone). Chordomas of the Skull Base, Clivus, and Intracranial Cavity
Chordomas comprise 6.15% of all skull base tumors and 0.1 to 0.2% of all intracranial tumors.8 – 11,15 They occur most often in the clivus, but can arise in other areas such Table 1 Symptoms and signs in patients with chordoma of the sacral region.
Symptoms and signs General low back pain Rectal dysfunction Constipation Sciatica Coccygeal pain Sacral pain Urinary incontinence Buttock pain Anal pain Perianal numbness Impotence Fecal incontinence Data compiled from references 10,11,63 – 65.
Percentage of patients 60 – 70 40 – 45 30 – 35 25 – 30 20 15 10 – 15 5–7 5–7 5 5 5
617
Table 2 Symptoms and signs in patients with chordoma of the skull base, clivus, and intracranial cavity.
Symptoms and signs
Percentage of patients
Diplopia Cranial nerve VI palsy Headache Cranial nerve IX, X, XI, XII palsy Cranial nerve II, III, IV, V1 , VII palsy Diplopia and headache Pyramidal tract dysfunction Facial pain Vertigo/tinnitus Dysphagia/hoarseness Alterations of vision Gait disturbance
50 – 90 45 – 75 25 – 60 25 – 40 15 – 25 15 – 20 15 – 20 12 – 15 12 – 15 12 – 15 12 – 15 12 – 15
Data compiled from references 8,9,24,28,67 – 69.
as the sphenoid sinus, cavernous sinus, occipital condyle, and sella.8 – 11,18,22,28,29,67 – 70 Depending on the primary site of tumor involvement and direction of growth (e.g., anterior, lateral, posterior), symptoms and signs may vary considerably. The mean age of patients with skull base chordomas is in the range of 38 to 45 years.64,65,67 In the majority of series, the most common symptoms are either diplopia or headache (see Table 2).8,9,24,28,67 – 70 Diplopia is the initial symptom in 50 to 90% of patients. The diplopia is usually horizontal and exacerbated by attempts at lateral gaze. Headache is noted at presentation in 25 to 60% of patients. In many patients, headache and diplopia develop simultaneously. Symptoms such as facial pain, vertigo, tinnitus, dysphagia, hoarseness, alterations of vision, and gait disturbance are present in 12 to 15% of patients.67 – 70 Infrequent complaints include hearing loss, dizziness, unilateral weakness, facial dysesthesias, and neck pain. On neurological examination, the most common findings are cranial nerve palsies (see Table 2).8,9,24,28,67 – 70 The VIth cranial nerve is involved most frequently, with abnormal function noted in 45 to 75% of patients. The deficit is usually unilateral, but can be bilateral in some cases. Abducens palsy can be associated with dysfunction of cranial nerves II, III, IV, V1 , and VII in 15 to 25% of patients.28,69,70 Although uncommon, isolated palsy of cranial nerves II, III, or IV is seen in some patients. Abnormalities of the lower cranial nerves (i.e., IX, X, XI, XII) are noted in 25 to 40% of patients.28,67 – 70 Similar to abducens palsy, the deficits are usually unilateral, but can be bilateral in some cases. Cranial nerves of the cerebellopontine angle (i.e., VII, VII) are rarely affected on examination at presentation, but can develop in patients with large tumors. Pyramidal tract dysfunction is present in 15 to 20% of patients and develops from tumors that compress the ventral surface of the brainstem.8,9,28 The findings may be unilateral or bilateral and in some cases are associated with ataxia. Furthermore, patients with brainstem compression by tumor may manifest inappropriate laughing or crying. The emotional lability is thought to occur from disturbance of ventral pontine tegmental pathways.9 Chordomas of the True Vertebrae
Chordomas are uncommon tumors of the vertebral column (usually the vertebral body), representing less than 5% of all
618
NEUROLOGICAL MALIGNANCIES
tumors in this region.16,17 Approximately 60% of vertebral chordomas arise in the lumbar region; 10 to 15% develop in the thoracic area, and 25 to 30% in the cervical spine. Ventral tumor growth will cause bone destruction and infiltration into paraspinal soft tissues, while dorsal expansion may cause nerve root displacement or spinal cord compression. The mean age of patients with spinal chordomas ranges from 45 to 50 years. Patients with these tumors often have a shorter duration of symptoms before diagnosis than patients with tumors of the sacrum, due to the smaller volume of bone in proximity to sensitive neural structures.66 In one series, the mean duration of symptoms prior to diagnosis was seven months.16 In the majority of cases (>90%), the initial symptom is localized pain in and around the involved vertebral body. There may be a radicular component to the pain from displacement or compression of nerve roots, with lancinating pain in a limb or anteriorly around the thorax. Other alterations of sensation, such as dysesthesias or sensory deficits, may occur. Cervical chordomas that grow ventrally and compress the esophagus may cause dysphagia.17 Occasionally, tumors can cause myelopathic weakness, gait ataxia, or sphincter dysfunction. Several authors have attempted to correlate various clinical parameters with overall survival and prognosis.28,29 Patients less than 40 years appear to have improved survival and a better prognosis. Forsyth et al. noted a significant difference in survival (5-year survival of 75 versus 30%) for patients less than 40 years (p < 0.0001).28 The presence of diplopia was also suggestive of a better prognosis and improved survival, especially when correlated with patient age.28 Female sex was associated with improved survival (median 158 months versus 86 months; p < 0.004) and a better prognosis in both
(a)
univariate and multivariate analyses by O’Connell et al.29 For patients with sacral and spinal chordomas, negative prognostic factors included larger tumor size, inadequate surgical margins, the presence of necrosis, a labeling index of greater than 5%, and local recurrence.66
RADIOLOGIC DIAGNOSIS Patients with a history and neurological examination suspicious for a chordoma of the skull base, vertebral column, or sacrum require a radiologic evaluation with either computed tomography (CT) or magnetic resonance imaging (MRI).8,71 – 80 CT and MRI are equivalent in their ability to delineate the presence of a tumor. Both modalities clearly demonstrate the mass within bone, bone erosion or destruction, and extension into soft tissues.71 – 74,78 – 80 Rarely, MRI may have trouble detecting small tumors confined within the margins of the clivus.72 On noncontrast CT, the tumor usually appears as a soft-tissue mass, isodense or hyperdense with neural tissues, causing destruction of the adjacent bone (see Figure 3). Bone-windowed CT scans demonstrate the precise amount of bone destruction caused by the tumor, with sharp margins (see Figure 3). Calcification is noted in 40 to 70% of chordomas (especially clival) with CT imaging. Small regions of sequestered bone can also be noted in approximately 15 to 20% of cases. MRI is inferior to CT in the ability to delineate the exact margins of bone destruction or the presence of calcification.71,72 With the administration of contrast, chordomas always demonstrate contrast enhancement. The amount of enhancement may vary, but is often quite dense and homogeneous. Sagittal and coronal reconstruction of CT images is sometimes helpful to better delineate the extent of skull base and sacral tumors. However, the ability of CT to
(b)
Figure 3 (a) Bone-windowed CT of the skull base demonstrating a large, eccentric chordoma of the clivus on the left side, which is causing severe erosion and destruction of surrounding bone. (b) Noncontrast CT of the skull base at the same level as figure 3(a), demonstrating the clival chordoma as a mass with soft-tissue density within the region of eroded bone.
CHORDOMAS
evaluate tumors in the sagittal and coronal planes is inferior to MRI. In general, MRI with sagittal, coronal, and axial sections clearly defines the margins of chordomas of the skull base, vertebral column, and sacrum.8,71 – 80 On T1-weighted images, 75% of tumors appear isointense, while 25% appear hypointense, compared to surrounding neural tissues (see Figure 4).8,72,80 With administration of gadolinium, chordomas usually enhance. As with CT, the degree of enhancement
(a)
619
is variable; in most tumors, the pattern is heterogeneous. On T2-weighted images, chordomas are always hyperintense to all surrounding structures. The pattern of hyperintensity is homogeneous in 20% of cases and heterogeneous in the remaining 80% (mild in 30%, marked in 50%).72 – 74 Tumor calcification, exact margins of eroded bone, and the presence of sequestered bone are noted in some tumors, but these are not demonstrated as well as with CT. However, the multiplanar capability of MRI allows for better visualization of
(b)
(c) Figure 4 (a) Noncontrast MRI (TR 7500, TE 102) of the brain, at the level of the upper clivus and orbits, demonstrating an isointense chordoma within the clivus that extends into the sphenoid sinus region and tracks along both optic nerves. (b) Gadolinium-enhanced MRI (TR 600, TE 23) of the brain, at the same level as figure 4(a), demonstrating heterogeneous enhancement of the chordoma within the clivus and sphenoid sinus. (c) Midsagittal, noncontrast MRI (TR 500, TE 16) of the brain, demonstrating the chordoma anterior to the brainstem within the clivus and extending into the sphenoid sinus region. The signal intensity of the mass is heterogeneous: hyperintense within the bone of the clivus and more isointense in the sinus.
620
NEUROLOGICAL MALIGNANCIES
the extent of tumor margins and infiltration into soft tissues than is possible on CT. Sagittal MRI clearly shows the anterior –posterior margins of tumor involvement. For chordomas of the skull base, MRI is helpful for determining the anterior extension of tumor into the sinuses or nasopharynx and posterior extension toward the brainstem.73,74 The posterior fossa is affected by tumor (e.g., brainstem compression, cranial nerve displacement) in approximately 80% of skull base tumors.74 Sagittal MRI is essential for tumors of the sacrum to determine the extent of the lesion anteriorly into the rectum and other soft tissues.78 Coronal MRI images are useful for assessing lateral extension of skull base tumors toward the cavernous sinuses. The cavernous sinuses are infiltrated by tumor in approximately 65% of skull base chordomas.74 Furthermore, with sacral chordomas, coronal MRI can determine involvement of sacral nerve roots within the neural foramina. MRI is far superior to CT in demonstrating the relationship of tumor to cranial nerves and vascular structures.71 – 74,80 The carotid and basilar arteries are delineated clearly on T2-weighted images because of the contrast between the flow void inside the vessels and surrounding high-signal tumor. Meyer et al. noted displacement of the carotid or basilar arteries in 57% of skull base chordomas.74 Furthermore, in 36% of their cohort, vascular encasement was present. Encasement of vessels by chordomas has also been reported by other authors.71 – 74,80 Universally, the lumen of encased vessels is not compromised by tumor, and they continue to have normal blood flow. Most authors do not report any MRI signal characteristics that can be used to differentiate between typical chordomas and those with chondroid regions.74 However, Sze et al. noted that chondroid chordomas were less intense on T2-weighted images than were conventional chordomas.72 Assessment of mean quantitative T1 and T2 relaxation values (msec) has shown that chondroid tumors often have shorter times than conventional chordomas. The use of MR angiography can further delineate the presence of arterial vascular encasement and luminal narrowing.80 In addition, MR venography can also clearly demonstrate any venous involvement by tumor (e.g., narrowing, occlusion).
TREATMENT The treatment of many chordomas is limited by the invasive and infiltrative nature of these tumors. The tumor is often too extensive at diagnosis for a complete, curative resection.8,65,68 Even when the lesion is small and radical surgery is attempted, local recurrence rates remain high (i.e., 50 to 100%). Therefore, the therapeutic approach for chordomas is primarily to maintain local control and minimize regional damage to neural structures.8,65,68 Despite the emphasis on local disease, systemic metastases can occur and are noted in 10 to 30% of patients.8,62,63 The most common sites for metastases are the lungs, regional lymph nodes, liver, bone, and skin. Infrequent sites include cardiac muscle, brain, adrenal glands, pancreas, pituitary gland, and eyelid.62,63 In the majority of patients, recurrence at the local site is most likely to affect morbidity and survival.
Surgical Resection
Most authors agree that surgical resection is an important aspect of the initial treatment of patients with chordoma.8,10,11,28,63,68,70,80 – 84 The most aggressive resection possible should be attempted after initial diagnosis, depending on the location (e.g., clivus, vertebral body, sacrum) and the extent of the tumor. It appears that the aggressiveness of resection has a critical impact on local control rates and may correlate with overall survival.8,28,68,70 In a review of 51 patients with intracranial chordomas, Forsyth et al. found that the extent of resection affected survival.28 In a univariate analysis, the extent of resection was significantly (p = 0.02) associated with survival. For patients receiving only biopsy, the 5- and 10-year survival rates were 36% and 0%, respectively. In the cohort of patients undergoing subtotal resections, the 5- and 10-year survival rates were 55% and 45%, respectively. This effect of resection on survival was most apparent in younger patients. Chordomas of the skull base and clivus can be grouped according to their size and extension into contiguous areas.68 Type I tumors are small and restricted to one compartment of the skull base (e.g., clivus or sphenoid sinus). Type II tumors are larger and extend to two or more contiguous areas of the skull base. Type III lesions are very extensive and involve several contiguous compartments of the skull base (e.g., clivus, sphenoid sinus, and middle fossa). Most series Type I chordomas are rare and usually amenable to radical resection using a single skull base approach.8,68 Type II tumors are most common (50 to 65%) and in many cases can also be radically resected using a single skull base procedure.68 The type III chordomas develop in 10 to 20% of patients and require two or more surgical procedures to attempt radical removal. There are numerous surgical approaches and techniques available for resection of skull base and clivus chordomas.8,68,70,81,82,85 – 91 The approach will depend on the location of the tumor and the degree of extension from the primary site. Most often, tumors are centered within the lower, middle, or upper clivus and extend into the cavernous sinus or petrous apex.68,81,82 The four most common approaches allow for an extensive resection of tumor either extradurally or intradurally. The subtemporal, transcavernous, transpetrous apex approach is used most often (30 to 35%) and provides access to the clivus, cavernous sinus, sella turcica, and petrous apex.8,68,90 The extended frontal approach is used in 25 to 30% of patients and is advantageous for tumors with extension into the orbits, ethmoid sinus, and anterior skull base.8,81 The subtemporal–infratemporal approach is utilized in approximately 20% of patients and offers excellent exposure of the middle fossa, clivus, and lateral skull base.8,81 For chordomas of the lower clivus, temporal bone, and occiput, the extreme lateral transcondyle and transjugular approach is used (approximately 15% of patients).8,81 Uncommon surgical approaches include the transoral, transmaxillary, transcervical–transclival, anterior cervical, and transsphenoidal procedures.8,70,82,85 – 91 Most authors would agree that the surgical approach is less important than the expertise of the
CHORDOMAS
surgical team and its ability to perform an extensive resection of the tumor.70 Studies using the most advanced skull base approaches for removal of chordomas report various results. Radical or total resection of tumor is achieved in 43.5 to 55% of patients, near-total or subtotal resection is noted in 40 to 47% of patients, and partial resection is attained in 8 to 10% of patients.8,68,81,82,70 In a series of patients with chordomas and chondrosarcomas involving the skull base and cavernous sinus, after a median follow-up of 24 months, Lanzino et al. noted three recurrences in 14 patients with subtotal or partial removal of tumor.81 No recurrences were observed in the group of patients that had undergone radical resections. In a study of skull base chordomas by Gayet al., there was a statistically significant difference (p < 0.05) between the risk of recurrence in patients with radical or near-total resections and patients with subtotal or partial resections.82 The overall recurrence-free survival estimates were 80% at three years and 76% at five years. In contrast, the survival estimates for patients that had recurrence of disease were 52% at two years and 26% at three years. Previous surgery or radiation therapy was associated with an increased risk of recurrence and surgical complications. Similar survival rates are described by Crockard et al. in a series of 42 patients with skull base chordomas, with 5- and 10-year rates of 77% and 69%, respectively.70 Overall, an aggressive surgical resection at the time of tumor diagnosis, regardless of the approach and technique used, has the most impact on subsequent local tumor control and survival. Sacral chordomas are often very large at diagnosis; however, most authors advise radical resection whenever possible.10,11,63,84,92 Similar to the experience with skull base tumors, recurrence-free survival is improved after radical or near-total resection. For tumors of the lower sacrum and coccygeal region, many authors recommend a posterior approach.11,84 Other investigators argue that a combined anterior –posterior approach is preferable.63 Tumors of the upper sacrum are resected most efficiently with a staged, combined anterior –posterior approach.10,11,63,84,92 Regardless of the approach used to resect the tumor, it is important to attempt preservation of the upper sacral nerve roots and the pudendal nerve. If the bilateral S-2 nerve roots are sectioned during surgery, urogenital and rectal function will be lost or impaired. If both S-2 nerve roots are preserved, 50% of patients will retain at least partial bladder and bowel control.83,91 To maintain normal bowel continence, preservation of at least one set of ipsilateral S-1, S-2, and S-3 nerve roots is recommended. The local recurrence rates are approximately 25 to 30% for tumors removed en bloc by radical resection.10,11,63 If the tumor is removed by subtotal or partial resection, local recurrence rates increase to approximately 60 to 65%. It is also recommended that chordomas of the true vertebrae be radically resected whenever feasible.11,17,83 For tumors of the cervical vertebrae, most authors recommend an anterior approach to perform a corporectomy, followed by bone grafting, if necessary.16 Thoracic tumors are best approached by thoracotomy or a staged procedure that combines a laminectomy and thoracotomy.11 Lumbar chordomas
621
will usually require an anterior approach; on occasion, a posteriolateral approach may be necessary.11 Radiation Therapy
Although radical resection is considered in each patient with a chordoma of the skull base, vertebrae, or sacrum, it is often impossible due to the invasive nature of these tumors. Therefore, radiation therapy to eradicate residual or recurrent disease is an important therapeutic consideration in many patients.8,10,11,28,65 Unfortunately, chordomas have proved to be relatively radioresistant tumors. The clinical results in most radiation therapy trials of chordomas have demonstrated only modest improvements in local control, recurrence-free survival, and overall survival.8,10,11,28,65,93 – 99 Early reports in the radiation oncology literature using photon-based megavoltage therapy suggested a dose-response relationship for chordoma.98,99 It was recommended that patients received at least 6000 to 7000 cGy to the tumor bed for optimal response. However, more recent studies have been unable to document a consistent dose-response relationship for chordoma using conventional photon techniques.8,99 – 101 In the reports by Cummings et al. and Saxton, doses of 2500 to 7000 cGy were used for patients with chordomas of various sites after surgical resection.100,101 Palliation of symptoms and improvement of relapse-free survival was as likely to occur with doses of 4000 to 5500 cGy as with higher doses. In an extensive review of reported dose-response data for photon techniques in treatment of cranial chordoma, Tai et al. concluded that no dose-response relationship was evident.99 Administration of doses in the range of 4500 to 5500 cGy were as effective as higher doses. In addition, the authors state that surgical resection in combination with irradiation significantly prolongs survival when compared to either modality used alone. In a study of 21 patients with chordomas of various sites, Keisch et al. concluded that irradiation prolonged the time to first relapse for tumors of the lower spine and sacrum, but not for tumors of the skull base.96 The overall 5- and 10-year actuarial survival rates were 74% and 46%, respectively. Forsyth et al. evaluated the results of 51 patients with intracranial chordomas and determined that conventional irradiation did not affect the overall survival of the cohort, but did prolong disease-free survival, especially in younger patients.28 The 5- and 10year disease-free survival in irradiated patients was 39% and 31%, respectively. A similar improvement of progressionfree survival, without a change in overall survival, has been reported by Thieblemont et al. in 26 patients with chordomas of various sites.98 Newer radiation therapy techniques currently being applied to patients with chordomas include intensity-modulated radiation therapy (IMRT) and conformal techniques. Several case reports have recently appeared using IMRT for chordomas of the sacrum and spine.102,103 IMRT was well tolerated and appeared to offer a more homogeneous radiation distribution pattern within the planning target volume. Conformal radiotherapy approaches are also being reported and have similar radiation distribution patterns to IMRT.104 Irradiation of chordomas with charged particles (i.e., protons, carbon, helium, neon) has shown promise as a
622
NEUROLOGICAL MALIGNANCIES
more efficacious therapeutic option.8,65,93 – 95,99,105 There are several radiobiological advantages of charged particles over photons. The high linear energy transfer (LET) of charged particles allows for a more defined and superior dose distribution (i.e., steeper fall-off in dose). Higher doses can be prescribed to the tumor volume with minimal risk of augmented toxicity to surrounding normal structures. AustinSeymour et al. have used fractionated proton irradiation for skull base chordomas and chondrosarcomas, administering a mean total dose of 69 GyE (Gray equivalent).94,95,99 The 5- and 10-year local control rates were 82% and 58%, respectively, while the 5- and 10-year disease-free survival rates were 76% and 53%, respectively. The median time to local failure was 53 months. In their opinion, these results represent a significant improvement over the results of conventional radiation techniques. A similar study in a series of 34 patients with skull base chordomas used a combination of high-energy photons and protons, in a two-thirds to onethird ratio of the total dose, respectively.106 The 3-year local control rate was 83.1%, with a 3-year overall survival rate of 91%. In a study using helium and neon particles, Berson et al. treated 25 patients with chordomas of the skull base and cervical spine and reported a 5-year local control rate of 55%.93,99 In a review of 14 patients with sacral chordomas treated with charged helium and neon particles, Schoenthaler et al. reported a 5-year local control rate of 55%.65 The 5and 10-year survival rates were 85% and 22%, respectively, with an overall median survival of 77 months. Several authors have reported their experience with carbon ion radiotherapy in chordomas.107,108 Schulz-Ertner et al. evaluated a series of 24 patients with skull base chordoma, using a median tumor dose of 60 GyE.107 With a mean follow-up of 13 months, the 2-year local control and progression-free survival rates were 83% and 83%, respectively. A similar study in 30 patients with unresectable sacral chordomas used carbon ion radiotherapy at a median dose of 70.4 GyE.108 At a median follow-up of 30 months, the 5-year local control and overall survival rates were 96% and 52%, respectively. Other methods of irradiation of chordoma include brachytherapy with radioactive seeds (e.g., iodine-125) and radiosurgery.8,94,109 – 111 It is difficult to evaluate the efficacy of these modalities because of the small number of patients that have been treated and the limited follow-up intervals reported.8,94,111 There may be a role for brachytherapy and radiosurgery in the palliation of residual and recurrent disease in carefully selected patients. One recent approach that has been applied to a small group of patients with skull base chordomas, involves the combination of maximal tumor resection and γ knife radiosurgery, with a mean treatment dose of 17 Gy.111 Although follow-up was limited, the local tumor control rate of 93.3% and mean tumor-free survival of 17 months were encouraging. Chemotherapy
The role of chemotherapy in the treatment of chordomas remains limited.8,11,83,98,112,113 The main indication for chemotherapy has been in patients with recurrent or widespread disease not amenable to further surgery or radiation therapy. In most cases, the regimens have been
designed to resemble protocols used for soft-tissue sarcomas. Unfortunately, very few patients have responded to this approach. Chemotherapeutic agents that have been used (as single agents or in combination) without success include methotrexate, vincristine, cisplatin, doxorubicin, etoposide, actinomycin D, and cyclophosphamide.11,83,98,112 Fleming et al. report two patients with malignant sacral chordomas and lung metastases in whom chemotherapy produced objective responses.112 One patient responded to a multiagent regimen consisting of etoposide, cisplatin, vincristine, dacarbazine, cyclophosphamide, and doxorubicin administered intravenously over three days every 4 to 5 weeks. The second patient responded to this multiagent regimen for five cycles and then progressed. A second regimen of single-agent continuous infusion ifosfamide was initiated, which produced dramatic shrinkage of the lung lesion. A more recent report has evaluated the efficacy of imatinib mesylate, a tyrosine kinase inhibitor with activity against cKIT, BCR-ABL, and platelet-derived growth factor receptors (PDGFR) that has been previously utilized for gastrointestinal stromal tumors (GIST) and high-grade gliomas.113,114 Six patients with advanced chordoma (5 sacral, 1 skull base) were treated with imatinib mesylate (800 mg/day). Pathologically, all of the tumors were noted to be positive for expression of PDGFR. In addition, four of the tumors were positive for phosphorylated PDGFR. Several of the patients had evidence of liquefaction on CT and MRI scans, as well as reduced glucose uptake on positron-emission tomography scans. Four of five symptomatic patients also reported subjective improvement, usually early in the course of treatment. A phase II trial with a larger cohort of patients has been initiated and remains open to further accrual.
CONCLUSIONS Chordomas are rare, locally invasive tumors that arise from embryonic rests of the primitive notochord, with the potential to develop anywhere along the midline of the axial skeleton. Patients with chordomas of the skull base, true vertebrae, or sacrum in whom a gross total resection has been performed can be followed closely with serial contrastenhanced MR imaging, without further treatment. For the more typical patient, with significant residual disease after surgery, some form of irradiation should be considered. The most effective radiotherapy techniques use charged particles (e.g., protons, carbon), which allow for higher doses to the tumor bed in a more well-defined dose distribution. However, access to facilities which can offer charged particle radiotherapy remain somewhat limited. If charged particle radiation therapy is not available, aggressive photon-based irradiation is still strongly recommended. Chemotherapy has a limited role in the treatment of chordoma, but should be considered for patients with recurrent or progressive disease that is refractory to further surgical intervention or irradiation.
CHORDOMAS
ACKNOWLEDGMENTS Dr. Newton was supported in part by grant P30CA16058, National Cancer Institute, Bethesda, MD, and the Esther Dardinger Neuro-Oncology Center Endowment Fund. The author would like to thank Ray Chaudhury, M.D. for assistance with the neuropathological materials and Ryan Smith for research assistance.
REFERENCES 1. Virchow R. Untersuchungen Uber die Entwickelung des Schadelgrundes. Berlin, Germany: George Reimer, 1857. 2. Luschka H. Die altersveranderungen der zwischen-wirbelknorpel. Virchows Arch Pathol Anat Physiol Klin Med 1864; 31: 396 – 9. 3. Muller H. Ueber das vorkommen von resten der chorda dorsalis bei menschen nach der geburt und uber ihr verhaltnis zu den gallertgeschwulsten am clivus. Ztschr Rat Med 1858; 2: 202. 4. Klebs E. Ein fall von ecchondrosis spheno-occipitalis amylacea. Virchows Arch Pathol Anat 1864; 31: 396. 5. Ribbert H. Ueber die echondrosis physalifora spheno-occipitalis. Centralbl Allg Pathol Pathol Anat 1894; 5: 457. 6. Ribbert H. Ueber die experimentelle erzeugung einer ecchondrosis physalifora. Verh Dtsch Kong Inn Med 1895; 13: 455. 7. Congdon CC. Proliferative lesions resembling chordoma following puncture of nucleus pulposus in rabbits. J Natl Cancer Inst 1952; 12: 893. 8. Gay E, Sekhar LN, Wright DC. Chordomas and chondrosarcomas of the cranial base. In Kaye H, Laws ER (eds) Brain Tumors. An Encyclopedic Approach. New York: Churchill Livingstone, 1995: Chap. 40: 777 – 794. 9. Miller NR. Secondary tumors of the central nervous system. In Miller NR (eds) Walsh and Hoyt’s Clinical Neuro-Ophthalmology, 4th ed. Baltimore, MD: Williams & Wilkins Company, 1988: Vol. 3, Chap. 52: 1662 – 1709. 10. Sundaresan N. Chordomas. Clin Orthop Relat Res 1986; 204: 135 – 42. 11. Healey JH, Lane JM. Chordoma: a critical review of diagnosis and treatment. Orthop Clin North Am 1989; 20: 417 – 26. 12. Salisbury JR. The pathology of the human notochord. J Pathol 1993; 171: 253 – 5. 13. Horwitz T. Chordal ectopia and its possible relation to chordoma. Arch Pathol 1941; 31: 354 – 62. 14. Salisbury JR, et al. Three-dimensional reconstruction of human embryonic notochords: clue to the pathogenesis of chordoma. J Pathol 1993; 171: 59 – 62. 15. McMaster ML, et al. Chordoma: incidence and survival patterns in the United States, 1973 – 1995. Cancer Causes Control 2001; 12: 1 – 11. 16. Klekamp J, Samii M. Spinal chordomas – results of treatment over a 17-year period. Acta Neurochir 1996; 138: 514 – 9. 17. D’Haen B, et al. Chordoma of the lower cervical spine. Clin Neurol Neurosurg 1995; 97: 245 – 8. 18. Yadav YR, et al. Cranial chordoma in the first decade. Clin Neurol Neurosurg 1992; 94: 241 – 6. 19. Coffin CM, et al. Chordoma in childhood and adolescence. A clinicopathologic analysis of 12 cases. Arch Pathol Lab Med 1993; 117: 927 – 33. 20. Katayama Y, et al. Intradural extraosseous chordoma of the foramen magnum region. Case report. J Neurosurg 1991; 75: 976 – 9. 21. Commins D, et al. Hypothalamic chordoma. Case report. J Neurosurg 1994; 81: 130 – 2. 22. Elias Z, Powers LK. Intrasellar chordoma and hyperprolactinemia. Surg Neurol 1985; 23: 173 – 6. 23. Kakuno Y, et al. Chordoma in the sella turcica. Neurol Med Chir 2002; 42: 305 – 8. 24. Rich TA, et al. Clinical and pathologic review of 48 cases of chordoma. Cancer 1985; 56: 182 – 7. 25. Schoedel KE, et al. Chordomas: pathological features; ploidy and silver nucleolar organizing region analysis. A study of 36 cases. Acta Neuropathol 1995; 89: 139 – 43.
623
26. Crapanzano JP, et al. Chordoma. A cytologic study with histologic and radiologic correlation. Cancer 2001; 93: 40 – 51. 27. Chetty R, Levin CV, Kalan MR. Chordoma: a 20-year clinicopathologic review of the experience at Groote Schuur Hospital, Cape Town. J Surg Oncol 1991; 46: 261 – 4. 28. Forsyth PA, et al. Intracranial chordomas: a clinicopathological and prognostic study of 51 cases. J Neurosurg 1993; 78: 741 – 7. 29. O’Connell JX, et al. Base of skull chordoma. A correlative study of histologic and clinical features of 62 cases. Cancer 1994; 74: 2261 – 7. 30. Kay PA, et al. Chordoma. Cytomorphologic findings in 14 cases diagnosed by fine needle aspiration. Acta Cytol 2003; 47: 202 – 20. 31. Heffelfinger MJ, et al. Chordomas and cartilaginous tumours at the skull base. Cancer 1973; 32: 410 – 20. 32. Berven S, et al. Clinical outcome in chordoma. Utility of flow cytometry in DNA determination. Spine 2002; 27: 374 – 9. 33. Ishida T, Dorfman HD. Chondroid chordoma versus low-grade chondrosarcoma of the base of the skull: can immunohistochemistry resolve the controversy? J Neurooncol 1994; 18: 199 – 206. 34. Bottles K, Beckstead JH. Enzyme histochemical characterization of chordomas. Am J Surg Pathol 1984; 8: 443 – 7. 35. Brooks JJ, LiVolsi VA, Trojanowski JQ. Does chondroid chordoma exist? Acta Neuropathol 1987; 72: 229 – 35. 36. Valderrama E, et al. Chondroid chordoma: electron-microscopic study of two cases. Am J Surg Pathol 1983; 7: 625 – 32. 37. Hruban RH, et al. Chordomas with malignant spindle cell components. A DNA flow cytometric and immunohistochemical study with histogenetic implications. Am J Pathol 1990; 137: 435 – 47. 38. Hruban RH, et al. Lumbo-sacral chordoma with high-grade malignant cartilaginous and spindle cell components. Am J Surg Pathol 1990; 14: 384 – 9. 39. Fukuda T, et al. Sacrococcygeal chordoma with a malignant spindle cell component. A report of two autopsy cases with a review of the literature. Acta Pathol Jpn 1992; 42: 448 – 53. 40. Tomlinson FH, et al. Sarcomatous transformation in cranial chordoma. Neurosurgery 1992; 31: 13 – 8. 41. Burger PC, Makek M, Kleihues P. Tissue polypeptide antigen staining of the chordoma and notochordal remnants. Acta Neuropathol 1986; 70: 269 – 72. 42. Meis JM, Giraldo AA. Chordoma. An immunohistochemical study of 20 cases. Arch Pathol Lab Med 1988; 112: 553 – 6. 43. Bouropoulou V, et al. Immunohistochemical investigation of chordomas: histogenetic and differential diagnostic aspects. Curr Top Pathol 1989; 80: 183 – 203. 44. Maiorano E, et al. Expression of intermediate filaments in chordomas. An immunocytochemical study of five cases. Pathol Res Pract 1992; 188: 901 – 7. 45. Plate KH, Bittenger A. Value of immunocytochemistry in aspiration cytology of sacrococcygeal chordoma. A report of two cases. Acta Cytol 1992; 36: 87 – 90. 46. Mori K, et al. Expression of E-cadherin in chordomas: diagnostic marker and possible role in tumor cell affinity. Virchows Arch 2002; 440: 123 – 7. 47. Hiroguchi H, et al. Expression of cell adhesion molecules in chordomas: an immunohistochemical study of 16 cases. Acta Neuropathol 2004; 107: 91 – 6. 48. Persons DL, Bridge JA, Neff JR. Cytogenetic analysis of two sacral chordomas. Cancer Genet Cytogenet 1991; 56: 197 – 201. 49. Gibas Z, Miettinen M, Sandberg AA. Chromosomal abnormalities in two chordomas. Cancer Genet Cytogenet 1992; 58: 169 – 73. 50. DeBoer JM, Neff JR, Bridge JA. Cytogenetics of sacral chordoma. Cancer Genet Cytogenet 1992; 64: 95 – 6. 51. Mertens F, et al. Clonal chromosome aberrations in three sacral chordomas. Cancer Genet Cytogenet 1994; 73: 147 – 51. 52. Bridge JA, Pickering D, Neff JR. Cytogenetic and molecular cytogenetic analysis of sacral chordoma. Cancer Genet Cytogenet 1994; 75: 23 – 5. 53. Butler MG, et al. Cytogenetic, telomere, and telomerase studies in five surgically managed lumbosacral chordomas. Cancer Genet Cytogenet 1995; 85: 51 – 7. 54. Gil Z, et al. Cytogenetic analysis of three variants of clival chordoma. Cancer Genet Cytogenet 2004; 154: 124 – 30.
624
NEUROLOGICAL MALIGNANCIES
55. Riva P, et al. Mapping of candidate region for chordoma development to 1p36.13 by LOH analysis. Int J Cancer 2003; 107: 493 – 7. 56. Matsuno A, et al. Immunohistochemical examination of proliferative potentials and the expression of cell cycle-related proteins of intracranial chordomas. Hum Pathol 1997; 28: 714 – 9. 57. Eisenberg MB, et al. Loss of heterozygosity in the retinoblastoma tumor suppressor gene in skull base chordomas and chondrosarcomas. Surg Neurol 1997; 47: 156 – 61. 58. Deniz ML, et al. Expression of growth factors and structural proteins in chordomas: basic fibroblast growth factor, transforming growth factor α, and fibronectin are correlated with recurrence. Neurosurgery 2002; 51: 753 – 60. 59. Pallini R, et al. Chordoma of the skull base: predictors of tumor recurrence. J Neurosurg 2003; 98: 812 – 22. 60. Naka T, et al. Skull base and nonskull base chordomas. Clinicopathologic and immunohistochemical study with special reference to nuclear pleomorphism and proliferative activity. Cancer 2003; 98: 1934 – 41. 61. Ogi H, et al. Cutaneous metastasis of CNS chordoma. Am J Dermatol 1995; 17: 599 – 602. 62. Hall WA, Clark HB. Sacrococcygeal chordoma metastatic to the brain with review of the literature. J Neurooncol 1995; 25: 155 – 9. 63. Bethke KP, Neifeld JP, Lawrence W. Diagnosis and management of sacrococcygeal chordoma. J Surg Oncol 1991; 48: 232 – 8. 64. Lybeert MLM, Meerwaldt JH. Chordoma. Report on treatment results in eighteen cases. Acta Radiol Oncol 1986; 25: 41 – 3. 65. Schoenthaler R, et al. Charged particle irradiation of sacral chordomas. Int J Radiat Oncol Biol Phys 1993; 26: 291 – 8. 66. Bergh P, et al. Prognostic factors in chordoma of the sacrum and mobile spine. A study of 39 patients. Cancer 2000; 88: 2122 – 34. 67. Watkins L, et al. Skull base chordomas: a review of 38 patients, 1958 – 1988. Brit J Neurosurg 1993; 7: 241 – 8. 68. Al-Mefty O, Borba LAB. Skull base chordomas: a management challenge. J Neurosurg 1997; 86: 182 – 289. 69. Volpe NJ, et al. Neuro-ophthalmologic findings in chordoma and chondrosarcoma of the skull base. Am J Ophthal 1993; 115: 97 – 104. 70. Crockard HA, et al. A multidisciplinary team approach to skull base chordomas. J Neurosurg 2001; 95: 175 – 83. 71. Oot RF, et al. The role of MR and CT in evaluating clival chordomas and chondrosarcomas. AJR Am J Roentgenol 1988; 151: 567 – 75. 72. Sze G, et al. Chordomas: MR imaging. Radiol 1988; 166: 187 – 91. 73. Larson TC, Houser OW, Laws ER. Imaging of cranial chordomas. Mayo Clin Proc 1987; 62: 886 – 93. 74. Meyers SP, et al. Chordomas of the skull base: MR features. AJNR Am J Neuroradiol 1992; 13: 1627 – 36. 75. Leproux F, et al. MRI of cranial chordomas: the value of gadolinium. Neuroradiol 1993; 35: 543 – 5. 76. Kumar AJ, et al. Imaging features of skull base tumors. Neuroimag Clin North Am 1993; 3: 715 – 34. 77. Ikushima I, et al. Chordomas of the skull base: dynamic MRI. J Comp Ass Tomogr 1996; 20: 547 – 50. 78. Rosenthal DI, et al. Sacrococcygeal chordoma: magnetic resonance imaging and computed tomography. AJR Am J Roentgenol 1985; 145: 143 – 7. 79. de Bru¨ıne FT, Kroon HM. Spinal chordoma: radiologic features in 14 cases. AJR Am J Roentgenol 1988; 150: 861 – 3. 80. Erdem E, et al. Comprehensive review of intracranial chordoma. Radiographica 2003; 23: 995 – 1009. 81. Lanzino G, et al. Chordomas and chondrosarcomas involving the cavernous sinus: review of surgical treatment and outcome in 31 patients. Surg Neurol 1993; 40: 359 – 71. 82. Gay E, et al. Chordomas and chondrosarcomas of the cranial base: results and follow-up of 60 patients. Neurosurgery 1995; 36: 887 – 97. 83. Sundaresan N, et al. Surgical treatment of spinal chordomas. Arch Surg 1987; 122: 1479 – 82. 84. Ozaki T, Hillmann A, Winkelmann W. Surgical treatment of sacrococcygeal chordoma. J Surg Oncol 1997; 64: 274 – 9. 85. Arnold H, Herrmann HD. Skull base chordoma with cavernous sinus involvement. Partial or radical tumour-removal? Acta Neurochir 1986; 83: 31 – 7.
86. Crumley RL, Gutin PH. Surgical access for clivus chordoma. The University of California, San Francisco, experience. Arch Otolaryngol Head Neck Surg 1989; 115: 295 – 300. 87. Sen CN, et al. Chordoma and chondrosarcoma of the cranial base: an 8-year experience. Neurosurgery 1989; 25: 931 – 41. 88. Goel A. Middle fossa sub-gasserian ganglion approach to clivus chordomas. Acta Neurochir 1995; 136: 212 – 6. 89. Maira G, et al. Surgical treatment of clival chordomas: the transsphenoidal approach revisited. J Neurosurg 1996; 85: 784 – 92. 90. Megerian CA, et al. The subtemporal-transpetrous approach for excision of petroclival tumors. Am J Otol 1996; 17: 773 – 9. 91. Jho HD, et al. Endoscopic transsphenoidal resection of a large chordoma in the posterior fossa. Acta Neurochir 1997; 139: 343 – 8. 92. Samson IR, et al. Operative treatment of sacrococcygeal chordoma. A review of twenty-one cases. J Bone Joint Surg 1993; 75: 1476 – 84. 93. Berson AM, et al. Charged particle irradiation of chordoma and chondrosarcoma of the base of skull and cervical spine: the Lawrence Berkeley laboratory experience. Int J Radiat Oncol Biol Phys 1988; 15: 559 – 65. 94. Austin-Seymour M, et al. Fractionated proton radiation therapy of chordoma and low-grade chondrosarcoma of the base of skull. J Neurosurg 1989; 70: 13 – 7. 95. Austin-Seymour M, et al. Fractionated proton radiation therapy of cranial and intracranial tumors. Am J Clin Oncol 1990; 13: 327 – 30. 96. Keisch ME, Garcia DM, Shibuya RB. Retrospective long-term followup analysis in 21 patients with chordomas of various sites treated at a single institution. J Neurosurg 1991; 75: 374 – 7. 97. Austin JP, et al. Probable causes of recurrence in patients with chordoma and chondrosarcoma of the base of skull and cervical spine. Int J Radiat Oncol Biol Phys 1993; 25: 439 – 44. 98. Thieblemont C, et al. Prognostic factors in chordoma: role of postoperative radiotherapy. Eur J Cancer 1995; 31: 2255 – 9. 99. Tai PTH, Craighead P, Bagdon F. Optimization of radiotherapy for patients with cranial chordoma. A review of dose-response ratios for photon techniques. Cancer 1995; 75: 749 – 56. 100. Saxton JP. Chordoma. Int J Radiat Oncol Biol Phys 1981; 7: 913 – 5. 101. Cummings BJ, Hodson DI, Bush RS. Chordoma: the results of megavoltage radiation therapy. Int J Radiat Oncol Biol Phys 1983; 9: 633 – 42. 102. Thilmann C, et al. Intensity-modulated radiotherapy of sacral chordoma. A case report and a comparison with stereotactic conformal radiotherapy. Acta Oncologica 2002; 41: 395 – 9. 103. Gabriele P, et al. Feasibility of intensity-modulated radiation therapy in the treatment of advanced cervical chordoma. Tumori 2003; 89: 298 – 304. 104. Jena R, et al. Conformal rotation therapy with central axis beam block is a feasible alternative to intensity-modulated radiotherapy for chordomas of the cervical spine. Clin Oncol 2004; 16: 449 – 56. 105. Taylor RE, et al. Proton therapy for base of skull chordoma: a report of the Royal College of Radiologists. The proton therapy working party. Clin Oncol 2000; 12: 75 – 9. 106. No¨el G, et al. Combination of photon and proton radiation therapy for chordomas and chondrosarcomas of the skull base: the Centre de Protontherpie D’Orsay experience. Int J Rad Oncol Biol Phys 2001; 51: 392 – 8. 107. Schulz-Ertner D, et al. Radiotherapy for chordomas and low-grade chondrosarcomas of the skull base with carbon ions. Int J Rad Oncol Biol Phys 2002; 53: 36 – 42. 108. Imai R, et al. Carbon ion radiotherapy for unresectable sacral chordomas. Clin Cancer Res 2004; 10: 5741 – 6. 109. Kumar PP, et al. Local control of recurrent clival and sacral chordoma after interstitial irradiation with iodine-125: New techniques for treatment of recurrent or unresectable chordomas. Neurosurgery 1988; 22: 479 – 83. 110. Kondziolka D, Lunsford LD, Flickinger JC. The role of radiosurgery in the management of chordoma and chondrosarcoma of the cranial base. Neurosurgery 1991; 29: 38 – 46. 111. Feigl GC, Bundschuh O, Gharabachi A. et al. Evaluation of a new concept for the management of skull base chordomas and chondrosarcomas. J Neurosurg 2005; 102: 165 – 70. 112. Fleming GF, et al. Dedifferentiated chordoma. Response to aggressive chemotherapy in two cases. Cancer 1993; 72: 714 – 8.
CHORDOMAS 113. Casali PG, et al. Imatinib mesylate in chordoma. Cancer 2004; 101: 2086 – 97. 114. Newton HB. Molecular neuro-oncology and the development of “targeted” therapeutic strategies for brain tumors. Part 1 – growth
625
factor and ras signaling pathways. Expert Rev Anticancer Ther 2003; 3: 595 – 614.
Section 10 : Neurological Malignancies
57
Meningeal Sarcomas
Nicholas G. Avgeropoulos and John W. Henson
INTRODUCTION Sarcomas are defined as tumors of “nonepithelial tissues of the body, exclusive of the reticuloendothelial system, glia, and supporting tissue of parenchymal organs.”1 Despite the relative rarity of these tumors, their behavior as destructive and locally invasive cancers with a propensity to metastasize is well known. When combined, the three most common systemic soft tissue sarcomas encountered in adulthood (malignant fibrous histiocytoma, liposarcoma, and fibrosarcoma) account for only 0.5% of all cancers.1,2 This chapter focuses on an even rarer group of sarcomas: those arising in the coverings of the nervous system. Recent classification schemes for meningeal sarcomas have focused on specific cellular differentiation (i.e. the tissue that the sarcoma resembles) instead of more traditional “cell of origin” approaches (see Table 1).3,4 This chapter reviews fibrosarcoma, mesenchymal chondrosarcoma, chondrosarcoma, rhabdomyosarcoma (RMS), leiomyosarcoma, meningeal sarcomatosis, malignant fibrous histiocytoma, angiosarcoma, liposarcoma, and osteosarcoma tumor types. Hemangiopericytoma, a debatably “nonsarcomatous” entity, is not described here.
EMBRYOLOGY Both mesodermal and ectodermal neural crest elements give rise to the mesenchymal components of the primitive meninx which then divide into the ecto- and endomeninx. The ectomeninx becomes the endorhachis, which forms the periosteum of the vertebral canal in the spine and periosteal dural layer in the cranium. The endomeninx forms the leptomeninx, which then separates into the arachnoid and pia mater. The arachnoid trabeculae merge with the pia, covering cells of reticuloendothelial origin and becoming an integral part of this innermost layer.5 – 8 The fact that deep-seated (i.e. intraparenchymal) mesenchymal sarcomas are seen at least as often as their meningeal counterparts invites an explanation. Most authors on the subject agree that primary intracranial sarcomas are derived from precursor elements of the dura mater, adventitia of intraparenchymal blood vessels, tela choroidea, choroid
plexus, leptomeninges, and pial extensions along the perivascular spaces.9 – 11 The diversity of these cellular constituents would attest to the pluripotent nature of undifferentiated mesenchyme. Others maintain that misplaced mesenchymal cell rests are ultimately responsible for tumor growth but concur that the division into meningeal, intraparenchymal, and intraventricular sarcomas is artificial.10,12 This latter viewpoint is supported by the observation that the pia and inner lamina of the arachnoid are not likely to extend beyond large and small caliber vessels to surround capillaries.8
EPIDEMIOLOGY The rarity and early description of central nervous system (CNS) sarcomas require that a review of this subject span many decades of the medical literature. The historical classification of medulloblastomas, glioblastomas, gliosarcomas, oligodendrogliomas, angiomas, aneurysms, parasitic cysts, lymphomas, meningiomas, and granulomas as sarcomas has often made it difficult to effectively analyze the nature of meningeal sarcomas.13 Virchow initially used the term “sarcoma” to describe malignant fleshy intracranial tumors as opposed to what he termed “gliomas,” which were felt to be less malignant.14 As such, older statistics list sarcomas as comprising a surprising 30–40% of brain tumors.3 Over time, classification schemes were devised on the basis of the histologic architecture (e.g. perivascular, perithelial, alveolar), on the site of origin within the cranial vault (e.g. dural, leptomeningeal), and according to histologic subtype (e.g. fibrous, spindle cell, and polymorphocellular).12 The use of histologically descriptive terms such as perithelial, periadventitial, alveolar, perivascular, round cell, reticuloendothelial, circumscribed, or monstrocellular have also obfuscated the understanding of these tumors. The tumors described as “periadventitial sarcoma” and “reticular sarcoma” have now been recognized as malignant lymphomas, “cerebellar arachnoidal sarcoma” has been more recently appreciated as desmoplastic medulloblastoma, and “monstrocellular sarcoma” or “circumscribed sarcoma of the blood vessels” is designated as giant cell glioblastoma.13 Additionally, the primary nature of this tumor has many times not
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
MENINGEAL SARCOMAS Table 1 Mesenchymal neoplasms not of meningothelial type.
Benign
Malignant
Fibrous histiocytoma Chondroma, osteochondroma, osteoma Cavernous hemangioma Lipoma Myxoma
Hemangiopericytoma Fibrosarcoma/malignant fibrous histiocytoma Mesenchymal chondrosarcoma Leiomyosarcoma Angiosarcoma Meningeal sarcomatosis Sarcoma, not otherwise specified Others
been confirmed in the literature by autopsy, or, in some cases, the possibility of intracranial extension from the skull or parameningeal sites was or could not be adequately excluded. Patients of all age-groups, including congenital and neonatal cases, have been afflicted with this tumor (see Table 2).15 – 26 Arguably, the pediatric population may account for a disproportionately high percentage of meningeal sarcomas.9,12 However, in an extensive review of fibrosarcomas, Kernohan and Uihlein found that the greatest intracranial incidence of these tumors was in the sixth decade of life.8 There does not appear to be a sex predilection or a consistently preferential site of occurrence for meningeal sarcomas.4,9 According to Zulch, the ratio of supra- to infratentorial tumors is roughly equal to the ratio of brain mass in these two compartments, respectively (about 3 : 1).3 However, Kernohan and Uihleins’ series demonstrated that the parietal area was overrepresented.8 An origin from the spinal meninges is rare but well described.27 – 29
627
Table 2 Incidence of intracranial sarcomas: summary of adult and pediatric literature.
Reference
Percentage of intracranial sarcomas Cohort/comments
Zulch4
1.9 – 4.3
Nichols and Wagner10
0.5
Kernohan and Uihlein8
1.3
Russell and Rubinstein9
1.2
Tomita and Gonzales-Crussi15
1.5
Paulus et al.13
<0.1
Jellinger and Sunder-Plasman16
0.2 – 0.9
6
ETIOLOGY/EXPERIMENTAL MODELS The causes of primary meningeal sarcomas are poorly understood but some insight can be derived from experimental models. Zimmerman and Arnold used the carcinogenic compound methylcholanthrene to create intracranial sarcomas and gliomas in varying proportions.30 V´azquez-L´opez was the first to use a cell free filtrate to produce sarcomas in fowl.31 Subsequent to this use of the Rous sarcoma virus, other researchers have created intracranial sarcomas with injections of polyoma virus into hamsters.32 Strontium-90 has been administered intravenously into pregnant rats, yielding metastasizing meningeal sarcomas in 6% of offsprings. Interestingly, 2% of these offsprings also developed simultaneous pituitary adenomas and meningeal sarcomas.33 In humans, exposure to therapeutic radiation may be a risk factor for intracranial sarcoma formation. The so-called “pituitary sarcomas,” which account for most radiationrelated intracranial sarcomas, are fibrosarcomas that arise in the suprasellar cistern after a latent interval of 2.5–27 years and a median time to appearance of 11 years after radiation exposure. These patients had frequently received a total radiation dose greater than 50 Gy but many had initial and cumulative tumor inducing doses of photon or proton irradiation ranging from doses as low as 2–10 Gy.33,34 The reported survival time for these patients with pituitary sarcomas was 6 months or less in over 90% of cases.33 An autopsy case of radiation-induced intracranial malignant
Fessard17
6
Hope et al.18
0.1
Nakamura and Becker19
0.1
Takaku et al.20
2.9
Davidson and Hope21
7
Raimond and Tomita22
1.5
a
1.9% in a series of 4000 intracranial tumors from all age-groups; 4.3% in a series of 9000 1.5% (9/584) of all primary brain tumors. Three of these, all from adults, were meningeal in origin 108/8070 all brain tumorsa Authors’ estimate from personal and other datasets Six meningeal sarcomas/402 intracranial tumors in children 19/25 000 brain tumor biopsies; 12/19 sarcomas were meningeal. Some of these cases overlap with Jellinger and Sunder-Plasmans’ review belowb One and possibly five out of 572 histologically confirmed intracranial neoplasms 0 – 16 years of age from an institutional database Retrospective review of 132 tumors in the first year of life; eight of these were sarcomas 4/66 cases, cerebral tumors in infancy (<2 years old)c 3/690 patients of all ages with primary intracranial tumors were meningeal sarcomasd 5/950 intracranial tumors of infancy and childhood (three meningeal sarcomas and two MFHs) 3/103 sarcomas from intracranial tumors in the neonatal period (literature review) 7% of meningeal tumors in childhood were sarcomas 5/341 childhood brain tumors were meningeal sarcomas; 3 of these were in the 0 – 1 years old groupe
This figure excludes reticulum cell sarcoma, giant cell fibrosarcoma, hemangiopericytoma, and circumscribed sarcoma of the cerebellum. b Did not include meningeal sarcomatosis. c Distinction between meningeal and intraparenchymal or intraventricular is not made. d These three meningeal sarcomas comprised one half of the six malignant tumors that were discovered. e Tumors with sarcomatous change are estimated to have a much higher frequency.23
628
NEUROLOGICAL MALIGNANCIES
fibrous histiocytoma that developed in the region of the sella turcica 11 years after high-dose radiation treatment of a chromophobe adenoma of the pituitary has also been reported.35 Specific reports of radiation-induced sarcomas are reviewed elsewhere in the chapter. Familial cases have been cited, although these are rare events. Gainer et al. reported cases of fibrosarcoma in a father and daughter and, in another family, in two sisters both in their seventh decade.36 Monosomy 22 has been reported in rhabdoid tumors.37 A genetic influence may be a factor in NF1, where a higher than expected incidence of dural fibrosarcomas has been noted.29,38 A reactive etiology is suggested by the finding of intracerebral sarcomas postoperatively. Here, meningothelial elements have been seen amidst sarcomatous cells and may represent the dedifferentiation of meningioma cells into more malignant cells or possibly the transformation of mesenchymal tissue into sarcomas secondary to irritation by the meningioma.39 Supporting this notion, Ho et al. reported an intracerebral malignant fibrous histiocytoma diagnosed at the operative site of a posterior communicating artery aneurysm clipped 3 months earlier.40 Similarly, Kristoferitsch and Jellinger reported a case of angiosarcoma occurring at the site of a chordotomy performed 5 years earlier.41,42 Finally, Reinhardt outlined a case of meningeal sarcoma in the center of which a metal wire was found, as the result of a boiler explosion 20 years earlier.9
CLINICAL PRESENTATION The presenting symptoms of meningeal sarcomas are most often secondary to extrinsic compression of neural structures. Reported signs and symptoms include headache ranging from dull to severe and diffuse, nausea, vomiting, papilledema, seizure activity, visual changes, diminished sensorium, root pains, aphasia, bulging fontanelle, failure to thrive, increasing head circumference, focal weakness or numbness, spinal cord sensory level, and paraparesis, among other findings.
RADIOGRAPHIC CHARACTERISTICS The characteristic magnetic resonance imaging (MRI) appearance of a meningeal sarcoma has been reported as T1 – dark and T2 – bright with ring or homogeneous enhancement (see Figure 1).43 However, enhancement patterns are highly variable on MRI and may be heterogeneous and nodular. Computerized axial tomography (CT) of three meningeal sarcomas in the series of Hope et al. revealed that, in distinction to meningiomas, two-thirds had fringed (poorly defined) margins, two-thirds had cysts, and all had heterogeneous enhancement.18,44 Sano’s group noted intratumoral cysts in 17% of their series of meningeal tumors in children, and all of these cases turned out to be meningeal sarcomas.45 The cysts may be mistaken for trapped cerebrospinal fluid (CSF), focal brain atrophy, or adjacent edema as alternative explanations to true cystic degeneration. There have also been several reports of subdural sarcomas discovered and originally presumed to be acute or chronic subdural hematomas.46 – 50
Figure 1 T1-weighted MRI with gadolinium contrast, coronal view of a 52-year-old male with a poorly differentiated meningeal sarcoma who presented after experiencing progressively severe headaches, periorbital pain, and proptosis. The scan demonstrates extra-axial regions of thickened, nodular soft tissue enhancement that transgresses into brain parenchyma along the left temporal and frontal lobes. Marked edema with resultant midline shift is appreciated.
Massive osteolysis of the skull has been seen in a case of meningeal sarcoma where skull bones were largely replaced by tumor.51 Skull erosion can be seen in 25% of these patients but there is generally no evidence of hyperostosis. In the series of Hope et al., two-thirds of the patients had bone destruction, with extracranial extension in one instance.18 Tumor vascularity, as determined by contrast-enhanced MRI/CT and angiography, may range from hypovascular to hypervascular with a tumor blush on angiography (see Figure 2).43,52,53 Latchaw et al. described angiographic findings of a case of meningeal sarcomatosis where he discovered direct correspondence between the sites of vascular irregularity, seen angiographically, and areas of microscopic vessel wall infiltration by tumor.54 The radiographic differential diagnosis of primary intracranial sarcomas includes extracranial sarcoma with metastases, intracranial extension from the skull or other parameningeal sites, neoplastic mesenchymal component in the context of a neuroectodermal tumor, malignant meningioma, or nonmalignant tumors that may simulate sarcomas because of pleomorphism (pleomorphic xanthoastrocytoma, benign fibrous histiocytoma).12,42,55 Special imaging characteristics of the various tumor types are included in the sections below.
GROSS PATHOLOGY Meningeal sarcomas involve the dura or leptomeninges solely or may extend into the surrounding tissues. They can also form large masses that are clearly demarcated from the adjacent brain. Relatively well-differentiated, dense, firm, grayish white fibrosarcomas tend to originate from
MENINGEAL SARCOMAS
629
Figure 2 A 55-year-old female with meningeal fibrosarcoma who presented for a third resection after experiencing progressive right hemiparesis and focal motor seizures. (a) T1-weighted MRI, coronal view – the scan reveals a large, lobulated, hypointense, extra-axial mass abutting the right falx and extending along the roof of the right lateral ventricle. Substantial edema and subfalcine herniation can be seen. (b) Cerebral angiography, anteroposterior view – right carotid artery injection yields a prominent tumor stain. The fibrosarcoma is fed by enlarged branches of the superficial-temporal as well as the middle-meningeal artery.
the dura.14,26 Sarcomas taking their origin from the leptomeninges or from within the brain are more frequently softer, more fleshy, and gray with a modestly granular cut surface. Areas of hemorrhage, necrosis, or cystic degeneration can be seen, particularly in more rapidly growing tumors.9,14,26,56,57 In pediatric primary meningeal tumors, Kolluri et al. noted cysts in 4 of 15 meningiomas and 2 of 4 meningeal sarcomas, thereby complementing the radiographic findings of Sano et al.45,58 Although they are at times well circumscribed at surgery or necropsy, these sarcomas are not encapsulated, infiltrate other tissues, and exhibit a high incidence of meningeal seeding.9,57
were growing relatively slowly, with well-developed connective tissue processes, no pleomorphism of the nuclei, and sparse mitotic figures, and the very malignant neoplasms at the other extreme (see Figure 3). Even in the most malignant fibrosarcomas of the cerebrum, reticulum is present and can be seen in the cytoplasm of many cells.8 The central portion of a fibrosarcoma may be better differentiated and frequently presents with comparatively more reticulin fibers while the growing edges of the tumor show the least degree of differentiation.9 Fibrosarcomas have no other characteristic immunohistochemical staining pattern, although most mesenchymal tumors are vimentin positive.
FIBROSARCOMA Christensen and Laras’ database on primary intracranial fibrosarcomas reflects the general experience with this tumor.12 Their series included 16 males and 9 females. The ages ranged from 9 months to 67 years, with an average age of 29.5 years and nine patients being below 20 years of age. There was also a tendency for the malignant forms of fibrosarcomas to develop earlier than the more differentiated types. Histologic parameters used in different grading systems include cellularity, differentiation, pleomorphism, mitotic rate, and necrosis, with authors placing emphasis on different parameters.59 An older, more commonly used classification scheme considered that the presence of mitotic figures was the best indicator of malignancy for intracranial fibrosarcomas.12 The “fibrous” type was the least aggressive, the “spindle cell” type was an intermediate grade, and the “undifferentiated” or “polymorphocellular” type was the most malignant. Although the concept of a three-tiered grading system is well accepted, these terms are not widely used in present day pathology. In Kernohan and Uihleins’ series of intracranial fibrosarcomas, there was a gradual transition between those that
Figure 3 Fibrosarcoma – a striking herringbone pattern is composed of interlacing bundles of spindle cells with a moderate amount of cytoplasm, ill-defined cell boundaries, and oval nuclei. Mitotic figures were seen at higher power.
630
NEUROLOGICAL MALIGNANCIES
The mode of spread of primary meningeal fibrosarcomas includes direct extension, via Virchow –Robin spaces, or into the cerebral cortex by fingerlike projections.8 Indeed, most fibrosarcomas show characteristic peripheral tongues of growth, which irregularly penetrate the surrounding parenchyma. Characteristically, distinct islands of neuroglia are encompassed by the advancing sarcoma, a feature which greatly helps in distinguishing invasive sarcomas from anaplastic gliomas.13 The tendency to recur locally after total macroscopic resection is typical of fibrosarcomas arising elsewhere in the body and tends to be the rule for primary meningeal fibrosarcomas as well.60 A review of nine cases of primary fibrosarcomas of the brain showed local recurrence in eight, distant recurrence in six, meningeal seeding in four, and systemic metastases in four patients.55 Many cases of extraneural metastases have been described in the literature as well12,14,61 – 64 Finally, cases of meningeal sarcoma thought to give rise to glioma have been reported.8,65,66
MESENCHYMAL CHONDROSARCOMA In 1959, Lichtenstein and Bernstein originally described mesenchymal chondrosarcoma as a tumor arising from the bone.67 Five years later, Dowling described the first case of dural mesenchymal chondrosarcoma.68 Subsequently, the dura has been the most commonly reported site of extraosseous origin of this tumor, occurring mainly in the first to fifth decades but more prominently in the first and second decades of life69 (see Table 3).70 – 74 There has been no reported sex preference. Extracranial metastases occurred in 12% of the patients in Hassounah’s series and have also been reported by Rollo et al.70,75 Grossly, the tumors are well circumscribed and often invasive of brain or overlying the bone. They have a grayish red knobby surface that is firm to palpation. Most
patients with reported radiologic studies showed some form of calcification of the tumor on skull X ray or CT of the head.70 Two histologic patterns predominate. The first is a highly undifferentiated stroma composed of small cells that resemble primitive mesenchymal cells. These cells vary from polygonal to fusiform in shape and occasionally show a palisading arrangement. They have hyperchromatic, moderately pleomorphic nuclei with distinct nucleoli and frequent mitoses. Narrow thin-walled vascular spaces (also known as staghorn channels) lined by flattened endothelium divide these cells into clusters and cords.76 As such, the main differential diagnoses on pathologic review include angioblastic meningioma or hemangiopericytoma. Ultrastructurally, round-to-ovoid cells with large nuclei are seen. Chromatin is irregularly clumped centrally and nucleoli are frequently seen. Scattered swollen mitochondria are also present.76 The second histologic pattern is composed of islands of hyaline cartilage in various stages of maturity interspersed in the cellular matrix. The transition from the mesenchymal components of the tumor to cartilage formation is usually abrupt (see Figure 4). Most cells within the cartilaginous islands appear immature, and ossification and calcification are generally limited to these areas. Oval chondrocytes have benign appearing large nuclei, prominent nucleoli, and small amounts of cytoplasm. Reticulin silver stains will outline delicate fibers.71
CHONDROSARCOMA The peak incidence of intracranial chondrosarcomas occurs during the sixth decade of life, whereas mesenchymal chondrosarcomas are tumors that tend to be found in young adults.72 Only 0.16% of all cranial and intracranial lesions are cartilaginous tumors, and chondrosarcomas constitute
Table 3 Mesenchymal chondrosarcomas: summary of reports from the literature.
Author
Age range (years)
Hassounah et al.70
11 – 51
Harsh and Wilson71
5 – 51
Heros et al.72
7 – 48
Scheithauer and Rubinstein73
5 – 48
Cohort 50 cases of primary intracranial and mesenchymal chondrosarcomas; of these, 13 were dural based 21 cases of mesenchymal chondrosarcoma with primary limited to the CNS. 12/16 intracranial cases were dural 18 cases of mesenchymal chondrosarcoma were reviewed. Of these, 13 had dural attachments 6/7 intracranial mesenchymal chondrosarcomas had a dural attachment. Of the five intraspinal examples, three had a dural attachment
MFH, malignant fibrous histiocytoma; CNS, central nervous system.
Figure 4 Mesenchymal chondrosarcoma – highly undifferentiated mesenchymal cells with hyperchromatic nuclei and frequent mitoses abut abruptly into immature cartilaginous islands. Narrow thin-walled vascular spaces stud the mesenchymal component of the tumor.
MENINGEAL SARCOMAS
about 14% of this subset.77 Dural-based chondrosarcomas accounted for 9 of 50 reported cases in the review by Cybulski et al.44 Although clinical outcomes of patients with chondrosarcoma have been described as indolent following surgery as monotherapy,78,79 Bosma et al.80 reviewed the case of a dural chondrosarcoma that required radiation treatment to attain clinical stability. Extracranial skeletal metastases have also been documented in this condition.53,81 Chondrosarcomas match the pattern of other chondromas radiographically. Calcifications visible on plain films as stippled, finely speckled, or amorphous clumps can be seen in more than 50% of patients. Additionally, bony destruction at the site of the chondrosarcoma was seen in 50% of the cases reported by Grossman and Davis, and in three out of four of Bahr and Gayler’s cases.82,83 Bahr and Gayler reported a case of meningeal chondrosarcoma that appeared as an avascular mass on cerebral angiography, while the case reported by El Gindi et al. demonstrated a vascular blush.53,83 On CT imaging, chondrosarcomas enhance irregularly but are sharply marginated. The most frequent site of origin for chondrosarcomas is the parasellar region followed by the cerebellopontine angle/petrous bone juncture.44 Grossly, the tumor is arranged in a lobular pattern with a well-delineated fibrous capsule. Intracranial chondrosarcomas generally occur as solitary lesions but they may also occur in conjunction with similar lesions in the rest of the skeleton (Ollier’s disease) or in association with a congenital mesodermal dysplasia associated with hemangiomas and tumors of bone (Maffucci’s syndrome).44 Histologic characteristics reveal a pseudoencapsulated cartilaginous tumor arranged in a lobular pattern with welldefined fibrous septae. The chondrocytes within the lobules resemble primitive cartilage, have plump atypical nuclei, and are composed chiefly of small, round-to-oval cells with vesicular, slightly hyperchromatic nuclei and a narrow rim of eosinophilic cytoplasm. Binucleated cells and small foci of calcification may also be seen. Portions of the tumor may be composed of spindle shaped cells or a loose reticular or myxoidal pattern but mitoses are not typically seen. As opposed to their mesenchymal counterparts, chondrosarcomas have stromal cells that are much larger and show greater pleomorphism, and the chondroid component is more anaplastic.72 The myxoid ground substance seen in regions of chondrosarcomas can show strong positive staining for both acid and sulfated mucopolysaccharide. Periodic acid-Schiff (PAS) positive, diastase labile granules highlight the presence of glycogen.84 Chondrosarcomas are cytokeratin and EMA negative, vimentin positive, and, sporadically, S-100 positive as well. Kernohan and Uihlein described a mixed fibrous chondrosarcoma and malignant astrocytoma in a 56year-old woman.8 Similarly, Katayama et al. described a case of meningocytic differentiation in a meningeal chondrosarcomatous tumor.85 On electron microscopy, some of the rough endoplasmic reticulum of the tumor cells can be seen to be partially dilated and to contain a homogenous amorphous material.86 Polygonal to oval and stellate cells with variable amounts of cytoplasm are present. Cytoplasmic membranes are well
631
defined but irregular and sometimes indented, giving a scalloped appearance.84
RHABDOMYOSARCOMA RMS, the most common soft tissue sarcoma of childhood, arises from primitive mesenchymal cells in a variety of tissues.87 Primary RMS of the CNS should be diagnosed only for brain tumors composed of purely mesenchymal derivatives consisting of both embryonal and mature striated muscle cells. In addition, these tumors must be differentiated from other primary and secondary brain tumors containing myocytic components such as medullomyoblastomas or teratomas.88 – 90 Dropcho and Allen reviewed 34 patients with primary intracranial RMS. In their series, 13 patients were less than 6 years old, 11 patients were 6 to 18 years of age, and 10 patients were over 18 years old.88 Involvement of the meninges at the time of diagnosis was a common occurrence in this series. In six of these patients, the tumor arose primarily within the meninges (either diffusely or as a focal mass) without involvement of brain parenchyma, and in six other patients there was continuous invasion of the brain and the overlying meninges at initial presentation. This is consistent with other reports.91 Dissemination of the RMS through the cranial and/or spinal leptomeninges was a pathologically proven complication in 11 patients (32%). Smith et al. have also reported a case of diffuse meningeal RMS.92 Legier and Wells reported five cases of cerebellar RMS, four of which were in patients between the ages of 2 and 10 years.93 RMS arising within the CNS does not appear to demonstrate any distinctive pathologic features in relation to its systemic counterpart. Spindle, oval, or racquet-shaped rhabdomyoblasts with abundant eosinophilic cytoplasm are noted on microscopic examination. Most tumors also contain areas of poorly differentiated embryonal or mesenchymal cells characterized by scant, indistinct cytoplasm and hyperchromatic nuclei (see Figure 5). Cross-striations, a trademark of the skeletal muscle, are not reliably characterized on routine staining, and the pathologist will routinely need to perform immunohistochemical analysis and electron microscopy. A diagnostic pitfall includes misdiagnosis of this tumor as a primitive neuroectodermal tumor (PNET).88 Immunohistochemistry is positive for actin, desmin, myoglobin, and vimentin, and negative for lymphoid, granulocytic, and nervous markers.39 Myoglobin is considered to be the most specific marker, while the others listed are useful in a more circumferential way.88 Electron microscopy will show abortive Z-lines and alternating thick and thin myofilaments.39,88 Primary intracranial RMS can metastasize. Namba et al. described a case where several firm metastatic nodules located in the spinal meninges and the cauda equina represented cerebrospinal fluid dissemination from the primary tumor.88 Conversely, intracranial extension from the skull or parameningeal sites must always be considered when confirming the primary nature of this tumor.13
632
NEUROLOGICAL MALIGNANCIES
(a)
(b)
(c)
Figure 6 (a) Leiomyosarcoma as seen on sagittal T1-weighted image before contrast administration demonstrates a dural-based mass lesion that is isointense to brain (arrow). After administration of gadolinium (b) the tumor demonstrates moderate homogeneous enhancement. The tumor is slightly hypointense to brain on T2-weighted image (c). Images courtesy of Geoffrey Young, MD, Division of Neuroradiology Brigham and Women’s Hospital, Boston, Massachusetts.
Figure 5 Rhabdomyosarcoma – highly pleomorphic cells range from plump, fusiform and straplike with prominent nuclei to diminutive with scant, indistinct cytoplasm and small hyperchromatic nuclei. Cross-striations are not reliably seen with routine staining.
LEIOMYOSARCOMA Although leiomyosarcoma may develop from any site where smooth muscle cells exist, the most common locations are the uterine retroperitoneum, the mesentery, and the mediastinum, with primary meningeal leiomyosarcoma being a rare occurrence. The MRI appearance of a patient with meningeal leiomyosarcoma is demonstrated in Figure 6. Asai et al. reported a case of leiomyosarcoma developing from the dura and extending extracranially.94 These authors contend that the marked periluminal tumor proliferation found in this case strongly supported the possibility that the tumor was of blood vessel origin and that the primary meningeal leiomyosarcoma had arisen from blood vessels of the dura mater or those of the skull including the periosteum. In addition to a meningeal origin, intracranial leiomyosarcoma could also potentially arise from components of the choroid plexus stroma or tela choroidea.95 Other cases of primary intracranial leiomyosarcoma include those described by Lee and Page96 who reported a rapidly expanding left parietal extra-axial lesion in an 8-year-old male, and Paulus et al.13 who reported a dural leiomyosarcoma in a series of 19 intracranial sarcomas (see Table 2). Sugita et al. reported two cases of primary meningeal sarcoma with leiomyoblastic differentiation.97 Bejjani et al.98 described a dural leiomyosarcoma in a human immunodeficiency virus (HIV) positive patient as did Zevallos-Giampietri et al.99 Microscopic examination shows a densely cellular background composed of wide intersecting fascicles of plump spindle cells. The cells have long round-ended nuclei and moderate eosinophilic cytoplasm, and mitotic figures are common (see Figure 7). Immunohistochemistry is usually positive for vimentin, smooth muscle –specific actin, and desmin, and negative for S-100 and GFAP. On electron microscopic examination, numerous intracytoplasmic thin filaments are found.95
Figure 7 Leiomyosarcoma – spindly smooth muscle cells with elongated, blunt nuclei are arranged in coarse bundles traversing at near right angles in this photomicrograph. Mitotic figures are noted.
MENINGEAL SARCOMATOSIS In a review of 10 primary diffuse tumors of the meninges, Black and Kernohan reported three patients less than 5 years of age, two patients between 10 and 20 years of age, four patients in their third decade, and one patient over 40 years of age.100 Adults are occasionally affected but most reported cases have been in infants, children, and young adults.7,101 – 104 Grossly, the leptomeninges appear milky, cloudy, and thickened, especially around the base of the brain. Diffuse tumor extends to the depths of the sulci and occasional small nodules of subarachnoid tumor may occur along the cerebrospinal axis.14,105 The main differential diagnoses include meningoencephalitis, medulloblastoma, and lymphoma.7,55 Microscopically, cells are round-to-oval with sparse cytoplasm without processes (see Figure 8). Areas between pia and arachnoid trabeculae are filled with tumor cells. Neoplastic cells are suspended in an abundant network of reticulin. Nuclei are round with heavily stained chromatin granules,
MENINGEAL SARCOMAS
Figure 8 Meningeal sarcomatosis – bands of cells with fibrillar cytoplasm are highlighted by oval, sometimes dense chromatin containing nuclei. The field is highly cellular.
and binucleated cells have also been identified. The number of mitotic figures seen varies, as does the degree of pleomorphism.13 Tumors can invade the brain or spinal cord by breaking directly through the pia or by spreading along the perivascular spaces.9 Immunohistochemistry shows focally positive staining for cytokeratin, EMA, and vimentin and diffuse staining for S-100. Ultrastructural examination shows plasma membrane interdigitations, desmosomes, gap junctions, and abundant intermediate filaments.56
MALIGNANT FIBROUS HISTIOCYTOMA Malignant fibrous histiocytomas and their variants in skin and soft tissue have been studied most extensively, and have also been referred to as “malignant fibrous xanthomas” and “fibrous xanthosarcomas” in the literature. These tumors accounted for 4 of 12 primary intracranial meningeal sarcomas in patients aged between 22 and 67 years in one series. Swamy et al. presented data on six patients with malignant fibrous histiocytoma primarily arising from the cranial meninges, including the group’s own case report.106 The age range of these cases was 12 to 54 years of age, and the series exhibited no particular sex preference. A case of meningeal malignant fibrous histiocytoma arising from a thoracolumbar myelomeningocele has also been reported.107 Grossly, these lesions are irregularly lobulated with multiple hemorrhagic areas. White firm areas are intermixed with well delineated, distinctly myxoid or mucoid areas.108,109 On microscopic examination, malignant fibrous histiocytomas are characterized by a variable mixture of bizarre fibroblasts, histiocytes, multinucleated giant cells, and foamy lipid-filled xanthomatous cells arranged in a storiform (pinwheel) pattern with spindle cell or myxoid stroma (see Figure 9). Mitoses and necrosis are frequently seen.15,40 Multinucleated tumor and Touton-type giant cells are present in moderate
633
Figure 9 Malignant fibrous histiocytoma – a characteristic storiform or pinwheel pattern is highlighted by star-shaped streams of fibroblasts. Large, rounded histiocytes with pleomorphic and sometimes multiple nuclei stand out from the background. Occasional lymphocytes with small nuclei are also appreciated.
numbers.108 In the series of Paulus et al., all storiform pleomorphic MFH cases showed islands of pleomorphic glial cells within the primary sarcoma.13 Immunohistochemical examination is positive for α-1 antichymotrypsin, vimentin, and lysozyme but negative for desmin, GFAP, and S-100 markers.42 Ultrastructurally, five different cell types are identified: fibroblastic, immature undifferentiated, xanthomatous, histiocytic, and multinucleated tumor giant cells. As such, malignant fibrous histiocytoma is considered routinely in the differential diagnosis of a pleomorphic sarcoma.108
ANGIOSARCOMA Angiosarcomas of the CNS are rare. Mena and Garcia described a 15-year-old girl who developed a sudden hemiparesis due to an angiosarcoma in the posterior frontal lobe.110 A second reported angiosarcoma arose in the spinal dura of a 60-year-old male 5 years after he had undergone chordotomy for intractable pain. A suture found in the center of this neoplasm raised the possibility that the response to a foreign body was involved in its pathogenesis.41
LIPOSARCOMA Cinalli et al. described the case of a 6-month-old girl who developed an asymmetric subdural fluid collection with several enhancing nodules bulging in the subdural cavity.50 Three months later, the patient underwent repeated craniotomies for a large frontoparietal lesion. Histologic examination revealed the diagnosis of liposarcoma. Kothandaram’s case of a 4-month-old infant presented in a similar fashion.47 Histologic examination shows pleomorphic cells with diameters varying from 20 to 100 µm. Larger cells have irregular nuclei and are often lobulated with prominent nucleoli. Mitotic figures are moderate in number. Some areas differ by their higher cellularity, showing rather closely attached rounded or ovoid cells with hyperchromatic nuclei
634
NEUROLOGICAL MALIGNANCIES
and only occasional cytoplasmic vacuoles. Scattered mitotic figures are seen in cellular areas.47,111
OSTEOSARCOMA Lam et al. described the case of a 64-year-old female with memory loss who was discovered to have a large, vascular, and partly calcified mass involving the anterior portion of the convexity of the left frontal lobe on CT scan.112 Upon resection, a highly cellular neoplasm characterized by the presence of irregular osteoid with occasional small foci of mineralization with single neoplastic osteoblasts or groups of these cells was seen. These extremely pleomorphic cells had hyperchromatic nuclei with one or two nucleoli and many mitoses. In a few less cellular areas, the stroma appeared myxoid with spindle or stellate tumor cells. Irregular sinusoidal capillaries were prominent with foci of hemorrhage and necrosis. There was also infiltration of a narrow rim of underlying cerebral cortex. The ultrastructural hallmark of early mineralization is the needle-shaped hydroxyapatite crystals deposited along collagen fibers. Lam’s case is very similar to that of a left frontoparietal area meningeal osteosarcoma described by Setzer et al.113 Osipov et al. described a case of radiationinduced osteosarcoma that occurred 10 years after treatment for a pituitary tumor.114 Kernohan and Uihlein recorded only three meningeal sarcoma cases containing foci of osteogenic tissue but were uncertain as to the exact nature of these neoplasms.8
TREATMENT Surgery Maximum feasible resection is the mainstay of diagnosis and treatment for all varieties of meningeal sarcoma. In most cases, strategic repeat operations serve as a focal point for the administration of multimodal treatments. After histologic confirmation, the physician should consider the possibility of CNS spread, a parameningeal focus, or extracranial metastases. Staging with an MRI of the neuraxis, CT of the chest, abdomen, and pelvis, a bone scan, and lumbar puncture for cytology (if possible) is prudent in order to fully assess tumor burden and optimize therapy. In some instances, the tumor is sufficiently circumscribed to permit total extirpation and a good clinical result.46 Christensen and Lara described a 21-year survival in a patient who required two gross total resections for treatment of a “fibrous” (low histologic grade) meningeal sarcoma.12 However, this is an exception to the otherwise poor survival of meningeal sarcoma patients, irrespective of treatment received (see section “Prognosis” below). Surgical cure, although sporadically mentioned in older literature, therefore raises the possibility of tumor misclassification.
Radiation Although they are often rapidly progressive tumors, meningeal sarcomas are not particularly radiosensitive. Gaspar et al. observed that doses of 64–66 Gy are required to treat microscopic residual extracranial fibrosarcomas.115
Evidence that lower doses of radiation (tolerated by the neuraxis) can control sarcomas is sparse.115,116 In Christensen and Laras’ series, intracranial fibrosarcoma patients were given “deep X-ray therapy” as soon as the diagnosis of sarcoma was histologically verified but it was not possible for these authors to state whether the treatment had any effect on overall survival.12 More recent approaches to radiation treatment have exploited technical advancements in stereotactic radiosurgery and improved methods of conformal photon therapy. The experience of Heros et al. with primary intracranial mesenchymal chondrosarcomas suggests the use of preoperative radiation therapy or embolization in lesions so vascular that they resemble arteriovenous malformations.72 The beneficial effects of radiation therapy in this tumor type has also been documented by Hoshino et al., who noted a 26% reduction in tumor volume on CT imaging.117 Harsh and Wilson reported good results from local radiation treatment of 50 Gy applied to three primary mesenchymal chondrosarcomas.71 With regard to skeletal mesenchymal chondrosarcomas, Huvos et al. discerned a “hemangiopericytomatoid” variant of systemic mesenchymal chondrosarcoma in eight patients, and ascertained that this subtype conferred a better response rate and prognosis to combination chemotherapy and radiation treatment.118 This has not been evaluated with respect to primary meningeal mesenchymal chondrosarcomas. Reports of radiation treatment for other primary meningeal sarcoma histologies are largely anecdotal. Suit et al. achieved local control with proton beam therapy in some cases of skull base chondrosarcomas by applying 50 Gy to the head and a 15 Gy boost to the tumor bed.119 Postoperative radiation therapy of 60 Gy was administered to the patient reported in Asai et al.’s review of primary leiomyosarcoma of the dura mater. No follow-up was available for this case, however.94 Tiberin et al. reported a case of brain sarcoma of meningeal origin that developed 6.5 years after 22 Gy cranial irradiation and multiagent chemotherapy was administered for treatment of acute lymphocytic leukemia.120 Brada et al. reviewed 334 patients with pituitary adenoma treated with conservative surgery and radiotherapy (median dose 45 Gy) and followed this cohort for 3760 person years.121 Five patients developed a second brain tumor, one of which was a meningeal sarcoma. Bojsen-Moller and Knudsen described the case of a 21-month-old boy from whom an ependymoma was removed, who was treated postoperatively with radiation of 43.5 Gy.122 Tumor recurrence was noted to be a meningeal sarcoma 7.5 years later. Differential diagnosis in these cases includes malignant meningioma and the neoplastic reaction that accompanies meningeal invasion by anaplastic gliomas.55
Chemotherapy Because of the rarity of this condition, there are no specific chemotherapy protocols of proven value for treatment of meningeal sarcomas. Chemotherapy regimens are individualized and directed at treatment of the histologic subtype and, in many instances, augmented with CNS penetrating agents. A case of a poorly differentiated meningeal fibrosarcoma was
MENINGEAL SARCOMAS
treated at the Massachusetts General Hospital with an institutional protocol of adjuvant vincristine 1.5 mg m−2 (2 mg), doxorubicin 37.5 mg m−2 , cyclophosphamide 1200 mg m−2 , and mesna alternating with etoposide 100 mg m−2 and ifosfamide 1800 mg m−2 . Although the patient remained clinically and radiographically stable for approximately 4 months while on this regimen, he showed progressive disease shortly afterwards. This chemotherapeutic approach is similar to the one used for treatment of a primary meningeal extraosseous Ewing’s sarcoma as reported by Stechschulte et al.123 After surgery and radiation treatment, their patient received ifosfamide, mesna, adriamycin, cytoxan, vincristine, and dactinomycin, as well as prophylactic intrathecal methotrexate, araC, and hydrocortisone, and demonstrated clinical response. Namba et al. treated primary intracranial RMS with intrathecal methotrexate as an adjunct to surgery and radiation. Unfortunately, their patient died from a pulmonary embolus a short time after the initiation of chemotherapy.88 Cinalli et al.’s patient with a primary meningeal liposarcoma responded radiographically to a combination of ifosfamide, carboplatin, and etoposide, but died following a recurrence months later.50
PROGNOSIS In general, the prognosis for patients with primary meningeal fibrosarcomas is poor. The literature is mostly anecdotal and frequently does not reveal details about exact histology or extent of resection, and often lacks chemotherapy or radiation therapy details and sites of relapse. As a rule, however, treatment is based on maximum feasible resection, involved field radiation or stereotactic radiosurgery (if warranted) to tolerance, and a “sarcoma-specific” chemotherapy regimen. In their review of nine patients with primary intracranial fibrosarcoma, Gaspar et al. described local recurrence in eight patients, distal recurrence in six patients, and systemic metastases in four patients.115 Eight of these patients died, with a median survival time of 7.5 months (1 day to 96 months), but a longer survival was observed in the subset of meningeal tumors. The treatment for most of the patients in this series included maximum feasible resection with radiation. Nichols and Wagner reviewed the course of leptomeningeal fibrosarcoma in three patients aged 37, 40, and 47 years.10 They reported a duration of symptoms of 19 years in the 40-year-old patient and more characteristic durations of 3 months and 2 weeks in the other two cases, respectively. All had postoperative survivals of less than 8 months but surgery with adjunctive radiation therapy appeared to yield an improved survival rate.10 Globus et al. reported on meningeal sarcomas discovered in the first two decades of life.124 The length of the clinical course was less than 6 months in six of eight cases, and 2 and 6 years in the remaining cases. All of these patients had signs of increased intracranial pressure, seizure activity, and signs of meningeal irritation at the time of admission.12 More optimistically, Kernohan and Uihlein reported that few of their patients with fibrosarcomas of “low grade malignancy” died within 2 years, and several of them were alive after many years.8
635
Harsh and Wilson reported three cases of intracranial mesenchymal chondrosarcoma where surgery and 50 Gy of radiation treatment had been utilized as therapy.71 In two of these three cases, local symptomatic recurrence followed within 1 year of the initial operation but the other patient was disease-free for at least 4.5 years after treatment. Hassounah et al. report more optimistic figures, stating a duration of presenting symptoms of up to 20 years with initial recurrences ranging from 2 months to 13 years after the first operation.70 In Dropcho and Allens’ series of 25 patients with primary intracranial RMS, the median survival recorded from the time of diagnosis was 7 months.88 Eighteen of these patients undergoing at least partial tumor resection with or without other therapy had a median survival of 10 months. Namba et al. observed that untreated patients with this tumor type usually died within a period of a few weeks to 5 months.88
SUMMARY Primary meningeal sarcomas are a rare and heterogeneous group of mesenchymal tumors that should be considered in the differential diagnosis when encountering an aggressive dural neoplasm in a patient of any age. Upon histologic confirmation of the diagnosis, radiographic staging and cytologic examination of the cerebrospinal fluid should be strongly considered. Multimodal treatment includes maximum feasible resection, involved field radiation treatment, and chemotherapy. Although the clinical course of certain “well-differentiated” sarcomas has been favorable, survival in general is poor.
REFERENCES 1. Enzinger FM, Weiss WS. Soft Tissue Tumors. St Louis, Missouri: CV Mosby, 1988: 1 – 16. 2. Brennan MF, Casper ES, Harrison LB. Soft tissue sarcoma. In DeVita VT, Hellman S, Rosenberg SA (eds) Principles of Oncology. New York: Lippincott-Raven, 1997: 1738 – 1788. 3. Scheithauer BW. Tumors of the meninges: proposed modifications of the World Health Organization classification. Acta Neuropathol 1990; 80: 343 – 54. 4. Zulch KJ. Brain Tumors. Their Biology and Pathology, 3rd ed. Berlin, Germany: Springer-Verlag, 1986: 383 – 393. 5. Weed LH. The meninges, with special references to the cell coverings of the leptomeninges. In Penfield W (ed) Cytology and Cellular Pathology of the Nervous System. New York: Paul B Hoeber, 1932, Vol. 2: 613 – 634. 6. Lopez de Faria J. Rhabdomyosarcoma of the cerebellum. Acta Pathol 1957; 63: 234 – 8. 7. Onofrio BM, Kernohan JW, Uihlein A. Primary meningeal sarcomatosis. A review of the literature and report of 12 cases. Cancer 1962; 15: 1197 – 208. 8. Kernohan JW, Uihlein A. Sarcomas of the Brain. Springfield, Illinois: Charles C Thomas, 1962. 9. Russell DS, Rubinstein LJ. Pathology of Tumours of the Nervous System, 5th ed. Baltimore, Maryland: Williams & Wilkins, 1989: 507 – 517. 10. Nichols P, Wagner JA Jr. Primary intracranial sarcoma; report of 9 cases with suggested classification. J Neuropathol Exp Neurol 1952; 11: 215 – 34. 11. Abbot KH, Kernohan JW. Primary sarcomas of the brain. Arch Neurol Psychiatr 1943; 50: 43. 12. Christensen E, Lara DE. Intracranial sarcomas. J Neuropathol Exp Neurol 1953; 12: 41 – 56.
636
NEUROLOGICAL MALIGNANCIES
13. Paulus WP, Slowik F, Jellinger K. Primary intracranial sarcomas: histopathological features of 19 cases. Histopathology 1991; 18: 395 – 402. 14. Rubinstein LJ. Tumors of the central nervous system. In Atlas of Tumor Pathology, Series 2, Fascicle 6. Washington, District of Columbia: Armed Forces Institute of Pathology, 1972: 190 – 204. 15. Tomita T, Gonzales-Crussi F. Intracranial primary nonlymphomatous sarcomas in children: experience with eight cases and review of the literature. Neurosurgery 1984; 14: 529 – 40. 16. Jellinger K, Sunder-Plasman H. Connatal intracranial tumors. Neuropadiatrie 1973; 4: 46 – 63. 17. Fessard C. Cerebral tumors in infancy. 66 clinico-anatomical case studies. Am J Dis Child 1968; 115: 302 – 98. 18. Hope JKA, et al. Primary meningeal tumors in children: correlation of clinical CT findings with histologic type and prognosis. AJNR Am J Neuroradiol 1992; 13: 1353 – 64. 19. Nakamura Y, Becker LE. Meningeal tumors of infancy and childhood. Pediatr Pathol 1985; 3(2 – 4): 341 – 58. 20. Takaku A, et al. Brain tumor in newborn babies. Childs Brain 1978; 4: 365 – 75. 21. Davidson GS, Hope JK. Meningeal tumors of childhood. Cancer 1989; 63: 1205 – 10. 22. Raimondi AJ, Tomita T. Brain tumors during the first year of life. Childs Brain 1983; 10: 193 – 207. 23. Crouse SK, Berg BO. Intracranial meningiomas in childhood and adolescence. Neurology 1972; 22: 135 – 41. 24. Cooper M, Dohn DF. Intracranial meningiomas in childhood. Cleve Clin Q 1974; 41: 197 – 204. 25. van Vliet MAT, et al. Congenital meningeal sarcoma – a case report. J Perinat Med 1983; 11: 249 – 53. 26. Reigh EE, Decker JT. Meningeal sarcoma in a two-week-old infant simulating hydrocephalus. J Neurosurg 1962; 19: 427 – 30. 27. Tarlov IM, Keener EB. Subarachnoid hemorrhage and tumor implants from spinal sarcoma in an infant. Neurology 1953; 3: 384. 28. Zwartverwer FL, et al. Meningeal sarcoma of the spinal cord in a newborn. Arch Neurol 1978; 35: 844. 29. Malat J, et al. Primary intraspinal fibrosarcoma. Neurosurgery 1986; 19: 434 – 6. 30. Zimmerman HM, Arnold H. Experimental brain tumors I. Tumors produced with methylcholanthrene. Cancer Res 1941; I: 919 – 38. 31. V´azquez-L´opez E. On the growth of the Rous sarcoma inoculated into the brain. Am J Cancer 1936; 26: 29 – 55. 32. Rabson AS, Kirschstin RL. Intracranial sarcomas produced by polyoma virus in Syrian hamsters. Arch Pathol 1960; 69: 663 – 71. 33. Kumar PP, et al. Radiation induced neoplasms of the brain. Cancer 1987; 59: 1274. 34. Shi T, Farrell MA, Kaufmann JCG. Fibrosarcomas complicating irradiated pituitary adenoma. Surg Neurol 1984; 22: 277 – 83. 35. Gonzalez-Vitale JC, Slavin RE, McQueen JO. Radiation-induced intracranial malignant fibrous histiocytoma. Cancer 1976; 37: 2960 – 3. 36. Gainer JV, Chou SM, Chadduck WM. Familial cerebral sarcomas. Arch Neurol 1975; 32: 665 – 8. 37. Biegel JA, et al. Monosomy 22 in rhabdoid atypical tumors of the brain. J Neurosurg 1990; 73: 710 – 4. 38. Savitz MH, et al. Fibrosarcoma of the spinal meninges in a case of neurofibromatosis. Mt Sinai J Med 1982; 49: 344 – 8. 39. Ferracini R, et al. Meningeal sarcoma with rhabdomyoblastic differentiation: case report. Neurosurgery 1992; 30: 782 – 5. 40. Ho YS, et al. Intracerebral malignant fibrous histiocytoma: case report and review of the literature. Neurosurgery 1992; 311: 567 – 71. 41. Kristoferitsch W, Jellinger K. Multifocal spinal angiosarcoma after chordotomy. Acta Neurochir (Wien) 1986; 79: 145 – 53. 42. Haddad GF, Al-Mefty O. Meningeal sarcomas. In Kaye AH, Laws ER (eds) Brain Tumors. An Encyclopedic Approach. New York: Churchill Livingstone, 1997: 713 – 721. 43. Lee Y-Y, VanTassel P, Raymond AK. Intracranial dural chondrosarcoma. AJNR Am J Neurorad 1988; 9: 1189 – 93. 44. Cybulski GR, et al. Falcine chondrosarcoma: case report and literature review. Neurosurgery 1985; 16: 412 – 5. 45. Sano K, et al. Characteristics of intracranial meningiomas in childhood. Childs Brain 1981; 8: 98 – 106.
46. Bailey OT, Ingraham FD. Intracranial fibrosarcomas of the dura mater in childhood: pathological characteristics and surgical management. J Neurosurg 1945; 2: 1 – 15. 47. Kothandaram P. Dural liposarcoma associated with subdural hematoma. Case report. J Neurosurg 1970; 33: 85 – 7. 48. Nussbaum ES, et al. Meningeal sarcoma mimicking an acute subdural hematoma on CT. J Comput Assist Tomogr 1995; 19: 643 – 5. 49. Tzonos T, Kraus B. Subduraler Erguss als Folge eines diffusen Leptomeningealsarkoms. Neurochirurgia 1972; 6: 227 – 31. 50. Cinalli G, et al. Subdural sarcoma associated with chronic subdural hematoma. J Neurosurg 1997; 86: 553 – 7. 51. Abhyanka SC, et al. Massive osteolysis of skull associated with meningeal sarcoma. J Indian Med Assoc 1985; 83(2): 67 – 9. 52. Guthrie BL, Ebersold MJ, Scheithauer BW. Neoplasms of the intracranial meninges. In Youmans JR (ed) Neurological Surgery, 3rd ed. Philadelphia, Pennsylvania: W. B. Saunders, 1990, Vol 5: 3250 – 3315. 53. El-Gindi S, Abd-El-Hafeez M, Sallanma M. Extracranial skeletal metastases from an intracranial meningeal chondrosarcoma: case report. J Neurosurg 1974; 40: 651 – 3. 54. Latchaw RE, Gabrielsen TO, Seeger JF. Cerebral angiography in meningeal sarcomatosis and carcinomatosis. Neuroradiology 1974; 8: 131 – 9. 55. Burger PC, Scheithauer BW, Vogel FS (eds) Surgical Pathology of the Nervous System and its Coverings, 3rd ed. New York: Churchill Livinstone, 1991: 112 – 123, 129 – 139. 56. Graham DI, Lantos PL (eds) Greenfield’s Neuropathology, 6th ed. New York: Oxford University Press, 1997: 748 – 749. 57. Reynier Y, et al. Meningeal fibrosarcomas. Neurochirurgie 1984; 30: 1 – 10. 58. Kolluri VRS, et al. Meningiomas in childhood. Childs Nerv Syst 1987; 3: 271 – 3. 59. Ironside JW. Classification of primary intracranial sarcomas and other central nervous system neoplasms. Histopathology 1991; 18: 483 – 6. 60. Leibel SA, et al. Soft tissue sarcomas of the extremities: survival patterns of failure with conservative surgery and postoperative irradiation compared to surgery alone. Cancer 1982; 50: 1076 – 83. 61. Khan WA, et al. Cytodiagnosis of a meningeal fibrosarcoma metastatic to the thyroid gland. Semin Diagn Pathol 2001; 18(2): 104 – 9. 62. Aung TH, Tse CH. Bifrontal meningeal fibrosarcoma in a patient with metastases to the liver, kidneys and suprarenal glands. Aust N Z J Surg 1993; 63: 746 – 8. 63. Caner GC, et al. Cabot case 24312. N Engl J Med 1938; 219: 169 – 73. 64. Campbell AN, et al. Extracranial metastases in childhood primary intracranial tumors. A report of 21 cases and review of the literature. Cancer 1984; 53: 974 – 81. 65. Henry JM, Leestma JE. Astrocytoma arising in meningeal fibrosarcoma. Acta Neuropathol 1973; 23: 334. 66. Lalitha VS, Rubinstein LJ. Reactive glioma in intracranial sarcoma: a form of mixed sarcoma and glioma (sarcoglioma). Report of eight cases. Cancer 1979; 43: 246 – 57. 67. Lichtenstein L, Bernstein D. Unusual benign and malignant chondroid tumors of bone. Cancer 1959; 12: 1142 – 57. 68. Dowling EA. Mesenchymal chondrosarcoma. J Bone Joint Surg Am 1964; 46: 747 – 54. 69. Forbes RB, Eljamel MS. Meningeal chondrosarcomas, a review of 31 patients. Br J Neurosurg 1998; 12(5): 461 – 4. 70. Hassounah M, et al. Primary cranial and intracranial chondrosarcoma: a survey. Acta Neurochir 1985; 87: 123 – 32. 71. Harsh GR IV, Wilson CB. Central nervous system mesenchymal chondrosarcoma. J Neurosurg 1984; 61: 375. 72. Heros RC, Martinez AJ, Ahn HS. Intracranial mesenchymal chondrosarcoma. Surg Neurol 1980; 14: 311 – 7. 73. Scheithauer BW, Rubinstein LJ. Meningeal mesenchymal chondrosarcoma: report of eight cases with review of the literature. Cancer 1978; 42: 2744 – 52. 74. Platania N, et al. Spinal meningeal mesenchymal chondrosarcoma. Report of a new case and review of the literature. J Neurosurg Sci 2003; 47(2): 107 – 10. 75. Rollo JL, Green WR, Kahn LB. Primary meningeal mesenchymal chondrosarcoma. Arch Pathol Lab Med 1979; 103: 239 – 43.
MENINGEAL SARCOMAS 76. Aucker KK, Horoupian DS. Dural mesenchymal chondrosarcoma: case report. J Neurosurg 1978; 48: 829 – 33. 77. Berkmen YM, Blatt ES. Cranial and intracranial cartilaginous tumors. Clin Radiol 1968; 19: 327 – 33. 78. Bergmann M, et al. Chondroid tumors arising from the meninges – report of 2 cases and review of the literature. Clin Neuropathol 2004; 23(4): 149 – 53. 79. Oruckaptan HH, et al. Parafalcine chondrosarcoma: an unusual localization for a classical variant. Case report and review of the literature. Surg Neurol 2001; 55(3): 174 – 9. 80. Bosma JJ, et al. Primary intradural classic chondrosarcoma: case report and literature review. Neurosurgery 2001; 48(2): 420 – 3. 81. Waga S, Masushima M, Ando K. Intracranial chondrosarcoma with extracranial metastases: case report. J Neurosurg 1972; 36: 790 – 4. 82. Grossman RI, Davis KR. Cranial computed tomographic appearance of chondrosarcoma of the base of the skull. Radiology 1981; 141: 403 – 8. 83. Bahr AC, Gayler BW. Cranial chondrosarcomas: report of four cases and review of the literature. Radiology 1977; 124: 151 – 6. 84. Smith TW, Davidson RI. Primary meningeal myxochondrosarcoma presenting as a cerebellar mass: case report. Neurosurgery 1981; 8: 577 – 81. 85. Katayama Y, et al. Meningeal chondrosarcomatous tumor associated with meningocytic differentiation. Surg Neurol 1987; 28: 375 – 80. 86. Nagata S, Sawada K, Kitamura K. Chondrosarcoma arising from the falx cerebri. Surg Neurol 1986; 25: 505 – 9. 87. Green DM, Tarbell NJ, Shamberger RC. Solid tumors of childhood. In DeVita VT, Hellman S, Rosenberg SA (eds) Principles of Oncology. New York: Lippincott-Raven, 1997: 2091 – 2130. 88. Dropcho EJ, Allen JC. Primary intracranial rhabdomyosarcoma: case report and review of the literature. J Neurooncol 1987; 5: 139 – 50. 89. Namba K, et al. Primary rhabdomyosarcoma of the tentorium with peculiar angiographic findings. Surg Neurol 1979; 11: 39 – 43. 90. Korinthenberg R, et al. Primary rhabdomyosarcoma of the leptomeninx. Clinical, neuroradiological and pathological aspects. Clin Neurol Neurosurg 1984; 86: 301 – 5. 91. Le Pessot F, et al. A case of primitive meningeal rhabdomyosarcoma. Histological, immunohistochemical and ultrastructural study. Ann Pathol 2000; 20(4): 353 – 6. 92. Smith MJ, Armbrustmacher MV, Violett TW. Diffuse meningeal rhabdomyosarcoma. Cancer 1981; 43: 2081 – 6. 93. Legier JF, Wells HA. Primary cerebellar rhabdomyosarcoma. Case report. J Neurosurg 1967; 26: 436 – 8. 94. Asai A, et al. Primary leiomyosarcoma of the dura mater. J Neurosurg 1988; 68: 308 – 11. 95. Louis D, et al. Primary intracranial leiomyosarcoma. J Neurosurg 1989; 71: 417 – 22. 96. Lee TT, Page LK. Primary cerebral leiomyosarcoma. Clin Neurol Neurosurg 1997; 99(3): 210 – 22. 97. Sugita Y, et al. Primary meningeal sarcomas with leiomyoblastic differentiation: a proposal for a new subtype of primary meningeal sarcomas. Am J Surg Pathol 2000; 24(9): 1273 – 8. 98. Bejjani GK, et al. Primary dural leiomyosarcoma in a patient infected with human immunodeficiency virus: case report. Neurosurgery 1999; 44(1): 199 – 202. 99. Zevallos-Giampietri EA, et al. Primary meningeal Epstein-Barr virus-related leiomyosarcoma in a man infected with human immunodeficiency virus: review of literature, emphasizing the differential diagnosis and pathogenesis. Appl Immunohistochem Mol Morphol 2004; 12(4): 387 – 91. 100. Black BK, Kernohan JW. Primary diffuse tumors of the meninges (so-called meningeal meningiomatosis). Cancer 1950; 3: 805 – 19. 101. Weglewski A, et al. Primary leptomeningeal sarcomatosis. Case report. Neurol Neurochir Pol 2003; 37(1): 251 – 8. 102. de Oliveira MJ, et al. Primary diffuse leptomeningeal sarcoma with rhabdomyoblastic differentiation. A case report and immunohistochemical study. J Neurol Sci 2004; 221(1 – 2): 79 – 82.
637
103. Uluc K, et al. Primary leptomeningeal sarcomatosis; a pathology proven case with challenging MRI and clinical findings. J Neurooncol 2004; 66(3): 307 – 12. 104. Xu F, et al. Primary meningeal rhabdomyosarcoma in a child with hypomelanosis of Ito. Arch Pathol Lab Med 2000; 124(5): 762 – 5. 105. Bailey P, Bucy PC. The origin and nature of meningeal tumors. Am J Cancer 1931; 15: 15 – 54. 106. Swamy KSN, Shankar SK, Asha T. Malignant fibrous histiocytoma arising from the meninges of the posterior fossa. Surg Neurol 1986; 25: 18 – 24. 107. Helle TL, Hanbery JW, Becker DH. Meningneal malignant fibrous histiocytoma arising from a thoracolumbar myelomeningocele: case report. J Neurosurg 1986; 58: 185 – 9. 108. Kalyanaraman UP, et al. Malignant fibrous histiocytoma of the meninges. Histological, ultrastructural, and immunocytochemical studies. J Neurosurg 1981; 55: 957. 109. Lam RM, Colah SA. Atypical fibrous histiocytoma with myxoid stroma. A rare lesion arising from dura mater of the brain. Cancer 1979; 43: 237. 110. Mena H, Garcia JH. Primary brain sarcomas. Light and electron microscopic features. Cancer 1978; 42: 1298 – 307. 111. Sima A, Kindblom LG, Pellettieri L. Liposarcoma of the meninges. Acta Pathol Microbiol Scand 1976; A84: 306 – 10. 112. Lam RMY, Malik GM, Chason JL. Osteosarcoma of meninges. Am J Surg Pathol 1981; 5: 203 – 8. 113. Setzer M, et al. Primary meningeal osteosarcoma: case report and review of the literature. Neurosurgery 2002; 51(2): 488 – 92; discussion 492. 114. Osipov V, et al. Post-radiation dedifferentiation of meningioma into osteosarcoma. BMC Cancer 2002; 2(1): 34. 115. Gaspar LE, et al. Primary cerebral fibrosarcomas: clinicopathologic study and review of the literature. Cancer 1993; 72: 3277 – 81. 116. Suit HD, et al. Preoperative, intraoperative, and postoperative radiation in the treatment of primary soft tissue sarcoma. Cancer 1985; 55: 2659 – 67. 117. Hoshino M, et al. A case of intracranial mesenchymal chondrosarcoma – changes observed by computed tomography before and after radiotherapy. No Shinkei Geka 1981; 9: 843 – 8. 118. Huvos AG, et al. Mesenchymal chondrosarcoma. A clinicopathologic analysis of 35 patients with emphasis on treatment. Cancer 1983; 51: 1230 – 7. 119. Suit HD, et al. Definitive radiation therapy for chordoma and chondrosarcoma of base of skull and cervical spine. J Neurosurg 1982; 56: 377 – 85. 120. Tiberin P, et al. Brain sarcoma of meningeal origin after cranial irradiation in childhood acute lymphocytic leukemia. J Neurosurg 1984; 61: 772 – 6. 121. Brada M, et al. Risk of second brain tumor after conservative surgery and radiotherapy for pituitary adenoma. Br Med J 1992; 304: 1343 – 6. 122. Bojsen-Moller M, Knudsen V. Radiation-induced meningeal sarcoma. A case report with a review of the literature. Acta Neurochir 1977; 37: 147 – 52. 123. Stechschulte SU, et al. Primary meningeal extraosseous Ewing’s sarcoma: case report. Neurosurgery 1994; 35(1): 143 – 7. 124. Globus JH, Levin S, Sheps JG. Primary sarcomatous meningioma. J Neuropathol Exp Neurol 1944; 3: 311.
FURTHER READING Meazza C, et al. Evolving treatment strategies for parameningeal rhabdomyosarcoma: the experience of the Istituto Nazionale Tumori of Milan. Head Neck 2005; 27(1): 49 – 57. Michalski JM, et al. Influence of radiation therapy parameters on outcome in children treated with radiation therapy for localized parameningeal rhabdomyosarcoma in Intergroup Rhabdomyosarcoma Study Group trials II through IV. Int J Radiat Oncol Biol Phys 2004; 59(4): 1027 – 38.
Section 10 : Neurological Malignancies
58
Atypical and Malignant Meningiomas Samer E. Kaba and Athanassios P. Kyritsis
INTRODUCTION AND DEFINITION Meningioma is one of the most common intracranial tumors. It originates from the meninges of the skull or spinal canal, and compresses the adjacent neural tissues causing a multitude of symptoms and signs, depending on the location of the tumor. Larger tumors can produce a remarkable degree of disability, and even mortality. Most meningiomas are histologically benign and potentially curable with surgical resection, but about 6% of meningiomas display malignant histological features, and are classified as atypical or anaplastic tumors. Most authors reserve the term “malignant meningioma” to describe tumors infiltrating the brain parenchyma. Brain invasion by itself, however, is not enough to classify the tumor as malignant unless it is associated with other cellular and structural features of malignancy. In this chapter we will use the World Health Organization (WHO) classification of meningiomas, which recognizes benign meningiomas (grade 1), atypical meningiomas (grade 2), and anaplastic meningiomas (grade 3). Also we will use the term “malignant meningioma” interchangeably with “anaplastic meningioma”, and refer to atypical and anaplastic meningiomas combined as “nonbenign meningiomas”. Surgical resection is still the mainstay of treatment of meningioma, and is associated with a high cure rate in benign meningiomas if the whole tumor is resected. Recurrence is very common in nonbenign meningiomas, even after total resection. In addition, total resection cannot be achieved in a large number of aggressive nonbenign meningiomas, where the recurrence rate reaches 100%. Radiation therapy was shown to be effective in delaying the recurrence in partially resected benign meningiomas and atypical/anaplastic meningiomas. Medical treatments such as hormonal therapy, interferon-α (INF-α), and hydroxyurea may have some role in the future management of recurrent unresectable meningiomas.
ANATOMY Meningiomas can be intracranial or spinal. Intracranial meningiomas are much more common and more likely to be atypical or malignant than the spinal ones. The most common locations of intracranial meningiomas are cerebral convexity, parasagittal, sphenoid wing, lateral ventricle, tentorium, and tuberculum sellae. Other less common locations include orbit, cerebellopontine angle, and olfactory groove (Table 3). Meningiomas in general are wellcircumscribed tumors attached to the dura. They compress the adjacent brain, but are usually separated from the underlying parenchyma, except for highly malignant tumors that invade the brain. Meningiomas of skull base, especially sphenoid wing ones, tend to grow in what is called en plaque pattern. These tumors grow within and expand the meninges without forming an exophytic mass pattern.1 The surface of these tumors can be smooth or nodular, depending on their location. Hyperostosis or erosion of the bone can be seen adjacent to meningiomas. While benign meningiomas can excavate and compress surrounding tissue, actual invasion of neural and vascular structure is associated with atypical and malignant variants.
EPIDEMIOLOGY AND BIOLOGY Epidemiology Meningiomas are relatively common tumors of the central nervous system, and the vast majority are histologically benign. Intracranial meningiomas constitute 18–20% of all primary brain tumors.2 The incidence of meningioma in the general population is about 2.5–6 per 100 000, while it reaches 10 per 100 000 in people older than 65 years of age.3 In a large series of 936 cases of intracranial meningiomas, 94.3% were histologically benign, 4.7% atypical, and 1% anaplastic (malignant).4 Similar figures were reported in a more recent series of meningioma patients, where 6.2% had atypical and 1.7% malignant tumors.5
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
ATYPICAL AND MALIGNANT MENINGIOMAS
The average age of patients with atypical and malignant meningiomas is 52 years, similar to benign meningiomas. A slight predominance of male gender was noted in most of the series in contrast with benign meningiomas, which occur more commonly in women. In Jaaskelainen’s series, 12% of men had histological anaplasia, while only 4% of the women did.3
Molecular Biology Meningiomas express a complex of active molecules, including sex hormone receptors and a variety of growth factors.5 It was noted for a long time that meningiomas are more common in women, they grow faster during pregnancy, and they appear to be associated with breast cancer in many cases.6 These observations prompted the investigation of the role of sex hormones in the development of meningioma. Progesterone receptors (PR) are expressed in 40–100% of meningiomas, and this expression increases in recurrent tumors and is thought to be mitogenic.7 PRs are localized to the nuclei and are transcriptionally active.8 It has been reported that atypical meningiomas are more often PR negative.9 Estrogen receptors (ER) are not consistently found in meningiomas and their expression rate ranges between 0 and 33% depending on the technique used.10 – 12 Therefore, it is unlikely that ER plays any important role in the development of meningiomas. More recent reports found androgen receptors (AR) in two-thirds of meningiomas, localized to the nuclei, suggesting an active role of these receptors in gene activation.13,14 Multiple in vitro studies showed the important role of growth factors in the development of meningiomas. EGF receptors are expressed in 85–100% of meningioma cultures taken from tumors.15,16 Weisman et al.17,18 reported a dose-dependent increase in DNA synthesis when EGF is added to meningioma cell cultures. Platelet-derived growth factor (PDGF) displayed a modulatory effect on EGF receptors, and was synergistic to EGF.19,20 Somatostatin receptors are usually found in the same tumors expressing EGF receptors, but their role in the development of meningioma is not clearly defined.21 PDGF receptors are expressed by both normal meningeal and meningioma cells, but csis/PDGF-2 gene is expressed only by meningioma cells.22 The coexpression of PDGF-2 and PDGF receptors suggests an autocrine loop that may contribute to the maintenance of these tumors. This hypothesis was corroborated by the work of Adams et al.,23 who showed that adding neutralizing antibody against PDGF abolished the stimulation of meningioma cell growth by meningioma-conditioned medium. Furthermore, meningioma cell proliferation in culture was inhibited by PDGF antagonist.24 High levels of basic fibroblast growth factor (bFGF) are found in meningioma cells.25 This growth factor is a potent mitogen and angiogenic agent and it is highly likely that it plays an important role in the development of meningioma.26 Dopamine and endothelin receptors were also found in meningiomas but their role in the development and growth of these tumors is not clear yet.27,28 Finally, the blood vessels of malignant meningiomas express plasminogen activator inhibitor type I more strongly than
639
benign meningiomas, indicating that it may be associated with angiogenesis.29
PATHOLOGY AND CLASSIFICATIONS The revised WHO classification of meningeal tumors distinguishes neoplasms from meningothelial origin, or meningiomas per se, from other nonmeningothelial tumors originating in the meninges (see Table 1). Meningothelial tumors include several benign variants, papillary meningioma, atypical meningioma, and anaplastic meningioma. Meningeal sarcomas and hemangiopericytoma, once classified as meningiomas, are now considered mesenchymal nonmeningothelial tumors.
Pathology Many different variants of meningiomas exist, with each variant having a distinguishing architecture and cellular appearance (see Table 1). Multiple types can coexist in the same tumor. Benign meningiomas, in general, display a uniform cellular appearance, whorling structures, and calcification. They have normal blood vessels, and no mitotic activity. A group of meningiomas display variable degrees of malignant features, including loss of architecture, increased cellularity, nuclear pleomorphism, mitotic figures, focal necrosis, and brain invasion. Depending on the abundance of malignant features, this group of tumors is further classified into atypical and anaplastic (malignant) meningiomas. None of Table 1 The WHO classification of tumors associated with the meninges.
Tumors of meningothelial cells
Nonmeningothelial tumors of the meninges
Meningioma
a. Meningothelia (syncytial) b. Transitional/mixed c. Fibrous (fibroblastic) d. Psammomatous e. Angiomatous f. Microcystic g. Secretory h. Lymphoplasmacytic rich i. Metaplastic variant (xanthomatous, myxoid, osseous, and so on)
Atypical meningioma
a. Variants of 1 a-i (above) b. Clear cell c. Chordoid
Anaplastic (malignant) meningioma
a. Variants of 1 a-i (above) b. Papillary c. Rhabdoid
Mesenchymal tumors
Benign neoplasms: a. Osteocartilaginous tumors b. Lipoma c. Fibrous histiocytoma Malignant neoplasms: a. Hemangiopericytoma b. Chondrosarcoma c. Mesenchymal chondrosarcoma d. Malignant fibrous histiocytoma e. Rhabdomyosarcoma f. Meningeal sarcomatosis
Tumors of uncertain origin
Hemangioblastoma
640
NEUROLOGICAL MALIGNANCIES
these features is enough by itself to establish the diagnosis, but rather the simultaneous presence of several features is needed.5 Invasion of bone, dura, or dural sinuses can be seen in benign meningiomas and is not by itself a marker of malignancy. It is noteworthy that the clinical and biological behavior of the tumor does not always correlate with the histological appearance, and seemingly less malignant tumors can progress more rapidly and invade surrounding tissues more aggressively than expected.
Atypical Meningioma (WHO Grade 2) This tumor is constituted of a patternless, sheetlike growth of an increased number of small cells with a high nuclear –cytoplasmic ratio and prominent nucleoli.5 A fair amount of mitotic activity can be seen. Focal necrosis and brain invasion are not usually seen in atypical meningioma.
Anaplastic (Malignant) Meningioma (WHO Grade 3) This tumor is highly malignant, exhibiting cellular anaplasia, frequent mitotic figures, and striking necrosis, on a patternless background. Frank extensive invasion of the brain is frequently seen. Proliferation markers, such as bromodeoxyuridine (BrdU), Ki67, and proliferating cell nuclear antigen (PCNA), are usually markedly elevated in atypical and malignant meningiomas compared with their benign counterpart, and can help in establishing the diagnosis (see further discussion under section “General Considerations”).30 – 32
Papillary Meningioma This is a group of meningiomas characterized by a papillary pattern intermixed with other more usual meningiomatous areas. The cells are usually uniform, long or tapered, and radiate toward blood vessels forming pseudorosettes. Although this tumor may not have the typical malignant features, it corresponds to WHO grades 2 or 3. Clinically, papillary meningiomas are usually aggressive and have the tendency to recur, invade the brain, and metastasize.33,34 These tumors are rare, and, unlike other meningiomas, can occur in young adults and children.
Rhabdoid Meningioma This is a new grade 3 meningioma according to the latest 2000 WHO classification, since in most cases it exhibits histological features of malignancy and follows an aggressive course. However, if the rhabdoid histological subtype is the only feature of the tumor without other signs of malignancy such as brain invasion and high proliferation rate, the expected behavior of the rhabdoid meningioma cannot be predicted.
ETIOLOGY Radiation The occurrence of meningioma following low-dose, mediumdose, and high-dose radiation therapy is well documented in the literature. For a tumor to be considered secondary to radiation, it has to fulfill three criteria: (i) arise in the
radiation field, (ii) develop many years after radiation, and (iii) appear histologically different from any primary tumor that existed before radiation. The overwhelming majority of radiation-induced meningiomas occurred after low-dose irradiation, less than 10 Gy, as had been used for treatment of tinea capitis.35 Radiation-induced meningiomas have the tendency to occur in younger patients, be atypical histologically, have a high bromodeoxyuridine (BUDR) labeling index, and to recur after resection.36 The latency between radiation and the development of these tumors ranges from 19.5 to 35 years depending on the patient’s age and radiation dose.37
Trauma There have been reports of increased recall of significant head trauma in meningioma patients,38 a finding that other case control studies failed to demonstrate.39 In addition, a large prospective study on patients with head trauma found no increased incidence of brain tumors after long periods of follow-up.40
Cytogenetics Both chromosomal loss and structural rearrangement are found in meningiomas. The most frequent chromosomal abnormality is chromosome 22 monosomy, which occurs in 70–80% of all meningiomas. Less often, a deletion of 22q occurs instead.10,41 The next most frequent abnormality is nonrandom loss or structural rearrangement of chromosome 14. In fact, loss of chromosome 14 is the most common chromosomal abnormality encountered in atypical and malignant meningiomas, suggesting a loss of a tumor suppressor gene located on the long arm of this chromosome.42 Other less frequent abnormalities reported are the loss of chromosomes 17 and Y43 in one study and the loss of chromosomes 8 and 1 in other studies.44,45 Structural rearrangement can also affect the same chromosomes occasionally. Hyperdiploidy of chromosome 22 is seen occasionally without loss of heterozygosity, suggesting microdeletions or inactivating mutations of the meningioma’s locus resulting in a similar effect caused by 22 monosomy.46 Because other chromosomal abnormalities are rarely found in the absence of 22 monosomy, this chromosome was implicated in the early stages of meningioma development. More recently, chromosome 1p was found to have frequent deletions of up to 17 possible tumor suppression genes, especially at 1p34 and 1p36.21,47 While histologically benign meningiomas show a normal karyotype or 22 monosomy, atypical and anaplastic meningiomas display more complex structural and numerical chromosomal abnormalities.19 Loss of heterozygosity of chromosome 10 was found in 1 of 2 atypical, 4 of 13 malignant, and 0 of 20 benign meningiomas. This loss of heterozygosity was not observed in tumors classified as malignant by invasive criteria only.20 These changes were shown to correlate with shorter time to recurrence and shorter survival.48 A higher incidence of loss of heterozygosity for three loci on chromosome 14q in atypical and malignant meningiomas compared with benign ones has been reported.24 These findings suggest that the progression of normal meningeal cells to a malignant phenotype is characterized genetically by an
ATYPICAL AND MALIGNANT MENINGIOMAS
early loss of a tumor suppressor gene on chromosome 22 followed by alterations of chromosomes 10 and 14, and other less common changes. Chromosome 7p loss is seen only in a small group of meningiomas in general, but is seen in the majority of radiation-induced meningiomas.49
Specific Gene Abnormalities in Meningiomas A presumed meningioma gene was localized on the long arm of chromosome 22, distal to the myoglobin locus, corresponding to the region 22q 12.3-qter.50,51 There was no difference noted in the cytogenetics of different histological variants of meningioma.52 The analysis of meningiomas associated with neurofibromatosis type II(NF2) revealed loss of genetic information in the same region of chromosome 22 involved in sporadic meningiomas. Meningioma locus, however, is distinct from the NF2 gene, despite the fact that NF2 gene is also mapped to the long arm of chromosome 22.53 The loss of expression of protein 4.1B has been observed in 60% of meningiomas regardless of the histological grade.54 A 6 to 8 fold increase in the expression of k-ras oncogene was observed in meningiomas with normal karyotypes of 22 monosomy.55 This increase in the expression of k-ras is probably secondary to the lack of activity of a tumor suppressor gene because it is not associated with any changes in k-ras gene itself. On the other hand, the locus of csis gene lies in the region of the putative meningioma locus of chromosome 22, and it was implicated in the pathogenesis of meningiomas.6,56 A 5- to 20-fold increase in the expression of c-myc has been observed in 63% of meningiomas, without amplification or rearrangement of its gene. The level of expression of c-myc was higher in cases with loss of heterozygosity of chromosome 22 than in those without, suggesting again the loss of a tumor suppressor gene on chromosome 22.57 Mutations of p53 gene are rare in benign meningiomas,58 but were found in malignant meningiomas, suggesting that p53 may be considered a marker for malignant transformation in meningiomas.59 In addition, the inactivation of p53 by p14ARF loss and MDM2mediated degradation of p53 may contribute to meningioma progression.60 Other changes in genes or gene products have been reported recently in meningioma cells including increased activity of telomerase (human telomerase reverse transcriptase (hTERT)), amplification of chromosome 17q23, and the modulation of meningioma cell growth in vitro by somatostatin and transforming growth factor (TGF-β).61
641
Table 2 The most common presenting symptoms of meningiomas (n = 193).56
Symptoms Paresis Headache Visual impairment Personality change/mental decline Ataxia Decreased level of consciousness Aphasia Focal seizure Generalized seizures
Malignant meningioma (%)
Benign meningioma (%)
43 36 29 21 21 14 14 14 7
19 36 16 22 15 7 10 15 19
Table 3 Specific neurological syndromes associated with meningiomas.
Tumor location
Syndrome
Findings
Tuberculum sellae
Chiasmal syndrome
Cavernous sinus
Cavernous syndrome
Sphenoidal wing
Superior orbital fissure syndrome
Olfactory groove
Foster Kennedy’s syndrome
Visual field defects (bitemporal hemianopsia, altitudinal hemianopsia, homonymous hemi- or quadrantanopsia), ± pituitary insufficiency Cranial nerves III, IV, VI, and V1 deficits ± Horner’s syndrome Ophthalmoplegia, exophthalmos, pain behind the eye Ipsilateral optic atrophy + contralateral papilledema, mental and personality changes, anosmia Cranial nerves VI, VII, and VIII deficits. Less commonly IX, X, and brain stem deficits
Cerebellopontine angle
making the diagnosis easier on a clinical basis. The most common examples are tuberculum sellae, olfactory groove, and cavernous meningiomas. The symptoms and signs of these syndromes, and the corresponding tumors, are illustrated in Table 3. No clinical features are specific for atypical or anaplastic meningiomas, but rapid progression of symptoms in the absence of intratumoral hemorrhage can suggest the malignant nature of the tumor.62
Diagnosis and Neuroimaging
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS
Computerized Tomography and Magnetic Resonance Imaging
Clinical Presentation
Atypical and anaplastic meningiomas can look identical to benign meningiomas on brain imaging. On nonenhanced CT scan, meningiomas appear as an isodense, well-demarcated mass. Magnetic resonance imaging (MRI) allows a more accurate evaluation of the size, location, and invasiveness of meningiomas, especially in the base of the skull. Meningiomas are usually isointense on both T1- and T2-weighted images. Occasionally they can be hypointense on T1 and hyperintense on T2.6 After the injection of contrast material, these tumors enhance homogeneously and intensely in most
The presenting symptoms of meningioma depend on the tumor’s location, size, and rate of growth. Small convexity meningiomas can be found incidentally on brain imaging studies, while those in the base of the skull tend to be symptomatic early in their course. The most common symptoms include headache, focal motor deficit, seizures, personality changes, and visual impairments. The common presenting symptoms of meningiomas are summarized in Table 2. Specific syndromes can result from tumors in certain locations,
642
NEUROLOGICAL MALIGNANCIES
of the cases. Atypical and malignant meningiomas are most commonly found in the convexity, parasagittal region, and in the sphenoid ridge. They can be multiple in about 20% of cases.56 Some imaging features suggest an atypical or anaplastic histology of the tumor. A nodular growth pattern, or what is called mushrooming, is seen in 67% of cases of atypical/anaplastic meningiomas.3,63,64 Indistinct margins and areas of hyperintensity on MRI were also reported.57 Heterogeneous enhancement on CT and MRI is a common feature in atypical/anaplastic meningiomas, seen in 40–50% of these tumors.3,57 Reactive vasogenic edema in the adjacent brain can be seen with all meningiomas, but is more common with higher grade meningiomas, where it is noted in up to 40% of tumors.57 Multiplicity, osteolysis, and soft tissue invasion are less consistent features of atypical and malignant meningiomas. These features may make the differentiation of nonbenign meningiomas and dural metastases difficult in some cases. Dynamic MRI can be a useful tool to distinguish between the two types of tumor before surgery.65 Angiography
The angiographic appearance of atypical and anaplastic meningiomas is similar to that of their benign counterpart. Most of these tumors derive their blood supply from a meningeal arterial branch.3 The need for standard angiography has declined rapidly with the availability of high resolution MRI scanning and MR angiography. However, angiography is still a useful tool when preoperative embolization of the tumor is planned.6 Positron-emission Tomography
Positron-emission tomography (PET) scanning in meningiomas has been studied rather extensively. Increased metabolic activity has been shown in all subtypes of meningiomas using quantitative and semiquantitative techniques.66,67 In addition, the tumor’s histopathology, aggressiveness, and proliferative rate were correlated with the metabolic rate on 2[18 F] fluoro-deoxy-D-glucose (FDG) PET.59,68,69 Preoperative FDG-PET scanning could distinguish benign meningiomas from higher grade ones with high specificity in one study.60 An increase in the metabolic rate on FDG-PET from baseline was reported in recurrent tumors.70 PET scanning, however, is still limited to research applications, and has no well-defined role in the clinical management of meningioma. Thallium 201 single photon emission computerized tomography (SPECT) scanning was useful in evaluating the aggressiveness and level of malignancy of meningioma in 59 lesions in 59 patients.71
gross total resection, and adjuvant therapy should be considered in the initial management of these tumors.
Surgery The goal of surgery in meningioma is to remove all the tumor mass and any dural attachment. This goal cannot be achieved in all cases of atypical and anaplastic meningioma.3 Tumors located at tuberculum sellae, cavernous sinus, and superior sagittal sinus are particularly difficult to resect completely because of their proximity to cranial nerves and major blood supply to the brain.72 The histological nature of the tumor cannot be predicted from the gross appearance during surgery, but atypical and anaplastic meningiomas are more likely to be soft, and adherent to the cortex. Intraoperative bleeding and duration of surgery were not different between benign and malignant meningiomas in a large series.3 The extent of surgical resection has been identified as one of the most important factors affecting the rate of recurrence and time to progression in meningioma. Simpson’s scale of surgical resection in meningioma is widely used because it correlates with clinical outcome (see Table 4). Younis et al.57 reported a median time to progression (MTP) of 35 and 5 months in totally and partially resected aggressive meningiomas, respectively. None of the patients who had partial resection was recurrence-free after 2 years. Jaaskelainen et al.3 reported recurrence rates of 38, 49, and 54% at 5, 10, and 15 years, respectively, in completely resected atypical meningiomas. The median time to recurrence was 2.4 years. Anaplastic meningioma recurred in 78% of the patients within 5 years, with a median time to recurrence of 3.5 years. In the same study, the recurrence rates of histologically benign meningioma at 5, 10, and 15 years were 3, 9, and 15%, respectively.3 In a more recent study, Palma et al.73 reported median times to recurrence of 5 and 2 years in atypical and malignant meningiomas. In this series, radical resection (Simpson grade 1) was correlated with a better clinical course, longer time to progression and longer survival.63 Similar findings were reported by other authors.74,75 Tumor location is another major factor in predicting tumor recurrence, with base of skull meningiomas being the most commonly recurring tumors, followed by sagittal and falx meningiomas, and finally the convexity meningiomas.63 The high rate of recurrence in malignant meningiomas can be explained at least partially by the microscopic infiltration of tumor cells in the dura and cerebral cortex.
Radiation Therapy The role of radiation therapy in the treatment of meningiomas in general has been under intense discussion over the past Table 4 Simpson’s grading of surgical resection in meningiomas.76
INITIAL MANAGEMENT OF ATYPICAL AND MALIGNANT MENINGIOMA
Grade 1
Surgical resection is still the mainstay of management, because total resection produces a high cure rate in benign meningiomas. Total resection, however, is not always possible, and residual tumors almost always recur. Atypical and anaplastic meningiomas have the tendency to recur even after
Grade 3
Grade 2
Grade 4 Grade 5
Gross total resection of tumor, with dural attachments and abnormal bone Gross total resection of tumor, coagulation of dural attachments Gross total resection of tumor, without resection or coagulation of dural attachments, or alternatively of its extradural extensions Partial resection of tumor Simple decompression or biopsy
ATYPICAL AND MALIGNANT MENINGIOMAS
decade. Most of the problem stems from the retrospective nature of most reports. Early studies showed conflicting results because they included different histological types of tumors, used a wide range of radiation sources and doses, and did not always use modern brain imaging to assess tumor progression. In more recent studies, however, a more uniform dose of radiation was used, brain imaging was used routinely to assess progression, and more strict classification criteria were applied. As a result, a longer time to progression was documented in patients treated with radiation, and the role of radiotherapy in the treatment of partially resected and recurrent meningiomas was established.77 – 79 The role of radiotherapy in the management of nonbenign meningiomas was even harder to evaluate because: (i) few studies addressed this group of patients separately, (ii) different studies used different criteria to define atypical and malignant meningiomas, and (iii) most studies included a certain number of meningeal sarcomas and hemangiopericytomas. Milosevic et al.80 reported the outcome of 59 cases of nonbenign meningiomas treated between 1966 and 1990 with adjuvant radiotherapy (see Table 5). This series included tumors designated as malignant meningioma based solely on brain invasion and/or hemangiopericytic features. The actuarial 5-year survival rate was 28%, and radiation doses higher than 50 Gy produced significantly longer survival.70 Similar observations were reported by Shimizu et al.,81 who found no difference between giving radiation as part of the initial therapy or at the time of recurrence. In a large series of radiotherapy-treated meningiomas 23 patients had malignant meningiomas. The 5-year progression-free survival rate of these patients was 48%, and the overall 5-year survival rate was 58%.67 Other authors reported less positive results of radiotherapy, but these studies had either a low number of cases or heterogeneous group of patients.3,57 More recently, Chamberlain82 reported the results of a prospective study using adjuvant radiotherapy and chemotherapy in 14 patients with malignant meningiomas. Gross total resection was achieved in only four patients, and a radiation dose of 59–60 Gy was given to all patients. In addition, all patients received CAV (cyclophosphamide, adriamycin, and vincristine) chemotherapy. The MTP was 4.6 years, and median survival was 5.3 years.72 Because the MTP reported
643
in this study is comparable to what was achieved by radiation alone, it is unlikely that chemotherapy had a major impact on the outcome. From this data, it is clear that adjuvant radiotherapy is beneficial in prolonging recurrence-free survival in patients with high-grade meningiomas. Unless a future well-controlled prospective study indicates otherwise, adjuvant radiotherapy should be given to all patients with atypical and anaplastic meningiomas, especially if a complete surgical resection (Simpson grade 1) could not be performed. Stereotactic radiotherapy delivered via γ -knife or linear accelerator is usually reserved for recurrent meningiomas. However this treatment modality can be considered an alternative to conventional radiation therapy for some deeply situated tumors, and even as an alternative to surgery for unresectable ones.83
Chemotherapy Although meningiomas were found to be sensitive to cytotoxic agents in vitro,84,85 few attempts have been made to evaluate the efficacy of adjuvant chemotherapy in high-grade malignant meningiomas. A report showed a relatively long progression-free survival in malignant meningioma patients treated with CAV after surgical resection and radiotherapy (see Table 5).83 Similar results, however, were observed when radiation was used alone. Based on the available evidence, chemotherapy cannot be recommended in the initial treatment of atypical and malignant meningiomas.
MANAGEMENT OF RECURRENT ATYPICAL AND MALIGNANT MENINGIOMAS General Considerations Recurrence is still the rule in atypical and malignant meningiomas. Recurrence can be recognized clinically, that is, the appearance of new symptoms and signs or the worsening of preexisting ones, or it can be seen on brain imaging before it produces clinical symptoms. In the latter case, definite tumor progression should be documented on serial studies before therapeutic intervention is planned, because the growth rate of meningiomas can vary over time. Recurrence is local in the
Table 5 Radiation therapy for atypical and malignant meningiomas.
Number of patients
Study 72a
Chamberlain Milosevic et al.70 Shimizu et al.71 Younis et al.57 Goldsmith et al.67 Jaaskelainen et al.3 Forbes and Goldberg69 a
14 59c 13 18c 23 5 4
Treatment: initial/recurrence Initial Both Both Both Initial Both Both
Prospective study. Cyclophosphamide + adriamycin + vincristine. Including patients with meningeal sarcomas and/or hemangiopericytomas. d Not all patients. e Not statistically different from nonirradiated group. b c
Median radiation dose (Gy) 60 50 54.4 52 54 59 52.8
Chemotherapy
Median time to progression (years)
Progression-free survival (5 year)
Yesb No No Yesd No No No
4.6 Not mentioned Not mentioned 2.5e 4.8 1.6 <3
Not mentioned 28% (total survival) 30% Not mentioned 48% Not mentioned Not mentioned
644
NEUROLOGICAL MALIGNANCIES
vast majority of the patients, but multiple intracranial lesions and even extracranial metastases have been reported.86,87 Fourteen percent of benign meningiomas transform into atypical or anaplastic histology upon recurrence, and 26% of atypical meningiomas transform into anaplastic meningiomas or even meningeal sarcoma.3,88 Certain factors are very important in determining the likelihood of, and the time of, recurrence of a given tumor. The most important factors are the extent of surgical resection and the location of the tumor, as mentioned above. Proliferation labeling indices have been used to predict recurrence in meningiomas. Hoshino et al.89 reported a clear correlation between BUDR labeling index (LI) and recurrence rate in meningiomas. In this study tumors with a BUDR LI ≥ 5% had a 100% recurrence rate, while tumors with an LI of 3–5% had a 55.6% recurrence rate. This observation was confirmed by other authors.90,91 PCNA LI was correlated with the tumor doubling time, and with resistance to radiation therapy.92,93 A PCNA LI >7% was associated with an almost fourfold increase in the rate of recurrence.9 A similar correlation with MIB-1 LI was found.8 We observed increasing BUDR LI in consecutive recurrences of a malignant meningioma.94
Surgery
same focused high-dose beam of radiation. Many authors reported favorable results using stereotactic radiosurgery in meningioma.96,97 Most reported series included a limited number of atypical and malignant meningiomas. Despite its promising results, stereotactic radiosurgery cannot be used in the majority of infiltrating skull-base meningiomas that are larger than 4 cm in diameter. Brachytherapy was also used in the treatment of meningiomas, with good results.98,99 Some of the reported series included histologically malignant meningiomas.100 The long-term results and side effects of these techniques are still not well characterized.
Chemotherapy Treatment with various chemotherapeutic agents has been tried for patients with recurrent, unresectable, previously irradiated meningiomas but has been largely ineffective.78,101 Scattered reports are available, however, on the efficacy of chemotherapy in individual cases.102 The experience of the MD Anderson Cancer Center using ifosfamide/mesna or dacarbazine/adriamycin regimens was not rewarding.103 Early reports suggested that hydroxyurea may have some efficacy against meningioma cells in vitro and in clinical settings.104,105 Larger subsequent studies suggested that although oral hydroxyurea has a stabilizing effect on benign recurrent or progressive meningiomas, it does not seem to be very effective against atypical meningiomas.106,107 Temozolomide failed to show any efficacy in a series of 16 patients with refractory recurrent meningiomas.108
Surgical resection is still the most effective treatment of recurrent meningiomas. With the availability of advanced brain imaging techniques, recurrences are diagnosed relatively early. Multiple surgical resections are not an unusual occurrence in patients with malignant meningiomas, especially those located at the base of the skull. Subsequent surgical resections are less likely to be more radical than the first one, because of the relentless infiltration of recurrent tumors into the surrounding neural and vascular structures. In addition, surgery produces fibrosis and other changes in the anatomy of the region rendering subsequent surgeries technically more difficult. Typically, the period to next recurrence becomes shorter after each resection. In a large series of meningiomas, the mean interval between the first and the second operations was 6 years, between second and third operations it was 3.4 years, and between fourth and fifth operations it was 1.6 years.95 Further surgery may not be feasible in a significant portion of cases because of the location of the tumor or the poor general medical condition of the patient.
Several factors suggest that sex hormones, that is, progesterone, estrogen, androgen, may play a role in stimulating the growth of meningiomas as mentioned above. Mifepristone (RU 486), an antiprogesterone agent, was shown to inhibit the growth of meningioma cells in vitro.109 Two small trials were conducted with this agent showed stabilization or minor decrease in the size of the tumor in about half of the patients.110,111 These preliminary positive results prompted the initiation of a large randomized trial to evaluate the efficacy of mifepristone in the treatment of recurrent meningiomas. Although the expression of ERs is less consistent, tamoxifen, an antiestrogen agent, induced transient stabilization or minor response in 9 of 10 patients with recurrent unresectable meningiomas.112
Radiation Therapy
Interferon-α
Evaluating the role of radiotherapy in the treatment of recurrent high-grade meningiomas is difficult because few studies dealt only with recurrent tumors (see Table 5). Considering currently available data, radiation therapy should be used after the first recurrence if it was not given initially. In most cases, however, the patient presents with further recurrence after receiving standard external beam radiation therapy, and in these cases other methods of radiation therapy may be used. Stereotactic radiosurgery has been used to treat unresectable meningiomas since the 1970s using cobalt (60 Co) γ -ray sources.6 Later development in radiation technology allowed the use of linear accelerators to deliver the
There is accumulating evidence, from in vitro and in vivo studies, suggesting that IFN-α may be active against meningiomas. Koper et al.113 reported that adding low concentrations of IFN-α to meningioma cell cultures produced a 70–100% inhibition of thymidine DNA incorporation. Bergstrom et al.114 observed an inhibitory effect of IFN-α on the metabolism of meningiomas as measured by PET, which was associated with decreased tumor growth rate in some patients. In addition, IFN-α has been shown to have a moderate antiangiogenic activity, which may be important in highly vascular tumors such as meningioma.115,116 WoberBingol et al.117 reported a marked effect of IFN-α-2b on one patient with meningioma.
Hormonal Therapy
ATYPICAL AND MALIGNANT MENINGIOMAS
645
Table 6 The effects of interferon-α-2b on atypical and malignant meningiomas.
Interferon-α treatment history
Previous therapy
Case
Histology
Number of surgical resections
1 2 3 4 5 6 7 8
Atypical Atypical Malignant Malignant Malignant Malignant Malignant Malignant
2 2 4 1 3 1 2 5
a b
Radiation therapy
Chemotherapy
Yes Yes Yes Yes Yes No Yes Yes
No No Yes No No No No No
RU 486
IFN daily dose (MU m−2 ) (per week)
Persistent side effects
Outcome
Duration of therapy (months)
No No No No Yes No No No
4×5 4×5 4×5 4×5 4×5 4×5 5×3 4×5
None None Fatigue None Leukopenia Fatigue Fatigue None
Lost to F/U Progressed Progresseda Progressed Progressed Stable Stableb Stable
14+ 2 12 + 12 N/A 12 13+ 2+ 6+
Tumor recurred only after stopping interferon, and was stable for another year after resuming IFN postoperatively. Interferon therapy stopped because of toxicity despite stable tumor.
We reported a positive response in five of six patients with recurrent unresectable meningiomas who were treated with recombinant INF-α-2b. Two of these meningiomas were histologically benign, one was atypical, and three were malignant.84 More patients have been treated with INF since this publication, eight of them with atypical or malignant tumors. Five of the eight patients experienced remissions lasting 6–14 months, and two are still on therapy. In two patients, the tumor continued to grow despite therapy, and one patient elected to stop the treatment after 2 months despite the observed stabilization of the tumor. The toxicity associated with prolonged use of IFN-α was moderate and generally well tolerated (Yung et al., private communication). These results are illustrated in Table 6. Other similar studies confirmed these findings. Muhr et al. reported stabilization in the size of the tumor in 9 of 12 patients. Three of the nine patients were followed for relatively long times (8, 8, and 4.5 years). Two of these patients are still stable on therapy.118 Furthermore, [11 C]-L-methionine PET scans done on these patients were able to predict responses and the suitability of patient for long-term treatment. These results suggest that IFN-α can produce long lasting remissions in rapidly progressing atypical and malignant meningioma, and may represent a valid therapeutic option for patients with recurrent or unresectable tumors.
AUTHORS’ RECOMMENDATIONS Patients of advanced age or poor surgical candidates with asymptomatic small benign meningiomas should be followed without therapy. In symptomatic patients complete surgical resection should be attempted which is often curative. For incompletely resected benign tumors, radiotherapy is recommended. For recurrent previously completely resected tumors reresection is recommended followed by radiotherapy. Radiotherapy could be administered either as conventional external beam irradiation or stereotactically. Atypical and malignant meningiomas should be maximally resected followed by adjuvant radiotherapy in an attempt to cure the disease. In rare instances of atypical meningiomas the radiotherapy can be postponed until recurrence, especially
if there are concerns about the side effects of the radiotherapy. INF-α or chemotherapy should be considered in all meningiomas (benign, atypical, or malignant) after multiple recurrences or when multifocal approaches and surgery or radiotherapy have failed or contraindicated.
EXTRACRANIAL METASTASES OF MENINGIOMA Extracranial metastases of meningiomas are well documented, occurring usually with local recurrence of the tumor.119 Metastases to all organ systems have been reported but the most common sites are lung (35%), bone (17%), liver (13%), and lymph nodes (11%).76 Dissemination of the tumor through the cerebrospinal fluid (CSF) can also occur, producing tumor depositions in the spinal cord, and other cerebral locations.66,120 Although most metastases are associated with histologically malignant meningiomas, occasional cases have been reported with atypical or even “benign” meningiomas.121,122 The rate of distant metastases ranged between 3 and 16% of atypical and malignant meningiomas3,57 and 13–41% of papillary meningiomas.12,123 There are no clear guidelines for the treatment of metastatic meningiomas. Radiotherapy, chemotherapy, and biological therapy can be used in different combinations depending on the individual circumstances. Long-term survival after distant metastases has been reported but is not the rule.124
PROGNOSIS Atypical and malignant meningiomas carry a poor prognosis for morbidity and survival. As discussed earlier in this chapter, most patients with high-grade meningiomas will experience multiple recurrences and death as a result of local invasion. The 5-year recurrence rates are 38 and 78% and median time to first recurrence is 2.4 and 3.5 years in atypical and malignant meningiomas, respectively.3 The recurrence rate in papillary meningioma is 59%.12 Most patients undergo multiple surgical resections and sustain variable degrees of neurological dysfunction. Loss of vision in one eye, ophthalmoplegia, seizures, and hemiparesis are a
646
NEUROLOGICAL MALIGNANCIES
few of the common sequelae of tumor invasion and repeated craniotomies. The overall survival of patients with nonbenign meningioma depends on the extent of surgical resection. In one series, the median survival of patients who had a total resection initially was 130 months (range 3–156), and that of those who had a partial resection was 46 months (range 5–46).57 With the advent of new radiation techniques and medical treatments, it is our hope that long-term control, and even cure, will be possible.
HEMANGIOPERICYTOMA The term “angioblastic meningioma” was widely used in the 1940s to describe a meningeal tumor that is highly vascularized and has a malignant aggressive course. Around the same time, a malignant and vascular soft tissue sarcoma, composed of cells resembling capillary pericytes, was described and called hemangiopericytoma. Later it was recognized that angioblastic meningiomas are simply hemangiopericytomas originating from the meninges.125,126 In the WHO classification, meningeal hemangiopericytomas are classified as mesenchymal nonmeningothelial tumors. Hemangiopericytomas are usually attached to the dura, not encapsulated, and sometimes invade the brain parenchyma. Histologically, they are composed of uniform polygonal cells with large hyperchromatic nuclei. These cells are arranged in sheets around abundant branching blood vessels. These tumors are highly vascular, and are often hemorrhagic. Mitotic figures, anaplastic cells, and focal necrosis are common features, while the typical meningothelial features are usually absent. Strong immunohistochemical staining for reticulin help distinguish these tumors from meningiomas.5 Meningeal hemangiopericytomas are rare, with an incidence of 2–4% of meningiomas.65 They occur more commonly in males in the third or fourth decade of life.127 These tumors occur in the convexity, tentorium, posterior fossa, and spinemimicking meningiomas in appearance and presentation. The period between presenting symptoms and diagnosis is typically shorter than for meningiomas. Radical surgical resection with amputation of all meningeal attachments is the treatment of choice. This is achieved, however, in only 50–67% of patients.114 Preoperative embolization is preferred because of the excessive bleeding associated with these tumors. Hemangiopericytomas have a great tendency to recur locally and metastasize after surgical resection. The recurrence rates are 65 and 76 in 5 and 10 years, respectively, with median recurrence-free survival of 40–50 months.114,128 Extraneural metastases occurred in 13, 33, and 64% of the patients 5, 10, and 15 years after diagnosis. The most common sites of metastases are bone, lung, and liver.114 Reviewing 118 cases reported until 1985, Schroder reported overall survival rates of 65, 45, and 15% at 5, 10, and 15 years, respectively.114 Postoperative radiotherapy, with doses higher than 50 Gy, resulted in clear and significant improvement in progressionfree and total survival times. In Guthrie’s series, the recurrence rates at 3 and 5 years were 30 and 50% in irradiated patients, and 50 and 100% in patients who did not
receive irradiation.114 Similar findings were reported by other authors.129,130 Therefore, involved field radiation therapy should be a part of the initial treatment of meningeal hemangiopericytomas. At recurrence, further surgical resection or stereotactic radiosurgery can be offered. Chemotherapy was not effective in most reported cases.131 – 133
REFERENCES 1. Lantos PL, Vandenberg SR, Kleihues P. Tumors of the nervous system. In Graham DI, Lantos PL (eds) Greenfield’s Neuropathology, 6th ed. London: Arnold, 1997: 583. 2. Central Brain Tumor Registry of the United States. CBTRUS First Annual Report, Chicago, IL, 1995. 3. Longstreth WT Jr, et al. Epidemiology of intracranial meningioma. Cancer 1993; 72: 639. 4. Jaaskelainen J, Haltia M, Servo A. Atypical and anaplastic meningiomas: radiology, surgery, radiotherapy, and outcome. Surg Neurol 1986; 25: 233. 5. Mahmood A, et al. Atypical and malignant meningiomas: a clinicopathological review. Neurosurgery 1993; 33: 955. 6. Bolger GB, et al. Chromosome translocation t(14;22) and oncogene (c-sis) variant in a pedigree with familial meningioma. N Engl J Med 1985; 312: 564. 7. Rubinstein AB, et al. Hormone receptors in initially excised versus recurrent intracranial meningiomas. J Neurosurg 1994; 81: 184. 8. Carroll RS, et al. Progesterone and glucocorticoid receptor activation in meningiomas. Neurosurgery 1995; 37: 92. 9. Brandis A, et al. Immunohistochemical detection of female sex hormone receptors in meningiomas: correlation with clinical and histological features. Neurosurgery 1993; 33: 212. 10. Poulsgard L, Schroder HD, Ronne M. Cytogenetic studies of 11 meningiomas and their clinical significance. II. Anticancer Res 1990; 10: 535. 11. Koehorst SG, et al. Detection of an oestrogen receptor-like protein in human meningiomas by band shift assay using a synthetic oestrogen responsive element (ERE). Br J Cancer 1993; 68: 290. 12. Koehorst SG, et al. Estrogen receptor variants mRNA in human meningiomas. Ann N Y Acad Sci 1993; 684: 222. 13. Carroll RS, et al. Androgen receptor expression in meningiomas. J Neurosurg 1995; 82: 453. 14. Maxwell M, et al. Expression of androgen and progesterone receptors in primary human meningiomas. J Neurosurg 1993; 78: 456. 15. Reubi JC, et al. Coincidence of EGF receptors and somatostatin receptors in meningiomas but inverse, differentiation-dependent relationship in glial tumors. Am J Pathol 1989; 134: 337. 16. Westphal M, Herrmann HD. Epidermal growth factor-receptors on cultured human meningioma cells. Acta Neurochir 1986; 83: 62. 17. Weisman AS, Raguet SS, Kelly PA. Characterization of the epidermal growth factor receptor in human meningioma. Cancer Res 1987; 47: 2172. 18. Weisman AS, Villemure JG, Kelly PA. Regulation of DNA synthesis and growth of cells derived from primary human meningiomas. Cancer Res 1986; 46: 2545. 19. Vagner-Capodano AM, et al. Correlation between cytogenetic and histopathological findings in 75 human meningiomas. Neurosurgery 1993; 32: 892. 20. Rempel SA, et al. Loss of heterozygosity for loci on chromosome 10 is associated with morphologically malignant meningioma progression. Cancer Res 1993; 53: 2386. 21. Sulman EP, White PS, Brodeur GM. Genomic annotation of the meningioma tumor suppressor locus on chromosome 1p34. Oncogene 2004; 23: 1014 – 20. 22. Maxwell M, et al. Human meningiomas co-express Platelet-derived Growth Factor (PDGF) and PDGF-receptor genes and their protein products. Int J Cancer 1990; 46: 16. 23. Adams EF, et al. Autocrine control of human meningioma proliferation: secretion of platelet-derived growth-factor-like molecules. Int J Cancer 1991; 49: 398.
ATYPICAL AND MALIGNANT MENINGIOMAS 24. Todo T, Adams EF, Fahlbusch R. Inhibitory effect of trapidil on human meningioma cell proliferation via interruption of autocrine growth stimulation. J Neurosurg 1993; 78: 463. 25. Takahashi JA, et al. Gene expression of fibroblast growth factors in human gliomas and meningiomas: demonstration of cellular source of basic fibroblast growth factor mRNA and peptide in tumor tissues. Proc Nat Acad Sci USA 1990; 87: 5710. 26. Jensen RL, et al. In vitro growth inhibition of growth factor-stimulated meningioma cells by calcium channel antagonists. Neurosurgery 1995; 36: 365. 27. Kitagawa N, et al. Expression of a functional endothelin (ETA) receptor in human meningiomas. J Neurosurg 1994; 80: 723. 28. Carroll RS, et al. Dopamine D1, dopamine D2, and prolactin receptor messenger ribonucleic acid expression by the polymerase chain reaction in human meningiomas. Neurosurgery 1996; 38: 367. 29. Kono S, et al. Immunohistochemical localization of plasminogen activator inhibitor type 1 in human brain tumors. J Neuropathol Exp Neurol 1994; 53: 256. 30. Hoshino T, et al. S-phase fraction of human brain tumors in situ measured by uptake of bromodeoxyuridine. Int J Cancer 1986; 38: 369. 31. Miyagami M, et al. Analysis of the proliferative potential of meningiomas with MIB-1 monoclonal antibodies. No to Shinkei 1996; 48: 39. 32. Cobb MA, et al. Significance of proliferating cell nuclear antigen in predicting recurrence of intracranial meningioma. J Neurosurg 1996; 84: 85. 33. Ludwin SK, Rubinstein LJ, Russell DS. Papillary meningioma: a malignant variant of meningioma. Cancer 1975; 36: 1363. 34. Pasquier B, et al. Papillary meningioma. Clinicopathologic study of seven cases and review of the literature. Cancer 1986; 58: 299. 35. Modan B, et al. Radiation-induced head and neck tumours. Lancet 1974; 1: 277. 36. Mack EE, Wilson CB. Meningiomas induced by high-dose cranial radiation. J Neurosurg 1993; 79: 28. 37. Ghim TT, et al. Childhood intracranial meningiomas after high-dose irradiation. Cancer 1993; 71: 4091. 38. Preston-Martin S, et al. Case-control study of intracranial meningiomas in women in Los Angeles County, California. J Natl Cancer Inst 1980; 65: 67. 39. Choi NW, Schuman LM, Gullen WH. Epidemiology of primary central nervous system neoplasms. II. Case-control study. Am J Epidemiol 1970; 91: 467. 40. Annegers JF, et al. Head trauma and subsequent brain tumors. Neurosurgery 1979; 4: 203. 41. Poulsgard L, Ronne M, Schroder HD. Cytogenetic studies of 19 meningiomas and their clinical significance. I. Anticancer Res 1989; 9: 109. 42. Menon AG, et al. Frequent loss of chromosome 14 in atypical and malignant meningioma: identification of a putative ‘tumor progression’ locus. Oncogene 1997; 14: 611. 43. Al Saadi A, et al. Cytogenetic studies of human brain tumors and their clinical significance. II. Meningioma. Cancer Genet Cytogenet 1987; 26: 127. 44. Rey JA, et al. Incidence and origin of dicentric chromosomes in cultured meningiomas. Cancer Genet Cytogenet 1988; 35: 55. 45. Rey JA, et al. Chromosomal involvement secondary to -22 in human meningiomas. Cancer Genet Cytogenet 1988; 33: 275. 46. Bello MJ, et al. Chromosome 22 heterozygosity is retained in most hyperdiploid and pseudodiploid meningiomas. Cancer Genet Cytogenet 1993; 66: 117. 47. Lomas J, et al. Methylation status of TP73 in meningiomas. Cancer Genet Cytogenet 2004; 148: 148 – 51. 48. Mihaila D, et al. Meningiomas: loss of heterozygosity on chromosome 10 and marker-specific correlations with grade, recurrence, and survival. Clin Cancer Res 2003; 9: 4443 – 51. 49. Rajcan-Separovic E, et al. Loss of 1p and 7p in radiation-induced meningiomas identified by comparative genomic hybridization. Cancer Genet Cytogenet 2003; 144: 6 – 11. 50. Dumanski JP, et al. Deletion mapping of a locus on human chromosome 22 involved in the oncogenesis of meningioma. Proc Nat Acad Sci USA 1987; 84: 9275.
647
51. Seizinger BR, et al. Molecular genetic approach to human meningioma: loss of genes on chromosome 22. Proc Nat Acad Sci USA 1987; 84: 5419. 52. Collins VP, Nordenskjold M, Dumanski JP. The molecular genetics of meningiomas. Brain Pathol 1990; 1: 19. 53. Pulst SM, et al. Familial meningioma is not allelic to neurofibromatosis 2. Neurology 1993; 43: 2096. 54. Gutmann DH, et al. Loss of DAL-1, a protein 4.1-related tumor suppressor, is an important early event in the pathogenesis of meningiomas. Hum Mol Genet 2000; 9: 1495 – 500. 55. Carstens C, et al. Human KRAS oncogene expression in meningioma. Cancer Lett 1988; 43: 37. 56. Smidt M, Kirsch I, Ratner L. Deletion of Alu sequences in the fifth c-sis intron in individuals with meningiomas. J Clin Invest 1990; 86: 1151. 57. Kazumoto K, et al. Enhanced expression of the sis and c-myc oncogenes in human meningiomas. J Neurosurg 1990; 72: 786. 58. Mashiyama S, et al. Detection of p53 gene mutations in human brain tumors by single-strand conformation polymorphism analysis of polymerase chain reaction products. Oncogene 1991; 6: 1313. 59. Wang JL, et al. Detection of TP53 gene mutation in human meningiomas: a study using immunohistochemistry, polymerase chain reaction/single-strand conformation polymorphism and DNA sequencing techniques on paraffin-embedded samples. Int J Cancer 1995; 64: 223. 60. Amatya VJ, Takeshima Y, Inai K. Methylation of p14(ARF) gene in meningiomas and its correlation to the p53 expression and mutation. Mod Pathol 2004; 17: 705 – 10. 61. Lusis E, Gutmann D. Meningioma: an update. Curr Opin Neurol 2004; 17: 687 – 92. 62. Rohringer M, et al. Incidence and clinicopathological features of meningioma. J Neurosurg 1989; 71: 665. 63. DeMonte F, al-Mefty O. Meningiomas. In Kaye AH, Laws ER (eds) Brain Tumors. New York: Churchill Livingstone, 1995: 675. 64. Younis GA, et al. Aggressive meningeal tumors: review of a series. J Neurosurg 1995; 82: 17. 65. Kremer S, et al. Contribution of dynamic contrast MR imaging to the differentiation between dural metastasis and meningioma. Neuroradiology 2004; 46(8): 642 – 8. 66. Nyberg G, et al. PET-methionine of skull base neuromas and meningiomas. Acta Otolaryngol 1997; 117: 482. 67. Di Chiro G, et al. Glucose utilization by intracranial meningiomas as an index of tumor aggressivity and probability of recurrence: a PET study. Radiology 1987; 164: 521. 68. Cremerius U, et al. Fasting improves discrimination of grade 1 and atypical or malignant meningioma in FDG-PET. J Nucl Med 1997; 38: 26. 69. Lippitz B, et al. PET-study of intracranial meningiomas: correlation with histopathology, cellularity and proliferation rate. Acta Neurochir Suppl 1996; 65: 108. 70. Shioya H, et al. Longitudinal analysis of glucose metabolism in recurrent meningioma. No to Shinkei 1994; 46: 1088. 71. Kinuya K, et al. Thallium-201 brain SPECT to diagnose aggressiveness of meningiomas. Ann Nucl Med 2003; 17(6): 463 – 7. 72. Philippon J, Cornu P. The recurrence of meningiomas. In al-Mefty O (ed) Meningiomas. New York: Raven Press, 1991: 87. 73. Palma L, et al. Long-term prognosis for atypical and malignant meningiomas: a study of 71 surgical cases. J Neurosurg 1997; 86: 793. 74. Mirimanoff RO, et al. Meningioma: analysis of recurrence and progression following neurosurgical resection. J Neurosurg 1985; 62: 18. 75. Thomas HG, Dolman CL, Berry K. Malignant meningioma: clinical and pathological features. J Neurosurg 1981; 55: 929. 76. Stoller JK, et al. Intracranial meningiomas metastatic to the lung. Cleve Clin J Med 1987; 54: 521. 77. Goldsmith BJ, et al. Postoperative irradiation for subtotally resected meningiomas. A retrospective analysis of 140 patients treated from 1967 to 1990 [published erratum appears in J Neurosurg 1994: 80(4):777]. J Neurosurg 1994; 80: 195. 78. Barbaro NM, et al. Radiation therapy in the treatment of partially resected meningiomas. Neurosurgery 1987; 20: 525.
648
NEUROLOGICAL MALIGNANCIES
79. Forbes AR, Goldberg ID. Radiation therapy in the treatment of meningioma: the Joint Center for Radiation Therapy experience 1970 to 1982. J Clin Oncol 1984; 2: 1139. 80. Milosevic MF, et al. Radiotherapy for atypical or malignant intracranial meningioma. Int J Radiat Oncol Biol Phys 1996; 34: 817. 81. Shimizu T, Iijima M, Tanaka Y. Radiotherapy for intracranial meningioma: special reference to malignant and high risk benign meningioma. Nippon Igaku Hoshaen Gakkai Zasshi 1995; 55: 1047. 82. Chamberlain MC. Adjuvant combined modality therapy for malignant meningiomas. J Neurosurg 1996; 84: 753. 83. Lunsford DL. Contemporary management of meningiomas: radiation therapy as an adjuvant and radiosurgery as an alternative to surgical removal? J Neurosurg 1994; 80: 187 – 90. 84. Taut FJ, Zeller WJ. In vitro chemotherapy of steroid receptor positive human meningioma low-passage primary cultures with nitrosoureamethionine-steroid conjugates. Clin Neuropharmacol 1996; 19: 520. 85. Tsuchida T, et al. Chemosensitivity of cultured meningiomas. Hum Cell 1995; 8: 155. 86. Russell T, Moss T. Metastasizing meningioma. Neurosurgery 1986; 19: 1028. 87. Philippon J, et al. Recurrent meningioma. Neurochirurgie 1986; 32(Suppl 1): 1. 88. LeMay DR, Bucci MN, Farhat SM. Malignant transformation of recurrent meningioma with pulmonary metastases. Surg Neurol 1989; 31: 365. 89. Hoshino T, et al. Proliferative potential of human meningiomas of the brain. A cell kinetics study with bromodeoxyuridine. Cancer 1986; 58: 1466. 90. Cho KG, et al. Prediction of tumor doubling time in recurrent meningiomas. Cell kinetics studies with bromodeoxyuridine labeling. J Neurosurg 1986; 65: 790. 91. Shibuya M, et al. Meningiomas: clinical implications of a high proliferative potential determined by bromodeoxyuridine labeling. Neurosurgery 1992; 30: 494. 92. Suwa T, et al. Invasive meningioma: a tumor with high proliferating and ‘recurrence’ potential. Acta Neurochir 1995; 136: 127. 93. Colvett KT, et al. High PCNA index in meningiomas resistant to radiation therapy. Int J Radiat Oncol Biol Phys 1997; 38: 463. 94. Kaba SE, et al. The treatment of recurrent unresectable and malignant meningiomas with interferon alpha-2b. Neurosurgery 1997; 40: 271. 95. Boker DK, Meurer H, Gullotta F. Recurring intracranial meningiomas. Evaluation of some factors predisposing for tumor recurrence. J Neurosurg Sci 1985; 29: 11. 96. Kondziolka D, et al. Gamma knife radiosurgery of meningiomas. Stereotact Funct Neurosurg 1991; 57: 11. 97. Kondziolka D, et al. Stereotactic radiosurgery of meningiomas. J Neurosurg 1991; 74: 552. 98. Kumar PP, et al. Brachytherapy: a viable alternative in the management of basal meningiomas. Neurosurgery 1991; 29: 676. 99. Vuorinen V, et al. Interstitial radiotherapy of 25 parasellar/clival meningiomas and 19 meningiomas in the elderly. Analysis of shortterm tolerance and responses. Acta Neurochir 1996; 138: 495. 100. Gutin PH, et al. Brachytherapy of recurrent tumors of the skull base and spine with iodine-125 sources. Neurosurgery 1987; 20: 938. 101. Stewart DJ, et al. Intraarterial cisplatin plus intravenous doxorubicin for inoperable recurrent meningiomas. JNO 1995; 24: 189. 102. Bernstein M, et al. Necrosis in a meningioma following systemic chemotherapy. Case report. J Neuro-surg 1994; 81: 284. 103. Kyritsis AP. Chemotherapy for meningiomas. JNO 1996; 29: 269. 104. Schrell UM, et al. Hydroxyurea for treatment of unresectable and recurrent meningiomas. I. Inhibition of primary human meningioma cells in culture and in meningioma transplants by induction of the apoptotic pathway. J Neurosurg 1997; 86: 845.
105. Schrell UM, et al. Hydroxyurea for treatment of unresectable and recurrent meningiomas. II. Decrease in the size of meningiomas in patients treated with hydroxyurea. J Neurosurg 1997; 86: 840. 106. Mason WP, et al. Stabilization of disease progression by hydroxyurea in patients with recurrent or unresectable meningioma. J Neurosurg 2002; 97: 341 – 6. 107. Newton HB, Slivka MA, Stevens C. Hydroxyurea chemotherapy for unresectable or residual meningioma. J Neurooncol 2000; 49: 165 – 70. 108. Chamberlain MC, Tsao-Wei DD, Groshen S. Temozolomide for treatment-resistant recurrent meningioma. Neurology 2004; 62: 1210 – 2. 109. Grunberg SM. Role of antiprogestational therapy for meningiomas. Hum Reprod 1994; 9(Suppl 1): 202. 110. Grunberg SM, et al. Treatment of unresectable meningiomas with the antiprogesterone agent mifepristone. J Neurosurg 1991; 74: 861. 111. Lamberts SW, et al. Mifepristone (RU 486) treatment of meningiomas. J Neurol Neurosurg Psychiatry 1992; 55: 486. 112. Goodwin JW, et al. A phase II evaluation of tamoxifen in unresectable or refractory meningiomas: a Southwest Oncology Group study. JNO 1993; 15: 75. 113. Koper JW, et al. Inhibition of the growth of cultured human meningioma cells by recombinant interferon-alpha. Eur J Cancer 1991; 27: 416. 114. Bergstrom M, et al. Modulation of tumor metabolism using hormonally acting drugs in patients with meningioma. J Endocrinol Invest 1989; 12(Suppl 2): 91. 115. Li VW, et al. Microvessel count and cerebral fluid basic fibroblast growth factor in children with brain tumours. Lancet 1994; 344: 82. 116. Folkman J, Klagsburn M. Angiogenic factors. Science 1987; 235: 442. 117. Wober-Bingol C, et al. Interferon-alfa-2b for meningioma. Lancet 1995; 345: 331. 118. Muhr C, et al. Meningioma treated with interferon-alpha, evaluated with [(11)C]-L-methionine positron emission tomography. Clin Cancer Res 2001; 7(8): 2269 – 76. 119. Salcman M. Malignant meningiomas. In al-Mefty O (ed) Meningiomas. New York: Raven Press, 1991: 75. 120. Kleinschmidt-DeMasters BK, Avakian JJ. Wallenberg syndrome caused by CSF metastasis from malignant intraventricular meningioma. Clin Neuropathol 1985; 4: 214. 121. Som PM, et al. ‘Benign’ metastasizing meningiomas. Am J Neuroradiol 1987; 8: 127. 122. Miller DC, et al. Benign metastasizing meningioma. Case report. J Neurosurg 1985; 62: 763. 123. Fukushima T, et al. Papillary meningioma with pulmonary metastasis. Case report. J Neurosurg 1989; 70: 478. 124. Wasserkrug R, Peyser E, Lichtig C. Extracranial bone metastases from intracranial meningiomas. Surg Neurol 1979; 12: 480. 125. Guthrie BL. Meningeal hemangiopericytomas. In Kaye AH, Laws ER (eds) Brain Tumors. New York: Churchill Livingstone, 1995: 705. 126. Begg CF, Garret R. Hemangiopericytoma occurring in the meninges. Cancer 1954; 7: 602. 127. Guthrie BL, et al. Meningeal hemangiopericytoma: histopathological features, treatment, and long-term follow-up of 44 cases. Neurosurgery 1989; 25: 514. 128. Schroder R, Firsching R, Kochanek S. Hemangiopericytoma of meninges. II. General and clinical data. Zentralbl Neurochir 1986; 47: 191. 129. Fukui M, et al. Irradiated meningiomas: a clinical evaluation. Acta Neurochir 1980; 54: 33. 130. Bastin KT, Mehta MP. Meningeal hemangiopericytoma: defining the role for radiation therapy. JNO 1992; 14: 277. 131. Borg MF, Benjamin CS. Haemangiopericytoma of the central nervous system. Australas Radiol 1995; 39: 36. 132. Staples JJ, et al. Hemangiopericytoma – the role of radiotherapy. Int J Radiat Oncol Biol Phys 1990; 19: 445. 133. Simpson D. The recurrence of intracranial meningiomas after surgical treatment. J Neurol Neurosurg Psychiatry 1957; 20: 22.
Section 10 : Neurological Malignancies
59
Primary Intracranial Germ Cell Tumors Jan Drappatz and Jay S. Loeffler
INTRODUCTION AND HISTORICAL BACKGROUND Intracranial germ cell tumors comprise a heterogeneous group of rare tumors that occur in children and adults. These tumors most commonly arise in the testes and ovaries, and the central nervous system is only one of several extragonadal sites of origin that also include the retroperitoneum, the sacrococcygeal region, the mediastinum, and rarely the nasopharynx.1 Germinomas comprise the majority of intracranial germ cell tumors.2 The other germ cell tumors, collectively termed nongerminomatous germ cell tumors, are thought to represent the malignant correlate of normal stages of embryonal development: the trophoblast (choriocarcinoma), the yolk sac endoderm (endodermal sinus tumor), the pluripotent stem cell of the embryo (embryonal carcinoma), and the embryonic differentiated cell (teratoma).3,4 The classification of intracranial germ cell tumors has not been uniform. In 1944, Russell initially noted a similarity between testicular germ cell tumors and several pineal tumors which she named atypical teratomas.5 Two years later, Friedman and Moore who examined 922 testicular tumors concluded that seminomas were tumors of primordial germ cells that should be referred to as germinomas.6 Subsequently, Friedman identified that the two-cell pattern pinealoma was in fact germinoma, resembling the morphology of the neonatal pineal gland.7 In 1953, Dixon and Moore published their hypothesis that germinoma and embryonal carcinoma which gives rise to choriocarcinoma and teratoma have their common origin in germ cells.8 The discovery of histologic resemblance between suprasellar and infundibular tumors, atypical teratomas of the pineal region, and certain mediastinal tumors led to the classification of all of these tumors as germ cell tumors.9,10 In the following years, Teilum established the endodermal sinus tumors (yolk sac tumors), and formulated the germ cell theory, that is, that germ cells give rise to germinomas and, along a different route, to embryonal carcinomas from which endodermal sinus tumors, choriocarcinomas, and teratomas
derive.11,12 The currently used histopathologic classification of intracranial germ cell tumors reflects the classification for testicular and ovarian germ cell tumors4 and has been adapted by the World Health Organization (WHO) classification of brain tumors (see Table 1). The WHO classification further subdivides intracranial germ cell tumors into germinomas, teratomas (mature, immature, malignant), embryonal carcinomas, endodermal sinus tumors (yolk sac tumors), choriocarcinomas, and the so-called mixed germinal tumors (tumors consisting of two or more of these tumor components).13
PATHOGENESIS It has been presumed that intracranial germ cell tumors originate from malignant transformation of germ cells misplaced during embryogenesis.14 This theory has been confirmed by chromosomal15 and imprinting analysis.16 While the etiology of germinoma is unknown, the relationship between the beginning of puberty and the increased incidence of intracranial germ cell tumors suggests a possible relation between increased neuroendocrine activity, and malignant de-differentiation in formerly inactive primordial germ cell remnants.2 It is uncertain why primordial germ cells endure in certain extragonadal midline structures including the pineal gland and suprasellar structures. Little is also known about the molecular mechanisms responsible for the development of these cells into neoplasms. Chromosomal abnormalities are thought to play a role in the development of intracranial germ cell tumors.17 – 19 While the majority of testicular germ cell tumors display a specific cytogenetic abnormalities of chromosome 12, and isochromosome 12p, these abnormalities are only infrequently seen in intracranial germ cell tumors.20,21 In contrast, an increased copy number of chromosome X was detected in 92% of 25 intracranial germ cell tumors, regardless of histological subtype.21 Along with the observed male predominance of intracranial germ cell tumors and the tendency in Klinefelter syndrome patients for these
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
650
NEUROLOGICAL MALIGNANCIES
Table 1 WHO classification of intracranial germ cell tumors.13
1. 2. 3. 4. 5.
6.
Germinomas Embryonal carcinoma Yolk sac tumor (endodermal sinus tumor) Choriocarcinoma Teratoma Immature Mature Teratoma with malignant transformation Mixed germ cell tumors
lesions, it suggests that relevant oncogenes reside on the sex chromosome. Genetic alterations in a tumor suppressor gene at the INK4a/ARF locus, whose gene product is known to have a major influence on p53-MDM2 interactions, were also found in 71% of 21 intracranial germ cell tumors, supporting that the p53-MDM2 gene interactions may be important in tumorigenesis of intracranial germ cell tumors.22
EPIDEMIOLOGY Intracranial germ cell tumors are extremely rare, accounting for only 1% of all brain tumors.23 The incidence of the different germ cell tumor subtypes varies among different age periods.16 Neonates are more commonly diagnosed with benign congenital teratomas. Endodermal sinus tumors and malignant teratomas dominate between the seventh month and third year of life. Other histological subtypes such as embryonal carcinomas and choriocarcinomas are unusual in this age-group, and germinomas almost never occur in infancy. In children up to 5 years, extragonadal germ cell tumors are more common than their gonadal counterparts and there is an equal sex distribution as concerns intracranial germ cell tumors. During adolescence and early adulthood, the incidence of gonadal germ cell tumors rises parallelly with the incidence of intracranial germinomas and they show a pronounced male predominance. Among the nongerminomatous tumors, choriocarcinomas, and embryonal carcinomas become more frequent while the incidence of teratomas is generally low.16,24 Germ cell tumors are overall rare tumors, accounting for 0.6% of all brain tumors.25 The incidence rates per 100 000 persons is 0.18 (0 to 19 years), and 0.10 (20 to 34 years) and 10-fold less for older adults (0.02).25 There is a markedly higher frequency in Japan, where germinomas account for up to 8% of all intracranial tumors.2,26 Overall, germinomas are the most common subgroup, comprising approximately 50 to 65% of all central nervous system (CNS) germ cell tumors. Jennings et al.2 retrospectively analyzed 389 primary intracranial germ cell tumors. The study cohort included 253 germinomas (65%), 70 teratomas (18%), 21 embryonal carcinomas (5%), 26 endodermal sinus tumors (7%), and 19 choriocarcinomas (5%). Nearly all primary intracranial germ cell tumors originated along an axis from the suprasellar regions (37%) to the pineal gland (48%). Germinomas preferentially involved the suprasellar region (57%), while the majority of nongerminomatous tumors (68%) occurred in the pineal region. In 5–10% of cases, the
suprasellar region and the pineal gland were involved simultaneously at the time of diagnosis, primarily in patients with germinomas. In contrast, Hoffman et al.27 and Ho and Liu28 made a conflicting observation. In their series, more germinomas were located in the pineal than the suprasellar region. In Jennings’ series, there was a definite male predominance for intracranial germ cell tumors: the ratio was 1.88 : 1 for germinomas and even greater for nongerminomatous germ cell tumors (3.25 : 1). In females, involvement of the suprasellar region was more common (75%), whereas among males, pineal involvement was more frequent (67%). The age span of patients with intracranial germ cell tumors ranged from newborn to 69 years and the peak incidence for both sexes was in early puberty.
PATHOLOGY Histologically, ultrastructurally, and histochemically, intracranial germ cell tumors are identical to those found in the testes and ovaries and other extragonadal sites29 – 32 and have been listed here in order of increasing malignancy. The CNS germinoma is identical to the testicular seminoma and ovarian dysgerminoma. These tumors consist of large, polygonal cells similar to primitive germ cells with clear cytoplasm and large, vesicular nuclei, prominent nucleoli and frequent mitoses. Scattered lymphocytes are generally present, representing an immune reaction directed at the tumor. Immunohistochemically, germinomas stain for placental alkaline phosphatase (PLAP). Infrequent syncytiotrophoblastic giant cells within the tumor stain with human choriogonadotropin. Germinomas may also be cytokeratin, epithelial membrane antigen, and vimentin positive.28,30 Among the nongerminomatous tumors, teratomas are divided into three subtypes according to the degree of tumor cell differentiation: mature and immature teratomas, and teratomas with malignant transformation. Mature teratomas contain elements derived from the three germ cell layers: endoderm, mesoderm, and ectoderm. They are often cystic, with an epithelial layer like the epidermis with its appendages and frequently include sebaceous fat, which has a characteristic appearance on magnetic resonance imaging (MRI). The embryonal carcinoma is extremely pleomorphic and shows a variety of patterns forming glands, tubules, and even primitive embryolike structures. Many mitotic figures are present and foci of hemorrhage or necrosis can be seen. Diagnostic features include Schiller-Duval bodies that are periodic acidSchiff (PAS) positive. Endodermal sinus tumors (yolk sac tumors) have a reticular pattern of loose meshwork of small spaces lined by primitive tumor cells, producing a sievelike appearance. Each cell has hyperchromatic nuclei. Both embryonal carcinomas and endodermal sinus tumors release α-fetoprotein (AFP). Choriocarcinomas are the highest malignant germ cell tumors and are composed exclusively of cytotrophoblasts and syncytiotrophoblasts that produce ßhuman chorionic gonadotropin (ß-HCG). Hemorrhage and necrosis are common. In mixed germ cell tumors, the
PRIMARY INTRACRANIAL GERM CELL TUMORS Table 2 Clinical signs and symptoms.2,33
DIAGNOSTIC CONSIDERATIONS
Signs and symptoms
Headache, nausea, vomiting, papilledema (raised intracranial pressure) Parinaud’s syndrome Diplopia Diabetes insipidus Visual impairment Amenorrhea Growth retardation Precocious puberty
651
Location Pineal
Suprasellar
+++
+
+++ ++ − − − + ++
− − +++ +++ + ++ +
most common components are germinomatous, followed by mature and immature teratomatous components.
CLINICAL PRESENTATION The clinical signs and symptoms depend on the location of the tumor and are summarized in Table 2. Tumors in the pineal region compress and obstruct the cerebral aqueduct, resulting in raised intracranial pressure due to obstructive hydrocephalus and present with headache, nausea, vomiting, and lethargy. They also compress the tectal plate producing the characteristic Parinaud’s syndrome with paralysis of upward gaze and accommodation and convergence-retraction nystagmus. However, less than 50% of patients present with a complete Parinaud’s syndrome. Diplopia is also common, most often due to fourth or sixth nerve palsies. Suprasellar tumors present most frequently with visual or endocrinologic signs and symptoms. Compression or infiltration of the optic chiasm results in bitemporal hemianopia and decreased visual acuity (in 33% in Jennings’ series,2 85% in Matsutani’s series33 ). Tumor invasion into the anterior pituitary produces endocrine deficits (in 33% of patients in Jennings’ series2 and 93% in Matsutani’s series33 ) such as delayed puberty in children and sexual dysfunction in adults due to gonadotropin deficiency. Children may also have growth retardation from growth hormone deficiency (in 30% of cases33 ). Panhypopituitarism and elevations in prolactin may arise.2 Many suprasellar germ cell tumor patients presents also with posterior pituitary dysfunction (41% in Jennings’ series,2 86% in Matsutani’s series33 ) resulting in vasopressin deficiency and symptoms such as polydipsia and polyuria in diabetes insipidus.2,34 Rarely, precocious puberty may develop due to hypothalamic injury or a tumor that is producing HCG.35 Endocrinopathies may precede the diagnosis for years.2 Patients with concurrent pineal and suprasellar tumors can have a combination of the above noted findings. Less common tumor locations such as the basal ganglia and thalamus, cerebellum and brain stem, and spinal cord may produce other cranial nerve deficits, ataxia, hemiparesis, myelopathic signs, or hemisensory deficits.2,27,36 – 39 In neonates and young infants, germ cell tumors (most often teratoma) present with less specific signs of raised intracranial pressure such as listlessness, failure to thrive, macrocephaly, and a bulging fontanelle. Frequently, these tumors are diagnosed by antenatal sonography.40
The differential diagnosis of intracranial germ cell tumors depends on the tumor location and varies with age and ethnicity of the patient population. In a Japanese case series, germ cell tumors comprised three quarters of pineal tumors.41 In contrast, only one-quarter to one-third of pineal region tumors in several North American and European case series were germ cell tumors, whereas the rest were pineal parenchymal or glial tumors.37,42 – 45 In the suprasellar region, one-third of tumors in children are germ cell tumors. The remainder is comprised of hypothalamic gliomas, hamartomas, or craniopharyngiomas.46 In adults, meningiomas, metastases, and large pituitary tumors are added to the list of suprasellar tumors. MRI with and without gadolinium is highly sensitive in the detection of intracranial germ cell tumors and is the imaging modality of choice.47,48 When MRI is not available, computed topography (CT) should be performed. Pineal calcifications on plain skull radiographs can also provide clues to the diagnosis of pineal region tumors as 70% of patients with pineal region tumors have calcifications, an otherwise uncommon finding in pediatric patients.49 The various germ cell tumors have characteristic MRI findings: Germinomas are usually solid and uniform. The signal intensity of the solid part is similar to that of gray matter and the tumors usually enhance50 (see Figure 1a –c). Calcifications and intratumoral cysts are seen in about half of the cases.50,51 Peritumoral edema is not usually present. Cerebral hemiatrophy has been described as a presenting characteristic in case reports of thalamic or basal ganglia germinomas.52 Multiple midline germinomas with lesions both in the pineal and suprasellar regions represent a unique presentation of germinoma. The pathology in these cases is always secure. Among nongerminomatous germ cell tumors, a cystic component is more frequently present while the solid part has similar signal intensity as gray matter.50 There is homogeneous enhancement. Invasion of adjacent structures is common. Teratomas are usually heterogeneous cystic masses with irregular shape, distinct tumor margins, and presence of fat and calcifications.48,50 Malignant teratomas tend to have smaller cystic components. Calcifications are less common, whereas peritumoral edema is more frequently seen. Since pineal parenchymal tumors and gliomas arising in pineal and suprasellar regions can have very similar signal characteristics, MRI does not substitute for a tissue diagnosis and in most clinical scenarios, tumor biopsy is required for diagnosis. Germ cell tumors possess specific tumor marker profiles, measurable in the serum and – with usually higher levels – in the CSF.53 AFP is a marker of yolk cells. Markedly elevated AFP levels are diagnostic of endodermal sinus tumors and more modest elevations are seen in mixed germ cell tumors, immature teratomas, and embryonal carcinomas. βHCG is produced by placental trophoblastic tissue. Strikingly increased levels of β-HCG are seen in choriocarcinomas. Mild elevations have also been seen in patients with germinomas that contain syncytiotrophoblastic giant cells.54 PLAP is
652
NEUROLOGICAL MALIGNANCIES
(a)
(b)
(c) Figure 1 MRI of the brain with and without gadolinium (T1-weighted images, coronal section, and FLAIR image, axial section) of a 19 year old woman with germinoma who presented with diabetes insipidus, hypothyroidism, and adrenal insufficiency. The contrast enhanced T1-weighted images demonstrate a brightly and uniformly enhancing suprasellar lesion expanding the optic chiasm (a). The lesion is isointense to gray matter on the T1-weighted image (b). FLAIR images demonstrate hyperintensity involving the optic chiasm and extending bilaterally into the optic nerves. The signal abnormality extends into the hypothalamus bilaterally and engulfs the fornices (c).
a rather nonspecific germ cell tumor marker and rarely helpful in distinguishing among the different histologies of germ cell tumors.55 Mature teratomas do not have a distinct biochemical signature. Table 3 summarizes the tumor markers in intracranial germ cell tumors. When an intracranial germ cell tumor is suspected, the evaluation of the patient includes a contrast-enhanced MRI of the brain and entire spine, as germ cell tumors can disseminate throughout the entire neuroaxis. Tumor marker assays (β-HCG, AFP, PLAP) in serum and CSF and CSF cytology complement the imaging workup. When CSF cannot be safely obtained by lumbar puncture due to the presence of obstructive hydrocephalus, it should be collected at the time
Table 3 Tumor markers in intracranial germ cell tumors.53 – 55
Tumor
β-HCG
AFP
PLAP
Germinoma Germinoma with syncytiotrophoblasts Teratoma (mature) Teratoma (immature and malignant) Embryonal carcinoma Choriocarcinoma Endodermal sinus tumor
− ++ − − + +++ +
− − − + + − +++
+ + − + + + +
of shunt placement. Positive cytology is a relatively common finding in patients with germ cell tumors.56 In patients
PRIMARY INTRACRANIAL GERM CELL TUMORS
with distinctive imaging findings, unambiguous tumor markers may obviate the need for a tissue diagnosis. In most cases, however, tissue diagnosis is necessary for optimal management. In the past, a common method to determine the histology of pineal region tumors without surgery was to observe their response to a trial of radiation. If the tumor was radiosensitive, it was presumed to be a germinoma. If there was no response, biopsy was recommended.57 Improvements in neurosurgical techniques and low operative morbidity of less than 5%44 and the relatively low incidence of radiosensitive germinomas in western populations have led to abandonment of this method which, however, is still prevalent in Japan.58 The preferred method for obtaining tissue has been a subject of debate among neurosurgeons. Stereotactic biopsies have been advocated by some because of lower morbidity,59,60 whereas most prefer open procedures to minimize the potential of sampling errors in heterogeneous tumors and to allow for more extensive resections.37,61
SURGERY There is general consensus regarding the need for surgery to obtain a tissue diagnosis unless specific tumor marker elevations are present. The role of surgery for pure germinomas, for which curative radiotherapy is available, is limited to stereotactic biopsy or open biopsy. In nongerminomatous tumors, there may be a therapeutic role for more extensive resections, but the available data is controversial.62 As nongerminomatous germ cell tumors are much less radiosensitive than germinomas and long-term survival has been poor,2 reduction of tumor mass has long been thought to increase the curative potential of subsequent chemo- and radiation therapy.34 Data from two case series suggest that total or subtotal resections improves survival,33,63 whereas pooled data from several European studies did not demonstrate a correlation between improved survival and extent of resection.64 It has been suggested that patients with aggressive resection have an improved chance of survival.65 Another surgical approach is that of “second-look resection” which applies to nongerminomatous tumors that had an incomplete response to initial chemotherapy.66 “Secondlook surgery” has a therapeutic as well as a diagnostic role. Residual radiographic abnormalities after chemotherapy are frequently found to only contain fibrotic tissue or mature teratoma components.67,68 Further chemotherapy would not benefit these patients. The growing teratoma syndrome refers to an enlarging mixed germ cell tumor with a secreting portion that responds to the chemotherapy and a non-secreting portion such as mature teratoma that continues to grow. Surgical resection is usually considered for the mature teratoma component.69
RADIATION For decades, radiotherapy has been the first-line therapy for intracranial germinoma and the majority of patients have had long-term survival.3,33,70 – 73 Therefore, any new treatment approach must be weighted against standard radiotherapy practice. The optimal radiation treatment has been a matter
653
of debate and data from randomized controlled trials is lacking. Germinomas have been traditionally treated using prophylactic craniospinal radiation with a boost to local residual disease, resulting in high cure rates with reported 5-year survival ranging from 80 to 100%, and 5-year disease-free survival of 70–90%3,33,70 – 74 irrespective of stage or completeness of staging at diagnosis. Craniospinal radiation remains the therapy of choice for metastasized germinomas, including positive CSF cytology.75,76 Since neuroaxis irradiation in children has long-term effects on growth, and cognitive and neuroendocrine development,77 – 80 other therapeutic avenues to reduce radiation exposure are being investigated. Different approaches, including lowering of the total radiation dose or reduction of the radiation volume, in combination with chemotherapy have been used to limit radiation-related morbidity. The spread in intracranial germinoma is assumed to occur along the subependymal lining of the walls of the third and fourth ventricles, leading to regional intraventricular relapse before spinal dissemination occurs.81 This underlines the importance of intraventricular radiation. A systematic review of all recent case series and trials by Rogers et al. found no significant difference in rates of spinal relapse between 343 patients who received craniospinal radiotherapy and 278 patients given whole-brain or wholeventricular radiotherapy plus boost. The overall relapse rate was 3.8% in the craniospinal radiation group and 7.6% in the whole-brain and ventricular radiation group. In both patient groups, spinal relapse happened in less than 3% of patients.82 Consequently, the authors conclude that craniospinal irradiation is not necessary in localized germinomas82 and that the routine use of ventricular radiotherapy is adequate. In contrast, local radiotherapy alone was found to be inadequate as the spinal relapse rate in the 133 patients receiving local radiation only was fourfold higher than with ventricular radiation (11.3%). More recent data advocate doses of 40 to 45 Gy to the primary tumor when radiotherapy is used alone.60,70,83 A study from the Harvard Joint Center of Radiation Therapy did not show a higher incidence of recurrence among 14 patients treated with whole-brain radiation and <50 Gy to the tumor volume compared to 21 patients who received whole-brain radiation and >50 Gy to the tumor volume arguing for lower dose radiation.73 The group at the University of Kyoto recommends a dose-tumor size schema in the radiotherapeutic approach to germinomas. For tumors <2.5 cm, 2.5 to 4.0 cm, and >4 cm, the recommended doses are 40, 45, and 50 Gy, respectively.70 For patients with CSF positive cytology, a 20 to 24-Gy-craniospinal dose is recommended. With such a dose to the remainder of the entire neuroaxis, the outcomes of patients with positive and negative cytology were similar.70 The treatment of nongerminomatous intracranial germ cell tumors is less clear. Nongerminomatous germ cell tumors are commonly less sensitive to irradiation than germinomas. The earlier case series by Jennings et al. reported longterm disease-free survival in less than 25% of patients with radiation alone.2 In later series, long-term survival ranged from 20 to 60% with radiotherapy.3,37,57,84 Craniospinal radiation is given when there is evidence of metastasis or
654
NEUROLOGICAL MALIGNANCIES
dissemination including positive CSF cytology. The radiation volumes for localized disease are debated. Wolden reported a 60% progression-free survival in 10 patients after involved field radiation of 50 to 54 Gy.74 Most other studies, however, support the use of larger fields84 – 87 and doses beyond 40 Gy. Since the majority of nongerminomatous tumors are relapsing with radiotherapy, chemotherapy should be part of the multimodality approach for nongerminomatous germ cell tumors.27,64,66,86
CHEMOTHERAPY PLUS COMBINED MODALITY TREATMENT CNS germinomas are very sensitive to platinum-based chemotherapy,85,88 but recurrence rates with chemotherapy alone are high.66,89 Balmaceda et al.66 and, more recently, Kellie et al.90 reported a complete response rate of 82% after initial chemotherapy using carboplatin, etoposide, and bleomycin and cyclophosphamide. Disease recurrence developed subsequently in more than 50% of these patients, resulting in an overall survival rate of only 84% at 2 years66 and 68% at 5 years.90 Overall, 10 and 16% of patients died of toxicity associated with the chemotherapy.66,90 In addition to considerable acute toxicity, disappointing long-term survival on chemotherapy-only regimens speaks strongly against the exclusion of radiotherapy from treatment protocols for intracranial germinoma. Consequently, chemotherapy is generally only used in combination with radiotherapy as part of combined modality strategies in which neoadjuvant chemotherapy replaces radiation resulting in a reduction in both radiation volume and dose.91 – 93 Combined modality treatment decreases the adverse effects from both individual therapies. While the initial response rates are high and frequently approach 80–100%, the overall relapse rate remains at about 15% in more recent studies.87 In the study by Matsutani et al.,92 56 (84%) of 67 patients with localized germinoma achieved complete responses after induction chemotherapy and then received 24 Gy radiotherapy to a field including the third and lateral ventricles. The relapse rate was 12.2% after 2.9 years, which is higher than with standard ventricular radiotherapy. Other studies report control rates comparable to craniospinal or ventricular radiotherapy.87,94 The study by Aoyama et al.87 used a regimen of induction chemotherapy for pure germinomas with etoposide (100 mg m−2 ) and cisplatin (20 mg m−2 ) (EP) for 5 days every 4 weeks for three to four cycles and for other germ cell tumors ifosfamide (900 mg m−2 ), cisplatin (20 mg m−2 ), and etoposide (60 mg m−2 ) (ICE) for 5 days every 4 weeks for three to four cycles followed by radiation. The radiation dose was 24 Gy for pure germinomas and up to 40–54 Gy for other germ cell tumors. Radiation was given only to the tumor bed as defined by pretreatment MRI unless definite predefined highly malignant criteria warranted craniospinal irradiation, such as positive CSF cytology. Results were excellent with regard to disease-free survival, overall survival, and relapse-free survival for all patients at 5 years, with rates of 100, 93, and 69%, respectively. Overall survival was again better for pure germinomas than for other germ cell tumors, with rates of 100 versus 75%. Given the
inconsistency of the data, the role of chemotherapy as part of reduced radiation protocols remains to be defined. Most nongerminomatous tumors are more radioresistant than pure germinomas and associated with a relatively poor outcome, ranging from 20 to 45% survival following treatment with full-dose neuroaxis radiation therapy alone.2,27,60 The combination of radiation therapy with platinum-containing multidrug chemotherapy regimens has been associated with improved outcomes.64,85,86,90,95 – 99 Unfortunately, most case series lump together a variety of different histologies when reporting results.
PROGNOSIS Patients with germinomas have generally an excellent prognosis with high cure rates, even if disseminated disease is present.3,33,70 – 73 As young patients are likely to survive for many years, the main concerns are the late sequelae of large-volume radiotherapy. Cranial and craniospinal radiation therapy are associated with significant neurodevelopmental disabilities, neuroendocrine dysfunction, and growth retardation in children. Neuropsychologic testing in adults treated with cranial radiation therapy also demonstrates the development of cognitive dysfunction. Patients also remain at risk of developing secondary tumors with a 20-year cumulative incidence of about 12%.100 The nongerminomatous germ cell tumors have a less favorable outcome with radiation alone. There is a clear relationship between histopathology and survival.2,3,27,33,89 In the study by Matsutani et al., patients with embryonal carcinoma, endodermal sinus tumor or choriocarcinoma had poor outcome with reported 3-year survival of 27%, whereas patients with mixed germinoma and mature or immature teratoma had a 3-year survival of 94%.3 In mixed tumors, patients with predominantly germinomatous or teratomatous components had a 70% 3-year survival, and patients with mixed tumors composed of predominantly malignant nongerminomatous elements had a 3-year survival of 9%.
RECOMMENDATIONS Tissue diagnosis is essential and can be safely achieved with current stereotactic and microsurgical techniques. Patients with a biopsy-confirmed diagnosis of intracranial germinoma should undergo a careful staging evaluation for CSF dissemination with contrast enhanced MRI of the entire neuroaxis and CSF cytology. In our practice, we presently recommend ventricular radiation to 25 Gy and a total tumor volume dose to 45 Gy in all patients. In patients with evidence of CSF dissemination or multiple midline lesions, we recommend spinal irradiation to a dose of 22 Gy. A randomized controlled collaborative group trial comparing craniospinal radiotherapy with whole-ventricular radiotherapy plus boost is needed to confirm that exclusion of craniospinal radiation is appropriate. Combination chemotherapy and radiation therapy has the potential to decrease the adverse effects from both individual therapies and may improve the rate of complete remission and overall survival but requires further study.
PRIMARY INTRACRANIAL GERM CELL TUMORS
REFERENCES 1. Brodeur GM, et al. Malignant germ cell tumors in 57 children and adolescents. Cancer 1981; 48(8): 1890 – 8. 2. Jennings MT, Gelman R, Hochberg F. Intracranial germ-cell tumors: natural history and pathogenesis. J Neurosurg 1985; 63(2): 155 – 67. 3. Matsutani M, Takakura K, Sano K. Primary intracranial germ cell tumors: pathology and treatment. Prog Exp Tumor Res 1987; 30: 307 – 12. 4. Takei Y, Pearl GS. Ultrastructural study of intracranial yolk sac tumor: with special reference to the oncologic phylogeny of germ cell tumors. Cancer 1981; 48(9): 2038 – 46. 5. Russell D. The pinealoma: its relationship to teratoma. J Pathol Bacteriol 1944; 56: 145 – 50. 6. Friedman NM, Moore RA. Tumors of the testis. A report on 922 cases. Milit Surg 1946; 99: 573 – 93. 7. Friedman N. Germinoma of the pineal. Its identity with germinoma (“seminoma”) of the testis. Cancer Res 1947; 7: 363 – 8. 8. Dixon F, Moore RA. Testicular tumors. A clinicopathological study. Cancer 1953; 6: 427 – 54. 9. Russell D. Ectopic pinealoma: its kinship to atypical teratoma. J Pathol Bacteriol 1954; 68: 125 – 9. 10. Friedman N. Comparative morphogenesis of extragonadal and gonadal teratoid tumors. Cancer 1951; 4: 256 – 76. 11. Teilum G. Classification of endodermal sinus tumour (mesoblastoma vitellinum) and so-called “embryonal carcinoma” of the ovary. Acta Pathol Microbiol Scand 1965; 64(4): 407 – 29. 12. Teilum G. Endodermal sinus tumors of the ovary and testis. Comparative morphogenesis of the so-called mesonephroma ovarii (Schiller) and extraembryonic (yolk sac-allantoic) structures of the rat’s placenta. Cancer 1959; 12: 1092 – 105. 13. Kleihues P, Burger PC, Scheithauer BW. The new WHO classification of brain tumours. Brain Pathol 1993; 3(3): 255 – 68. 14. Nichols CR, Fox EP. Extragonadal and pediatric germ cell tumors. Hematol Oncol Clin North Am 1991; 5(6): 1189 – 209. 15. Chaganti RS, Rodriguez E, Mathew S. Origin of adult male mediastinal germ-cell tumours. Lancet 1994; 343(8906): 1130 – 2. 16. Schneider DT, et al. Epidemiologic analysis of 1,442 children and adolescents registered in the German germ cell tumor protocols. Pediatr Blood Cancer 2004; 42(2): 169 – 75. 17. Albrecht S, et al. Cytogenetic demonstration of gene amplification in a primary intracranial germ cell tumor. Genes Chromosomes Cancer 1993; 6(1): 61 – 3. 18. de Bruin TW, et al. Isochromosome 12p-positive pineal germ cell tumor. Cancer Res 1994; 54(6): 1542 – 4. 19. Taylor MD, et al. Molecular genetics of pineal region neoplasms. J Neurooncol 2001; 54(3): 219 – 38. 20. Rickert CH, et al. Comparative genomic hybridization in pineal germ cell tumors. J Neuropathol Exp Neurol 2000; 59(9): 815 – 21. 21. Okada Y, et al. Hypomethylated X chromosome gain and rare isochromosome 12p in diverse intracranial germ cell tumors. J Neuropathol Exp Neurol 2002; 61(6): 531 – 8. 22. Iwato M, et al. Alterations of the INK4a/ARF locus in human intracranial germ cell tumors. Cancer Res 2000; 60(8): 2113 – 5. 23. Grant FC. A study of the results of surgical treatment in 2,326 consecutive patients with brain tumor. J Neurosurg 1956; 13(5): 479 – 88. 24. Pinkerton CR. Malignant germ cell tumours in childhood. Eur J Cancer 1997; 33(6): 895 – 901; discussion -2. 25. 2004 – 2005 Statistical Report: Primary Brain Tumors in the United States: Central Brain Tumor Registry of the United States; http://www.cbtrus.org/reports//2005, 2005. 26. Mori K, Kurisaka M. Brain tumors in childhood: statistical analysis of cases from the Brain Tumor Registry of Japan. Childs Nerv Syst 1986; 2(5): 233 – 7. 27. Hoffman HJ, et al. Intracranial germ-cell tumors in children. J Neurosurg 1991; 74(4): 545 – 51. 28. Ho DM, Liu HC. Primary intracranial germ cell tumor. Pathologic study of 51 patients. Cancer 1992; 70(6): 1577 – 84. 29. Markesbery WR, et al. Ultrastructural study of the pineal germinoma in vivo and in vitro. Cancer 1976; 37(1): 327 – 37.
655
30. Bjornsson J, et al. Intracranial germ cell tumors: pathobiological and immunohistochemical aspects of 70 cases. J Neuropathol Exp Neurol 1985; 44(1): 32 – 46. 31. Bentley AJ, et al. A comparative morphological and immunohistochemical study of testicular seminomas and intracranial germinomas. Histopathology 1990; 17(5): 443 – 9. 32. Beeley JM, et al. Ectopic pinealoma: an unusual clinical presentation and a histochemical comparison with a seminoma of the testis. J Neurol Neurosurg Psychiatry 1973; 36(5): 864 – 73. 33. Matsutani M, et al. Primary intracranial germ cell tumors: a clinical analysis of 153 histologically verified cases. J Neurosurg 1997; 86(3): 446 – 55. 34. Allen JC. Controversies in the management of intracranial germ cell tumors. Neurol Clin 1991; 9(2): 441 – 52. 35. Rivarola MA, et al. Endocrine disorders in 66 suprasellar and pineal tumors of patients with prepubertal and pubertal ages. Horm Res 1992; 37(1 – 2): 1 – 6. 36. Kobayashi T, et al. Unilateral germinomas involving the basal ganglia and thalamus. J Neurosurg 1981; 55(1): 55 – 62. 37. Edwards MS, et al. Pineal region tumors in children. J Neurosurg 1988; 68(5): 689 – 97. 38. Abay EO II, et al. Pineal tumors in children and adolescents. Treatment by CSF shunting and radiotherapy. J Neurosurg 1981; 55(6): 889 – 95. 39. Chute DJ, et al. Primary intramedullary spinal cord germinoma: case report. J Neurooncol 2003; 63(1): 69 – 73. 40. Isaacs H Jr. Perinatal (fetal and neonatal) germ cell tumors. J Pediatr Surg 2004; 39(7): 1003 – 13. 41. Sano K, Matsutani M. Pinealoma (Germinoma) treated by direct surgery and postoperative irradiation. A long-term follow-up. Childs Brain 1981; 8(2): 81 – 97. 42. Packer RJ, et al. Pineal region tumors of childhood. Pediatrics 1984; 74(1): 97 – 102. 43. Lapras C, et al. Direct surgery for pineal tumors: occipitaltranstentorial approach. Prog Exp Tumor Res 1987; 30: 268 – 80. 44. Bruce JN, Stein BM. Surgical management of pineal region tumors. Acta Neurochir (Wien) 1995; 134(3 – 4): 130 – 5. 45. Bruce DA, Allen JC. Tumor staging for pineal region tumors of childhood. Cancer 1985; 56(7 Suppl): 1792 – 4. 46. Sung DI. Suprasellar tumors in children: a review of clinical manifestations and managements. Cancer 1982; 50(7): 1420 – 5. 47. Tien RD, Barkovich AJ, Edwards MS. MR imaging of pineal tumors. AJR Am J Roentgenol 1990; 155(1): 143 – 51. 48. Fujimaki T, et al. CT and MRI features of intracranial germ cell tumors. J Neurooncol 1994; 19(3): 217 – 26. 49. Smirniotopoulos JG, Rushing EJ, Mena H. Pineal region masses: differential diagnosis. Radiographics 1992; 12(3): 577 – 96. 50. Liang L, et al. MRI of intracranial germ-cell tumours. Neuroradiology 2002; 44(5): 382 – 8. 51. Moon WK, et al. Intracranial germinomas: correlation of imaging findings with tumor response to radiation therapy. AJR Am J Roentgenol 1999; 172(3): 713 – 6. 52. Kim CH, et al. Cerebral germinoma with hemiatrophy of the brain: report of three cases. Acta Neurochir (Wien) 2002; 144(2): 145 – 50; discussion 50. 53. Allen JC, et al. Alphafetoprotein and human chorionic gonadotropin determination in cerebrospinal fluid. An aid to the diagnosis and management of intracranial germ-cell tumors. J Neurosurg 1979; 51(3): 368 – 74. 54. Inamura T, et al. Human chorionic gonadotropin in CSF, not serum, predicts outcome in germinoma. J Neurol Neurosurg Psychiatry 1999; 66(5): 654 – 7. 55. Shinoda J, et al. Placental alkaline phosphatase as a tumor marker for primary intracranial germinoma. J Neurosurg 1988; 68(5): 710 – 20. 56. Schulte FJ, et al. Pineal region tumours of childhood. Eur J Pediatr 1987; 146(3): 233 – 45. 57. Nakagawa K, et al. Radiation therapy of intracranial germ cell tumors with radiosensitivity assessment. Radiat Med 1992; 10(2): 55 – 61. 58. Paulino AC, Wen BC, Mohideen MN. Controversies in the management of intracranial germinomas. Oncology (Williston Park) 1999; 13(4): 513 – 21; discussion 521 – 2, 528 – 3.
656
NEUROLOGICAL MALIGNANCIES
59. Linstadt D, et al. Radiotherapy of primary intracranial germinomas: the case against routine craniospinal irradiation. Int J Radiat Oncol Biol Phys 1988; 15(2): 291 – 7. 60. Dearnaley DP, et al. Pineal and CNS germ cell tumors: Royal Marsden Hospital experience 1962 – 1987. Int J Radiat Oncol Biol Phys 1990; 18(4): 773 – 81. 61. Baumgartner JE, Edwards MS. Pineal tumors. Neurosurg Clin N Am 1992; 3(4): 853 – 62. 62. Sawamura Y, et al. Management of primary intracranial germinomas: diagnostic surgery or radical resection? J Neurosurg 1997; 87(2): 262 – 6. 63. Shinoda J, et al. Prognostic factors and therapeutic problems of primary intracranial choriocarcinoma/germ-cell tumors with high levels of HCG. J Neurooncol 2004; 66(1 – 2): 225 – 40. 64. Calaminus G, et al. Intracranial germ cell tumors: a comprehensive update of the European data. Neuropediatrics 1994; 25(1): 26 – 32. 65. Nam DH, et al. Treatment of intracranial nongerminomatous malignant germ cell tumor in children: the role of each treatment modality. Childs Nerv Syst 1999; 15(4): 185 – 91. 66. Balmaceda C et al., The First International Central Nervous System Germ Cell Tumor Study. Chemotherapy without irradiation – a novel approach for newly diagnosed CNS germ cell tumors: results of an international cooperative trial. J Clin Oncol 1996; 14(11): 2908 – 15. 67. Friedman JA, et al. Management of malignant pineal germ cell tumors with residual mature teratoma. Neurosurgery 2001; 48(3): 518 – 22; discussion 22 – 3. 68. Brandes AA, Pasetto LM, Monfardini S. The treatment of cranial germ cell tumours. Cancer Treat Rev 2000; 26(4): 233 – 42. 69. Yagi K, et al. Growing teratoma syndrome in a patient with a nongerminomatous germ cell tumor in the neurohypophysis – case report. Neurol Med Chir (Tokyo) 2004; 44(1): 33 – 7. 70. Shibamoto Y, Takahashi M, Abe M. Reduction of the radiation dose for intracranial germinoma: a prospective study. Br J Cancer 1994; 70(5): 984 – 9. 71. Ogawa K, et al. Long-term results of radiotherapy for intracranial germinoma: a multi-institutional retrospective review of 126 patients. Int J Radiat Oncol Biol Phys 2004; 58(3): 705 – 13. 72. Bamberg M, et al. Radiation therapy for intracranial germinoma: results of the German cooperative prospective trials MAKEI 83/86/89. J Clin Oncol 1999; 17(8): 2585 – 92. 73. Hardenbergh PH, et al. Intracranial germinoma: the case for lower dose radiation therapy. Int J Radiat Oncol Biol Phys 1997; 39(2): 419 – 26. 74. Wolden SL, et al. Radiation therapy for primary intracranial germ-cell tumors. Int J Radiat Oncol Biol Phys 1995; 32(4): 943 – 9. 75. Shirato H, et al. Analysis of long-term treatment of intracranial germinoma. Int J Radiat Oncol Biol Phys 1997; 37(3): 511 – 5. 76. Shibamoto Y, et al. The role of cerebrospinal fluid cytology in radiotherapy planning for intracranial germinoma. Int J Radiat Oncol Biol Phys 1994; 29(5): 1089 – 94. 77. Jenkin D, et al. Pineal region germinomas in childhood treatment considerations. Int J Radiat Oncol Biol Phys 1990; 18(3): 541 – 5. 78. Sutton LN, et al. Quality of life of adult survivors of germinomas treated with craniospinal irradiation. Neurosurgery 1999; 45(6): 1292 – 7; discussion 7 – 8. 79. Spiegler BJ, et al. Change in neurocognitive functioning after treatment with cranial radiation in childhood. J Clin Oncol 2004; 22(4): 706 – 13. 80. Benesch M, et al. Tumor- and treatment-related side effects after multimodal therapy of childhood intracranial germ cell tumors. Acta Paediatr 2001; 90(3): 264 – 70.
81. Bouffet E, et al. Combined treatment modality for intracranial germinomas: results of a multicentre SFOP experience. Societe Francaise d’Oncologie Pediatrique. Br J Cancer 1999; 79(7 – 8): 1199 – 204. 82. Rogers SJ, Mosleh-Shirazi MA, Saran FH. Radiotherapy of localised intracranial germinoma: time to sever historical ties? Lancet Oncol 2005; 6(7): 509 – 19. 83. Aoyama H, et al. Pathologically-proven intracranial germinoma treated with radiation therapy. Radiother Oncol 1998; 47(2): 201 – 5. 84. Haas-Kogan DA, et al. Radiation therapy for intracranial germ cell tumors. Int J Radiat Oncol Biol Phys 2003; 56(2): 511 – 8. 85. Yoshida J, et al. Prognosis of intracranial germ cell tumours: effectiveness of chemotherapy with cisplatin and etoposide (CDDP and VP-16). Acta Neurochir (Wien) 1993; 120(3 – 4): 111 – 7. 86. Robertson PL, DaRosso RC, Allen JC. Improved prognosis of intracranial non-germinoma germ cell tumors with multimodality therapy. J Neurooncol 1997; 32(1): 71 – 80. 87. Aoyama H, et al. Induction chemotherapy followed by low-dose involved-field radiotherapy for intracranial germ cell tumors. J Clin Oncol 2002; 20(3): 857 – 65. 88. Allen JC, Kim JH, Packer RJ. Neoadjuvant chemotherapy for newly diagnosed germ-cell tumors of the central nervous system. J Neurosurg 1987; 67(1): 65 – 70. 89. Matsutani M. Clinical management of primary central nervous system germ cell tumors. Semin Oncol 2004; 31(5): 676 – 83. 90. Kellie SJ, et al. Intensive cisplatin and cyclophosphamide-based chemotherapy without radiotherapy for intracranial germinomas: failure of a primary chemotherapy approach. Pediatr Blood Cancer 2004; 43(2): 126 – 33. 91. Sawamura Y, et al. Induction chemotherapy followed by reducedvolume radiation therapy for newly diagnosed central nervous system germinoma. J Neurosurg 1998; 88(1): 66 – 72. 92. Matsutani M, et al. Combined treatment with chemotherapy and radiation therapy for intracranial germ cell tumors. Childs Nerv Syst 1998; 14(1 – 2): 59 – 62. 93. Baranzelli MC, et al. Nonmetastatic intracranial germinoma: the experience of the French Society of Pediatric Oncology. Cancer 1997; 80(9): 1792 – 7. 94. Buckner JC, et al. Phase II trial of primary chemotherapy followed by reduced-dose radiation for CNS germ cell tumors. J Clin Oncol 1999; 17(3): 933 – 40. 95. Patel SR, et al. Cisplatin-based chemotherapy in primary central nervous system germ cell tumors. J Neurooncol 1992; 12(1): 47 – 52. 96. Kida Y, et al. Chemotherapy with cisplatin for AFP-secreting germcell tumors of the central nervous system. J Neurosurg 1986; 65(4): 470 – 5. 97. Itoyama Y, et al. Treatment of intracranial nongerminomatous malignant germ cell tumors producing alpha-fetoprotein. Neurosurgery 1995; 36(3): 459 – 64; discussion 64 – 6. 98. Gobel U et al., GPOH MAKEI and the MAHO study groups. Germcell tumors in childhood and adolescence. Ann Oncol 2000; 11(3): 263 – 71. 99. Chang TK, Wong TT, Hwang B. Combination chemotherapy with vinblastine, bleomycin, cisplatin, and etoposide (VBPE) in children with primary intracranial germ cell tumors. Med Pediatr Oncol 1995; 24(6): 368 – 72. 100. Shibamoto Y, et al. Intracranial germinoma: radiation therapy with tumor volume-based dose selection. Radiology 2001; 218(2): 452 – 6.
Section 10 : Neurological Malignancies
60
Primary Central Nervous System Lymphoma Lisa M. DeAngelis
INTRODUCTION Primary central nervous system lymphoma (PCNSL) was first described as perithelial sarcoma by Bailey.1 It was subsequently classified as reticulum cell sarcoma because of the characteristic reticulum it deposited around blood vessels.2 Later, it was described as a microglioma after the presumed cell of origin, the microglia.3 Henry et al.4 classified these tumors as primary malignant lymphomas of the central nervous system (CNS) because of their histologic similarities to extraneural malignant lymphoma. The lymphoid nature of PCNSL was established unequivocally by modern immunohistochemical techniques.5 – 8
EPIDEMIOLOGY PCNSL accounts for 1–2% of non-Hodgkin’s lymphoma (NHL) and at least 0.5–1.2% of all primary brain tumors.9,10 The National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) database revealed a threefold increase in the incidence of PCNSL between 1973–1984 and 1985–1997 among presumed immunocompetent patients.11 This increase reflected an overall increase in all extranodal lymphomas, but the rise was proportionately greatest in the CNS. However, the rate of increase was slow after 1985. A second study demonstrated a decreased rate in patients over the age of 60 after 1995, although the rate remained elevated in younger patients.12 The increased incidence has not been seen worldwide. Data from Denmark, Scotland, and Alberta, Canada fail to show a significant change in the incidence of PCNSL in immunocompetent individuals,13 – 15 raising the question of an environmental contribution to tumorigenesis. The reason for the apparent increased incidence in immunocompetent patients in the United States cannot be explained by advances in technology, such as computed tomography (CT) or magnetic resonance imaging (MRI), physician referrals, or changes in tumor nosology. PCNSL has been associated with a variety of congenital (Wiskott-Aldrich syndrome, ataxia –telangiectasia) and
acquired (human immunodeficiency virus [HIV], renal transplant recipients) immunodeficiency states.9,10,16 It is particularly common in HIV-infected individuals, where the incidence is 1.6–9.0%.17,18 After toxoplasmosis, PCNSL is the second most common intracranial mass lesion found in HIV-infected patients. Prior to the introduction of the new anti-HIV agents, there was growing evidence that PCNSL was rising in the acquired immunodeficiency syndrome (AIDS) population.19 However, the incidence of PCNSL in AIDS patients has fallen significantly since the introduction of highly active antiretroviral therapy (HAART) which helps restore immune function.20 PCNSL can also occur as a secondary malignancy. DeAngelis21 reported that 13% of the patients with PCNSL had a prior malignancy but a more recent literature review22 found this number to be 7.9%. Median latency from initial tumor to diagnosis of PCNSL was 8 years.22 Whether PCNSL develops as a consequence of genetic predisposition or prior antineoplastic treatment is not clear, although no patient received prior cranial radiotherapy. Nonetheless, it is important to keep PCNSL in mind as a possible cause of an intracranial mass lesion in patients with a history of systemic cancer because treatment for brain metastasis markedly differs from treatment for PCNSL.
PATHOLOGY PCNSLs are usually diffuse intermediate or high-grade NHLs, the majority of which are diffuse large cell, large cell immunoblastic or lymphoblastic subtypes. Phenotypically, more than 95% are of B cell origin, CD20+ (see Figure 1),5 – 8,23 but T cell tumors, CD3+ or CD45RO+, are occasionally reported.24 Cytogenetic abnormalities in PCNSL have been reported involving chromosomes 1, 6, 7, and 14.25 These abnormalities are similar to those seen in nodal B cell lymphomas. Kumanishi et al.26 reported p15 and p16 deletions on chromosome 9 by Southern blot analysis in four out of five PCNSL tumors.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
658
NEUROLOGICAL MALIGNANCIES
Figure 1 Immunohistochemistry of PCNSL showing CD20 positivity.
Figure 2 PCNSL hematoxylin and eosin stain showing angiocentricity.
Because PCNSL is confined to the nervous system, a single extranodal site, it is classified as a stage IE NHL; however, it can be disseminated throughout the nervous system, involving the brain, leptomeninges, spinal cord, or eyes.5,16 There may be involvement of one or all of the parts of the nervous system, but PCNSL is usually manifest as a brain tumor. There is a single brain lesion in approximately 70% of immunocompetent individuals8,27 but multifocal lesions are almost always seen in immunocompromised individuals.27 PCNSL usually involves deep structures, frequently periventricular in location, and often abutting the ventricles. Grossly, it is a soft, gray, ill-defined solid lesion. In immunocompetent patients, hemorrhage, cysts, and necrosis are rare, but necrosis is frequently seen in AIDS-related PCNSL. Occasionally, extensive subependymal spread is seen. Microscopically, PCNSL is composed of large cells with round nuclei, vesicular chromatin, and prominent nucleoli; mitoses are frequently seen.28 PCNSL is highly infiltrative, poorly demarcated, and angiocentric.28 The latter is a characteristic feature where aggregates of cells surround blood vessels and infiltrate vessel walls without endothelial reaction or thrombosis (see Figure 2). Vascular proliferation is not seen. Small reactive T lymphocytes, histiocytes, and reactive astrocytes are seen frequently.28 PCNSL is usually leukocyte common antigen positive, which helps distinguish it from gliomas or carcinomas.28 Ultrastructurally, PCNSL lacks intermediate filaments, specific organelles, and intracellular junctions.28 Leptomeningeal infiltration is present in nearly all cases of parenchymal involvement,29 but is usually patchy. Cerebrospinal fluid (CSF) cytology is positive in 26% of patients, but 42% have demonstrable leptomeningeal involvement when pathologic and radiographic examination of the meninges are combined.30 Primary leptomeningeal lymphoma is rare, accounting for only 7% of all PCNSL cases;31 isolated dural-based lesions have also been reported.32 Ocular lymphoma typically involves the retina, choroid, or vitreous. Large gray patches of subretinal and retinal infiltrates or thickened choroid are seen on examination. Lymphoma cells are found between Bruch’s membrane and the retinal pigment epithelium. Evaluation of vitreal specimens
may show malignant lymphocytes along with reactive lymphocytes, histiocytes, necrotic debris, and fibrous material. Immunohistochemistry for B cell markers or identification of a monoclonal κ or λ light chain may be needed to make the diagnosis if standard cytology is inconclusive.33 The origin of the malignant lymphocytes in PCNSL is unknown because the tumor develops in an environment normally devoid of lymphoid tissue. There are several theoretic mechanisms.5 The first suggests that a normal but reactive population of lymphocytes is attracted into the CNS where a second event then transforms a clone of the inflammatory cell population into neoplastic cells. However, the reactive lymphocytes that traffic through the CNS are overwhelmingly T cells, which does not explain this predominantly B cell tumor. The second possibility is that extranodal B lymphocytes, with a propensity for the CNS because of a CNS-specific binding marker, are activated; they then proliferate and transform into neoplastic cells. These cells migrate through the bloodstream into the CNS where they multiply and become tumors, with the original site of disease remaining obscure. However, there is no identifiable difference in the adhesion molecules and the integrins seen in PCNSL and systemic NHL.34 A third possibility is that neoplastic lymphocytes in the periphery are eradicated but persist in the CNS because of poor immune surveillance.27 There are no data to support any of these hypotheses and the pathogenesis of PCNSL remains unknown. The pathogenesis of PCNSL in immunodeficient patients is better understood. The Epstein-Barr virus (EBV) plays an important role in initiating the development of PCNSL in immunocompromised patients. The EBV genome was found in 91% of AIDS-related PCNSLs and 100% of post–organ transplant patients; alternatively, EBV has been found in only 17% of PCNSL in immunocompetent patients.35 In the normal host, EBV-driven B cell lymphoproliferation is inhibited by T cells. This control is absent in immunocompromised patients, thereby facilitating B cell proliferation that can lead to an individual clone evolving into a malignancy. The CNS is a preferred site for EBV-driven lymphomas, presumably because of decreased immune surveillance. Some allogeneic bone marrow transplant recipients have had regression of
PRIMARY CENTRAL NERVOUS SYSTEM LYMPHOMA
their tumor when treated with donor cytotoxic T lymphocytes, which reconstitutes T cell control over the EBV-driven B cells.36 Human herpes virus-6 (HHV-6) has been suggested as an etiologic agent, but confirmatory studies have been negative thus far.37 HHV-8 has been studied as a potential source of chronic antigenic stimulation leading to PCNSL but no association with AIDS or non-AIDS patients was found.38,39 Recent data suggest that Bcl-6 expression, indicating a germinal center phenotype, predicts improved survival in PCNSL patients treated with high-dose methotrexate.40 Proto-oncogenes Bcl-1 and Bcl-2 have no detectable rearrangement in PCNSL.37
CLINICAL PRESENTATION PCNSL occurs in all age-groups with a peak in the sixth and seventh decades for immunocompetent individuals (mean age 55 years) and in the third to fourth decades in immunocompromised individuals (mean age 30 years). There is a 3 : 2 male-to-female ratio among immunocompetent patients but in those with AIDS more than 90% are men. Symptom duration prior to diagnosis averages 2.8 months for immunocompetent patients and 1.8 months for immunocompromised patients.27 The clinical presentation of PCNSL is similar to other intracranial mass lesions (see Table 1). Cognitive and personality changes, reflecting frontal lobe predilection, are the most common symptoms but patients frequently have lateralizing signs as well. Seizures occur in 10% of patients at presentation, a lower percentage than seen with gliomas or brain metastases, due to the subcortical location of most PCNSL lesions. In AIDS patients, however, seizures occur in 25% at presentation.27 Headache is common but other symptoms of raised intracranial pressure are rare. More than 40% of patients have dissemination of PCNSL to the leptomeninges, yet symptoms of leptomeningeal involvement are rare.30 This is in contradistinction to systemic NHL where CNS metastasis primarily affects the leptomeninges, causing multifocal neurologic symptoms. Patients with primary leptomeningeal lymphoma present Table 1 Symptoms and signs of PCNSL.
Cerebral 1. Personality/cognitive changes 2. Lateralizing, i.e. hemiparesis, aphasia 3. Seizures 4. Headache Ocular 1. Floaters 2. Blurred or cloudy vision 3. Decreased visual acuity Leptomeningeal 1. Headache 2. Cranial neuropathy 3. Radiculopathy Spinal cord 1. Back pain 2. Limb paresis 3. Sensory level or paresthesias 4. Bowel or bladder dysfunction
659
with symptoms of increased intracranial pressure along with multifocal signs indicating cranial nerve or multilevel root involvement. These patients do not develop parenchymal or systemic lymphoma.31 Ocular lymphoma can occur in isolation or concomitant with brain lymphoma. It can originate in the eye, but 50–80% of these patients subsequently develop cerebral involvement. This occurs a mean of 23 months (range 1–84 months) after ocular diagnosis.41 Alternatively, 12–20% of patients with brain lymphoma have ocular involvement at diagnosis. Symptoms are usually ‘floaters’, blurred or cloudy vision, and diminished visual acuity, but approximately 50% of patients are asymptomatic. Eye pain and conjunctival hyperemia are rare. Involvement may be unilateral or bilateral but is usually asymmetric. Mean time from symptom onset to diagnosis of intraocular lymphoma is 21.4 months33 because ocular lymphoma is frequently mistaken for chronic vitreitis or uveitis. Fewer than 1% of PCNSL patients present with spinal cord involvement; they may have back pain and symptoms and signs of myelopathy. Lesions are usually seen in the lower cervical or upper thoracic region.42
DIAGNOSTIC CONSIDERATIONS PCNSL is confined to the CNS and does not represent metastatic disease. Therefore, an extent of disease evaluation at diagnosis should focus on the nervous system (see Table 2). All patients should have an enhanced cranial MRI. Every patient should have an ophthalmological evaluation, including a slit lamp examination. MRI of the spine should be performed in patients with signs or symptoms of back pain, myelopathy, or radiculopathy. CSF should be obtained from all patients and analyzed for cell count, protein, glucose, cytology, and tumor markers (lactate dehydrogenase [LDH] isoenzymes, β-glucuronidase, and β-2microglobulin). Immunocytochemical analysis43 and polymerase chain reaction (PCR) for detection of immunoglobulin gene rearrangements44 may be a useful adjunct in the diagnosis of lymphomatous meningitis. An association between interleukin-10 (IL-10) and PCNSL has been reported. Patients with ocular lymphoma have an elevated IL-10 to IL-6 ratio in the vitreal fluid, and those with brain or ocular lymphoma have an elevated IL-10 to IL-6 ratio in the CSF.45 Elevated IL-10 levels correlate with the presence of malignant cells. In patients with AIDS, examination of the CSF for EBV by PCR may establish the diagnosis of PCNSL.46 Demonstrating hypermetabolic lesions on Table 2 Evaluation of patients with PCNSL.
Enhanced brain MRIa Ophthalmologic and slit lamp examination CSF analysis HIV test Enhanced spinal MRI if clinically indicated; a CT scan only if MRI contraindicated. 6. Chest, abdomen, pelvic CT scan 7. Consider body PET scan 8. Bone marrow 1. 2. 3. 4. 5.
a
CT scan only if MRI contraindicated.
660
NEUROLOGICAL MALIGNANCIES
single positron emission computed tomography (SPECT) or positron emission tomographic (PET) scans differentiate PCNSL from infection in AIDS patients with an intracranial mass. When SPECT or PET is positive in combination with identification of EBV in the CSF of an AIDS patient, the diagnosis of PCNSL can be established with accuracy, thus avoiding a brain biopsy.47 Systemic evaluation occasionally identifies another site of disease in PCNSL patients. Comprehensive testing for sites of systemic NHL in 128 patients with PCNSL yielded evidence of systemic disease in only five patients.48 All sites of disease were identified by abdominal CT scan or bone marrow biopsy,48 thereby excluding the need for additional testing. Our own experience confirms the relatively low yield of systemic evaluation in patients with PCNSL although body PET can occasionally reveal a site of disease not appreciated on other imaging studies. In older series, when patients did not undergo systemic workup, only 7% of PCNSL patients ever developed autopsy-proven systemic disease, demonstrating the rarity of multiorgan involvement even late in the course of the illness.4,49
IMAGING MRI is the optimal imaging technique. CT scan should be used only if MRI is unavailable or contraindicated such as in patients with pacemakers. On MRI, PCNSL lesions are typically isointense to hypointense on T1-weighted images. Enhancement tends to be solid and homogeneous in immunocompetent patients (see Figure 3) but irregular and heterogeneous, often with a ring-like pattern, in immunodeficient patients. This ring enhancement is due to central necrosis of the tumor, which makes radiographic distinction from CNS infections difficult. On T2-weighted images, PCNSL is usually diffusely hyperintense but may also be hypointense to isointense.50 Hypointensity on T2 may help differentiate this lesion from gliomas or demyelinating lesions. Peritumoral edema is variable and may be less seen in gliomas or metastases of comparable size. Also, the amount of mass effect is less than expected given the size of the lesion.50 Calcification, hemorrhage, and cyst formation are rare and should raise the suspicion of an alternative pathologic process. Nonenhancing tumors have been reported in approximately 10% of PCNSL patients, both immunocompetent and immunodeficient.27,51 This can present as a focal lesion(s) or as a diffuse infiltrative process, lymphomatosis cerebri. PCNSL tends to be supratentorial, periventricular, and involve the deep structures such as the basal ganglia.27 Ependymal contact is seen in approximately 50% of patients.8 The frontal lobe is the most commonly involved region of the brain, followed in decreasing order by the temporal, parietal, and occipital lobes.8,50 Other areas of involvement include the corpus callosum, hypothalamus and septum, brainstem, and cerebellum.8 On magnetic resonance spectroscopy (MRS) PCNSL has a higher choline/creatinine ratio than all grades of glioma.52 In addition, PCNSL often has marked elevation of lipid and lactate, which may identify patients with a worse prognosis.52,53 On PET, PCNSL is hypermetabolic with a
Figure 3 Typical contrast enhanced MRI showing involvement of the splenium of the corpus callosum. Note the diffuse, homogeneous enhancement pattern.
high fluorodeoxyglucose (FDG) uptake. With methioninePET,54 there is increased lesional uptake and this tracer may be better at delineating lesion size when compared to CT or MRI.
TREATMENT Prognostic Factors Although patient numbers are small, several authors have attempted to define prognostic factors that influence survival. Age and performance status are consistently recognized as influential prognostic factors, with patients younger than 60 and those with a Karnofsky performance status (KPS) greater than 70 surviving longer.13,16,49,55 A prognostic index has been proposed incorporating age, performance status, CSF protein concentration, serum LDH level and tumor involving deep structures on a 5-point scale predicting outcome, but this has yet to be validated.56
Supportive Care Untreated, PCNSL is rapidly fatal (see Table 3). Henry et al.4 found a median survival of 3.3 months for patients treated with supportive care only. A recent analysis of 50 English language papers between 1980 and 1995 found a median survival of 2 months for patients who underwent a biopsy only and no treatment.57
PRIMARY CENTRAL NERVOUS SYSTEM LYMPHOMA Table 3 Therapeutic outcomes for patients with non-AIDS – PCNSL.
Median survival (months)
Reference
Therapy
Supportive care Reni et al.57
None
2.0
Surgery alone Reni et al.57
Surgery
1.5
Radiotherapy alone Nelson et al.49
Radiation
12.0
Standard lymphoma regimens Lachance et al.58 O’Neill et al.59 Schultz et al.60 Brada et al.61
CHOP/RT CHOP/RT CHOD/RT MACOP-B/RT
16.5 10.4 16.0 14.0
MTX-based regimens Glass et al.62 Glass et al.63 DeAngelis et al.64 Abrey et al.65
MCHOD/RT MTX/RT MTX/RT/cytarabine MVP/RT/cytarabine
25.5 33.0 42.5 60.0
Chemotherapy only McAllister et al.66 Batchelor et al.67 Herrlinger et al.68
BBBD/MTX MTX (8 g/m2 ) MTX (8 g/m2 )
40.7 23+ 25
RT, radiation therapy; CHOP, cyclophosphamide, doxorubicin, vincristine, and prednisone; CHOD, cyclophosphamide, doxorubicin, vincristine, and dexamethasone; MACOP-B, methotrexate, doxorubicin, cyclophosphamide, vincristine, prednisone, and bleomycin; MCHOD, methotrexate, cyclophosphamide, doxorubicin, vincristine, and decadron; MTX, methotrexate; MVP, methotrexate, vincristine, and procarbazine; BBBD, blood – brain barrier disruption.
Surgery Surgical resection does not prolong survival (see Table 3). Median survival with surgery alone ranges from 1 month69 to 6 months.4,70 Unlike gliomas, survival is not affected by gross total or incomplete resection.69 Surgical resection may even decrease median survival, from 2 months with biopsy only, to 1.5 months.57 Furthermore, resection may worsen a patient’s neurologic function due to the deep location of most PCNSL lesions. In a patient with acute neurologic deterioration due to herniation, a resection may be indicated for immediate decompression; however, biopsy is the preferred method for tissue diagnosis and is usually safe regardless of tumor site. A frozen section could be performed intraoperatively and if consistent with PCNSL, the neurosurgeon should proceed no further.
Radiation Therapy PCNSL is an exquisitely radiosensitive tumor and radiation therapy prolongs survival (see Table 3). In a large multicenter Radiation Therapy Oncology Group (RTOG) trial, Nelson et al.49 found a median survival of 11.6 months from the onset of treatment and 12.6 months from the time of diagnosis. Out of the 41 patients, 21 failed locally at the original site of disease, four failed locally and had distant CNS or extraneural disease, and three had only distant extraneural disease. The RTOG data are similar to historical series where median survivals range from 10 to 18 months69 – 72 despite the RTOG’s use of high-dose and large-volume radiation therapy (RT). Despite CSF dissemination in many patients,
661
craniospinal (CS) RT does not improve survival, as seen in a report by Reni et al.57 where median survival for patients treated with whole-brain radiation therapy (WBRT) was 17 months but only 14 months for those receiving CS RT. Brada et al.61 also found no survival advantage to CS versus cranial radiotherapy alone. In addition, CS RT irradiates a large volume of bone marrow, compromising subsequent administration of chemotherapy. WBRT is the recommended treatment because of the multifocal and infiltrative nature of PCNSL and the widespread microscopic disease seen in patients at autopsy. Shibamoto et al. reviewed patients with PCNSL treated with focal RT using margins <4 cm compared to those treated with margins ≥4 cm. Patients treated with the smaller margins had a significantly higher incidence (83%) of out-field recurrences than patients with larger margins (22%).73 These data support the use of the whole-brain (WB) port to achieve local control. The optimal dose of WBRT has not been defined but most data indicate the need for at least 4000 cGy.49,69 In the RTOG study, 4000 cGy WBRT and a 2000 cGy boost to the tumor did not result in improved disease control since most patients relapsed within the boosted field.49 These data confirm our own experience using 4000 cGy WBRT and a 1440 cGy boost.74 Therefore, patients do not appear to benefit from doses greater than 5000 cGy. Furthermore, Blay et al.55 noted increased delayed neurotoxicity in patients radiated with greater than 5000 cGy. Currently, when using RT, we treat patients with 4500 cGy WBRT and no boost. Efforts to reduce the dose of WBRT to decrease the incidence of neurotoxicity have produced mixed results. The RTOG conducted a study using a high-dose methotrexatebased regimen plus 4500 cGy WBRT.75 Patients who achieved a complete response (CR) from chemotherapy received only 3600 cGy WBRT. Survival, disease control, and neurotoxicity were identical regardless of WBRT dose. However, Bessell et al. used a different preradiation chemotherapy protocol and reduced the WBRT dose from 4500 cGy to 3060 cGy in those who achieved a CR.76 No difference in outcome was seen in patients older than 60 years but younger patients had significantly longer survival if the full dose of WBRT was used. There was no information concerning neurotoxicity. These data are inconclusive regarding the ability of chemotherapy to permit safe dose reduction of WBRT. In patients with ocular lymphoma, RT should include the posterior two-thirds of the globe to a dose of 3000–4000 cGy.41 At these doses improvements in visual acuity are seen, but frequently patients relapse within the brain, and this is the ultimate cause of death. Ocular toxicities include epithelioid keratopathy, posterior subcapsular cataracts, and radiation retinopathy or optic neuropathy. If cerebral involvement is present at the same time, the treatment plan should include both regions simultaneously to eliminate overlapping fields and minimize toxicity to the optic nerve and retina.
Chemotherapy Like systemic NHL, PCNSL is highly sensitive to glucocorticoids and has been called a “ghost tumor”77 because the
662
NEUROLOGICAL MALIGNANCIES
lesion may vanish with steroid use. About 40% of patients treated with steroids will have tumor shrinkage, but the tumor usually recurs within a short period.77 Half of 20 patients treated with steroids had an evaluable response ranging from complete to minor, with two having a CR lasting 6 and 15 months prior to needing further therapy.64 There are other reports of long-term remissions in patients treated with glucocorticoids lasting from 6 to 60 months but this is rare.78 Some patients may have repeated responses to steroids.79 Corticosteroids are chemotherapeutic agents for PCNSL and lesion disappearance is due to cell lysis, not reconstitution of the blood–brain barrier (BBB). Because steroids are oncolytic, they should be withheld until a tissue diagnosis is established. Furthermore, steroid response should not be used as a diagnostic test for PCNSL because other CNS processes, such as multiple sclerosis or sarcoidosis, can have a similar radiographic appearance and regress after steroid administration. Advanced systemic NHL is optimally treated with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) chemotherapy.80 Lachance et al.58 treated six PCNSL patients with pre-RT CHOP (see Table 3). All responded initially but developed progressive disease, frequently between the second and third cycle of chemotherapy. Median survival was 8.5 months for the group but 16.5 months for the four patients who were able to receive RT after chemotherapy. The authors terminated the trial early because patients did so poorly. O’Neill et al.59 completed a multicenter trial using preirradiation CHOP. The median survival for the 46 patients was 41.7 weeks but survival was better for patients younger than 60 years of age (47.4 weeks) than for those older than 60 (32.9 weeks). In addition, significant chemotherapy-related toxicities were seen and only 54% of patients ever proceeded to RT in this study. The RTOG60 used preirradiation CHOD (substituting dexamethasone for prednisone) and found a median survival of 16 months and a median diseasefree survival of 9.2 months. In a randomized control trial that was terminated before completion because of poor accrual, the addition of CHOP to RT did not prolong survival over WBRT alone.81 Brada et al.61 used MACOP-B (methotrexate [0.4 g m−2 ], doxorubicin, cyclophosphamide, vincristine, prednisone, and bleomycin) followed by whole brain or CS radiation in 10 patients. There was a median survival of only 14 months compared to those treated with RT only who had a median survival of 16 months. These studies demonstrate that the addition of WBRT to chemotherapy regimens efficacious for systemic NHL offers no survival advantage over RT alone. These regimens use anthracyclines and cyclophosphamide, neither of which effectively penetrates the intact BBB. Patients may have an initial response because of BBB disruption in areas of bulky disease, but recurrence is seen, often in areas remote from the initial tumor site, where lymphoma was present behind a relatively intact BBB. Ervin and Canellos82 had one of the earliest reports of a patient with recurrent PCNSL, after surgery and radiation, who had a CR to high-dose methotrexate (3 g m−2 ) and remained disease-free for at least 12 months. Glass et al.62 added MTX (3.5 g m−2 ) to CHOD prior to WBRT. They
found a median survival of 25.5 months, which was improved over the CHOD or CHOP regimens used by others. Glass et al.63 used the same dose of methotrexate (3.5 g m−2 ) prior to RT in 25 patients, of whom 14 had a CR to chemotherapy and eight had a PR; 20 of the 25 had a CR after RT. Median survival for the group was 33 months but for responders the median survival was 42.5 months. In 1992, DeAngelis et al.74 reported results using MTX (1 g m−2 ), WBRT, and then highdose cytarabine (3 g m−2 ) to treat 31 patients. They found a median disease-free survival of 41 months and a median overall survival of 42.5 months. At recent follow-up, seven of the original 31 patients were still alive (47–126 plus months), six of whom were less than 50 years of age at diagnosis.83 The 5-year survival rate is 22%, markedly better than the 3–4% rate with RT alone.49 Blay et al.55 found overall survival was significantly improved in patients treated with methotrexate-based regimens (1.5 g m−2 or higher) compared to other regimens, irrespective of age. There was also a significant survival advantage in standard-dose cytarabinecontaining regimens. The use of single agent, very highdose methotrexate (8 g m−2 ) with the intention of deferring radiotherapy has produced conflicting results. Batchelor et al. treated 25 patients with methotrexate alone for over a year and reported a 74% response rate and median progressionfree survival of 12.8 months with overall survival not reached at 23 months follow-up.67 In a similar study, Herrlinger et al. had a comparable relapse-free survival of 13.7 months in patients with a CR but 38% of patients progressed and overall survival was only 25 months.68,84 They abandoned the study early because of patients’ poor response. Multidrug regimens including methotrexate have been used to treat PCNSL. Methotrexate (3.5 g m−2 ), procarbazine, and vincristine is an effective combination given prior to WBRT, achieving a median survival of 60 months.65 A large multicenter trial using a comparable regimen, but with a lower dose of methotrexate (2.5 g m−2 ), achieved a median survival of 37 months.75 Despite initial high response rates using any high-dose methotrexate-based regimen, some patients will progress while many will relapse. A recent study examined the potential role for highdose chemotherapy followed by autologous stem cell rescue without cranial RT. Overall event-free survival was only 5.6 months for all patients and 9.3 months for patients who underwent transplantation.85 However, overall survival had not been reached with a median follow-up of 28 months. This approach warrants further investigation but should remain confined to a clinical trial. An alternative approach has been promoted by McAllister et al.66 They treated patients with sequential cyclophosphamide, intra-arterial (IA) mannitol for BBB disruption and then IA methotrexate (2.5 g), followed by 14 days of procarbazine and dexamethasone. Seventy-four patients were treated and the median survival was 40.7 months; however, more than 25% received WBRT. These results are better than WBRT alone but not superior to standard systemic high-dose methotrexate regimens. Furthermore, this technique is invasive and has acute morbidity, such as acute arterial injuries, stroke, and seizures.
PRIMARY CENTRAL NERVOUS SYSTEM LYMPHOMA
Recurrent Disease Most patients with PCNSL will develop recurrence and there is no standard therapeutic approach. In part, prior treatment and location of relapse, e.g. isolated ocular relapse determine the choice of therapy. However, most relapses occur within the brain and within the first 2 years of diagnosis. Radiotherapy is an option if not previously administered. Some chemotherapeutic regimens that are reportedly useful include rituximab and temozolomide;86 topotecan;87 and repeating methotrexate.88
Neurotoxicity Leukoencephalopathy is a serious complication of effective PCNSL treatment, but apparent only when the patient is in a durable remission.55,83 Therefore, it is rarely observed in patients treated with WBRT alone because they die of tumor progression, but it has been seen in patients treated with combined modality therapy, especially high-dose methotrexate-based regimens. There is synergistic toxicity when methotrexate is combined with WBRT, which can be minimized when the methotrexate is administered prior to RT. However, it has also been observed in patients treated with single-agent methotrexate and no cranial irradiation.68
663
Blay et al.55 found that RT followed by chemotherapy and radiation doses greater than 5000 cGy were risk factors for late neurotoxicity. The projected incidences of neurotoxicity in their series were 4, 8, and 26% at 12, 24, and 68 months, respectively. Treatment-related leukoencephalopathy occurs primarily in patients older than 60 years of age. These patients present with a syndrome similar to normal pressure hydrocephalus with significant cognitive impairment, gait ataxia, and incontinence; some improve with ventriculoperitoneal shunting.88 Given the high incidence of this treatment complication in older patients, we treat these patients with chemotherapy alone (see Figure 4) and avoid WBRT. In older patients, eliminating WBRT does not compromise outcome.65
OCULAR LYMPHOMA Ocular lymphoma is uncommon, so defining an optimal therapy is difficult. Hormigo et al.41 reported 31 patients with ocular lymphoma who were treated with a variety of regimens including topical or systemic steroids, WBRT, ocular RT, and chemotherapy. Most of the patients had improvement in their symptoms but relapse was common in both the brain and the eyes. Baumann et al.89 reported
Figure 4 Patient with multifocal PCNSL before (left) and after (right) chemotherapy with high-dose methotrexate, procarbazine, and vincristine, and no radiation therapy. There is a complete regression of all lesions. The patient was off corticosteroids when response was assessed.
664
NEUROLOGICAL MALIGNANCIES
a patient with ocular lymphoma who responded to highdose cytarabine and had therapeutic drug levels measured in intraocular fluid. Cytarabine was then used as sole treatment in six patients with ocular lymphoma; one complete and four partial responses were achieved prior to RT.90 Batchelor et al. treated nine patients with ocular lymphoma using single agent high-dose methotrexate (8 g m−2 ). All patients achieved a therapeutic concentration of drug in both the aqueous and vitreous humor four hours after completion of the infusion. Seven patients had an ocular response but three relapsed requiring orbital radiotherapy.91 Recently, patients with ocular lymphoma have been successfully treated with intravitreal methotrexate.92 Response is usual and adverse reactions are rare.
50, and a single cerebral lesion fared better. Three patients of Baumgartner et al.93 received high-dose methotrexate and RT and had survivals of 122, 245, and 380 days. Chamberlain treated four AIDS patients with concomitant WBRT and hydroxyurea followed by chemotherapy with procarbazine, lomustine, and vincristine.96 The median survival for the group was 14.5 months (range of 11–16 months). These studies suggest there is a role for combined modality therapy or possibly chemotherapy alone in AIDS-PCNSL patients but it should be reserved for those with good KPS and no concomitant opportunistic infections. In post-allogeneic bone marrow transplant lymphoproliferative disorders, donor cytotoxic T lymphocytes sensitized to EBV have caused tumor regression.36 Whether this type of therapy can be modified and used in AIDS is being explored.
PCNSL AND AIDS AIDS-related PCNSL is more aggressive than PCNSL in the immunocompetent population. Baumgartner et al.93 found clinical and radiographic response to WBRT in 76 and 69% of AIDS patients, respectively (see Table 4). Median survival for those without treatment was 27 days, but 119 days for those completing the RT protocol. In patients treated with RT, the cause of death was opportunistic infection in 87%, whereas progressive PCNSL caused death in 77% when untreated. Ling et al.19 reported a 71% mild to excellent clinical response in AIDS patients treated with WBRT, but median survival was only 3 months. Because PCNSL in AIDS patients is associated with severe immune suppression, treatment directed against the HIV infection is essential. The institution of HAART is associated with significantly longer survival, independent of specific treatment for the lymphoma.94 Experience with chemotherapy in AIDS-related PCNSL is more limited than in the immunocompetent population. Jacomet et al.95 treated 15 AIDS patients with high-dose methotrexate (3 g m−2 ) and observed a CR in 47% of patients with a median survival of 19 months. Combined chemotherapy with intravenous methotrexate (3 g m−2 ) intrathecal – methotrexate, procarbazine, thio-TEPA, and WBRT was used in 10 patients with AIDS.92 Eight patients completed both chemotherapy and RT; six had a CR with a median survival of 7 months and two patients survived more than 1 year. Although the sample was small, it appeared that patients with a CD4 count greater than 50 cells mm−3 , KPS greater than Table 4 Treatment efficacy for AIDS-related PCNSL.
Reference 93
Baumgartner et al. Baumgartner et al.93 Skiest et al.94 Jacomet et al.95 Forsyth et al.92 Chamberlain96
Number of patients
Treatment
Median survival (days)
17 29 12 13 7 8 4
None RT None RT MTX MTX/RT RT/PCV
27 (8 – 127) 119 (33 – 380) 29 About 350 570 (170 – 575) 210 (60 – 1620) 435 (330 – 480)
RT, radiation therapy; MTX, methotrexate; PCV, procarbazine, lomustine, and vincristine.
REFERENCES 1. Bailey P. Intracranial sarcomatous tumors of leptomeningeal origin. Arch Surg 1929; 18: 1359 – 402. 2. Yuile CL. Case of primary reticulum cell sarcoma of the brain. Relationship of microglia cells to histiocytes. Arch Pathol 1938; 26: 1037 – 44. 3. Russell DS, Marshall AHE, Smith FB. Microgliomatosis. A form of reticulosis affecting the brain. Brain 1948; 71: 1 – 15. 4. Henry JM, et al. Primary malignant lymphomas of the central nervous system. Cancer 1974; 34: 1293 – 302. 5. Hochberg FH, Miller DC. Primary central nervous system lymphoma. J Neurosurg 1988; 68: 835 – 53. 6. Nakhleh RE, et al. Central nervous system lymphomas. Immunohistochemical and clinicopathologic study of 26 autopsy cases. Arch Pathol Lab Med 1989; 113: 1050 – 6. 7. Taylor CR, et al. An immunohistological study of immunoglobulin content of primary central nervous system lymphomas. Cancer 1978; 41: 2197 – 205. 8. Tomlinson FH, et al. Primary intracerebral lymphoma: a clinicopathological study of 89 patients. J Neurosurg 1995; 82: 558 – 66. 9. Jellinger KA, Radaskiewicz TH, Slowik F. Primary malignant lymphomas of the central nervous system in man. Acta Neuropathol 1975; (suppl 6): 95 – 102. 10. Zimmerman HM. Malignant lymphomas of the nervous system. Acta Neuropathol 1975; (suppl 6): 69 – 74. 11. Olson JE, et al. The continuing increase of primary central nervous system non-Hodgkins lymphoma. A surveillance, epidemiology, and end results analysis. Cancer 2002; 95: 1504 – 10. 12. Kadan-Lottick NS, Skluzacek MC, Gurney JG. Decreasing incidence rates of primary central nervous system lymphoma. Cancer 2002; 95: 193 – 202. 13. Krogh-Jensen M, et al. Clinicopathological features, survival, and prognostic factors of primary central nervous system lymphomas: trends in incidence of primary central nervous system lymphomas and primary malignant brain tumors in a well-defined geographic area. Leuk Lymphoma 1995; 19: 223 – 33. 14. Yau Y, et al. Primary lymphoma of the central nervous system in immunocompetent patients in south-east Scotland. Lancet 1996; 348: 890. 15. Hao D, et al. Is the incidence of primary CNS lymphoma increasing? A population-based study of incidence, clinicopathological features, and outcomes in Alberta from 1975 to 1996. Ann Neurol 1997; 42: 537. 16. DeAngelis LM, Yahalom J. Primary central nervous system lymphoma. In DeVita VTJ, Hellman S, Rosenberg SA (eds) Principles and Practice of Oncology Updates, 7th ed. Philadelphia, Pennsylvania: JB Lippincott, 2000: 2012. 17. Rosenblum ML, et al. Primary central nervous system lymphomas in patients with AIDS. Ann Neurol 1988; (suppl 23): S13 – 6. 18. Welch K, et al. Autopsy findings in the acquired immune deficiency syndrome. JAMA 1984; 252: 1152 – 9.
PRIMARY CENTRAL NERVOUS SYSTEM LYMPHOMA 19. Ling SM, et al. Radiotherapy of primary central nervous system lymphoma in patients with and without human immunodeficiency virus. Ten years of treatment experience at the University of California San Francisco. Cancer 1994; 73: 2570 – 82. 20. Wolf T, et al. Changing incidence and prognostic factors of survival in AIDS-related non-Hodgkin’s lymphoma in the era of Highly Active Antiretroviral Therapy (HAART). Leuk Lymphoma 2005; 467: 207 – 15. 21. DeAngelis LM. Primary central nervous system lymphoma as a secondary malignancy. Cancer 1991; 67: 1431 – 5. 22. Reni M, et al. Primary brain lymphomas in patients with a prior or concomitant malignancy. J Neurooncol 1997; 32: 135 – 42. 23. Bashir R, et al. Immunophenotypic profile of CNS lymphoma: a review of eighteen cases. J Neurooncol 1989; 7: 249 – 54. 24. Gijtenbeek JM, Rosenblum MK, DeAngelis LM. Primary central nervous system T-cell lymphoma. Neurology 2001; 57(4): 716 – 8. 25. Itoyama T, et al. Primary central nervous system lymphomas. Cancer 1994; 73: 455 – 63. 26. Kumanishi T, et al. Primary malignant lymphoma of the brain: demonstration of frequent p16 and p15 gene deletions. Jpn J Cancer Res 1996; 87: 691 – 5. 27. Fine HA, Mayer RJ. Primary central nervous system lymphoma. Ann Intern Med 1993; 119: 1093 – 104. 28. Paulus W, et al. Malignant lymphomas In Kleihues P Cavenee WK (eds) World Health Organization Classification of Tumours, Pathology & Genetics, Tumours of the Nervous System, Lyon, France: IARC Press, 2000: 198 – 203. 29. Schaumberg HH, Plank CR, Adams RD. The reticulum cell sarcomamicroglioma group of brain tumors. A consideration of their clinical features and therapy. Brain 1972; 95: 199 – 212. 30. Balmaceda C, et al. Leptomeningeal tumor in primary central nervous system lymphoma: recognition, significance, and implications. Ann Neurol 1995; 38: 202 – 9. 31. Lachance DH, et al. Primary leptomeningeal lymphoma: report of 9 cases, diagnosis with immunocytochemical analysis, and review of the literature. Neurology 1991; 41: 95 – 100. 32. Miranda RN, et al. Stage 1E non-Hodgkin’s lymphoma involving the dura. A clinicopathologic study of five cases. Arch Pathol Lab Med 1996; 120: 254 – 60. 33. Whitcup SM, et al. Intraocular lymphoma: clinical and histopathologic diagnosis. Ophthalmology 1993; 100: 1399 – 406. 34. Paulus W, Jellinger K. Comparison of integrin adhesion molecules expressed by primary brain lymphomas and nodal lymphomas. Acta Neuropathol 1993; 86: 360 – 4. 35. Morgello S. Pathogenesis and classification of primary central nervous system lymphoma: an update. Brain Pathol 1995; 5: 383 – 93. 36. Papadopoulos EB, et al. Infusions of donor leukocytes to treat Epstein – Barr virus-associated lymphoproliferative disorders after allogeneic bone marrow transplantation. N Engl J Med 1994; 330: 1185 – 91. 37. Jellinger KA, Paulus W. Primary central nervous system lymphomas – new pathological developments. J Neurooncol 1995; 24: 33 – 6. 38. Corby JR, Garl PJ, Kleinschmidt – Demasters BK. HHV-8 DNA in CNS lymphomas from AIDS and non-AIDS patients. Neurology 1998; 50: 335 – 40. 39. Montesinos-Rongen M, et al. Human herpes virus-8 is not associated with primary central nervous system lymphoma in HIV-negative patients. Acta Neuropathol 2001; 102: 489 – 95. 40. Braaten KM, et al. BCL-6 expression predicts improved survival in patients with primary central nervous system lymphoma. Clin Cancer Res 2003; 9: 1063 – 9. 41. Hormigo A, et al. Ocular presentation of primary central nervous system lymphoma: diagnosis and treatment. Br J Haematol 2004; 126: 202 – 8. 42. Schild SE, et al. Primary lymphoma of the spinal cord. Mayo Clin Proc 1995; 70: 256 – 60. 43. Li C.-Y, et al. Diagnosis of B-cell non-Hodgkin’s lymphoma of the central nervous system by immunocytochemical analysis of cerebrospinal fluid lymphocytes. Cancer 1986; 57: 737 – 44. 44. Gleissner B, et al. CSF evaluation in primary CNS lymphoma patients by PCR of the CDR III IgH genes. Neurology 2002; 58: 390 – 6.
665
45. Whitcup SM, et al. Association of interleukin 10 in the vitreous and cerebrospinal fluid and primary central nervous system lymphoma. Arch Ophthalmol 1997; 115: 1157 – 60. 46. Arribas JR, et al. Detection of Epstein – Barr virus DNA in cerebrospinal fluid for diagnosis of AIDS-related central nervous system lymphoma. J Clin Microbiol 1995; 33: 1580 – 3. 47. Antinori A, et al. Value of combined approach with thallium-201 singlephoton emission computed tomography and Epstein-Barr virus DNA polymerase chain reaction in CSF for the diagnosis of AIDS-related primary CNS lymphoma. J Clin Oncol 1999; 17: 554 – 60. 48. O’Neill BP, et al. Occult systemic Non-Hodgkin’s Lymphoma (NHL) in patients initially diagnosed as Primary Central Nervous System Lymphoma (PCNSL): how much staging is enough? J Neurooncol 1995; 25: 67 – 71. 49. Nelson DF, et al. Non-Hodgkin’s lymphoma of the brain: can high dose, large volume radiation therapy improve survival? Report on a prospective trial by the Radiation Therapy Oncology Group (RTOG): RTOG 8315. Int J Radiat Oncol Biol Phys 1992; 23: 9 – 17. 50. Koeller KK, Smirniotopoulos JG, Jones RV. Primary central nervous system lymphoma: radiologic – pathologic correlation. Radiographics 1997; 17: 1497 – 526. 51. DeAngelis LM. Cerebral lymphoma presenting as a nonenhancing lesion on computed tomographic/magnetic resonance scan. Ann Neurol 1993; 33: 308 – 11. 52. Harting I, et al. Differentiating primary central nervous system lymphoma from glioma in humans using localized proton magnetic resonance spectroscopy. Neurosci Lett 2003; 342: 163 – 6. 53. Raizer JJ, et al. Proton magnetic resonance spectroscopy in immunocompetent patients with primary central nervous system lymphoma. J Neurooncol 2005; 71(2): 173 – 80. 54. Ogawa T, et al. Methionine PET for follow-up of radiation therapy of primary lymphoma of the brain. Radiographics 1994; 14: 101 – 10. 55. Blay J.-Y, et al. High-dose methotrexate for the treatment of primary cerebral lymphomas: analysis of survival and late neurologic toxicity in a retrospective series. J Clin Oncol 1998; 16: 864 – 71. 56. Ferreri AJ, et al. Prognostic scoring system for primary CNS lymphomas: the International Extranodal Lymphoma Study Group experience. J Clin Oncol 2003; 21: 266 – 72. 57. Reni M, et al. Therapeutic management of primary central nervous system lymphoma in immunocompetent patients: results of a critical review of the literature. Ann Oncol 1997; 8: 227 – 34. 58. Lachance DH, et al. Cyclophosphamide, doxorubicin, vincristine, and prednisone for primary central nervous system lymphoma: shortduration response and multifocal intracerebral recurrence preceding radiotherapy. Neurology 1994; 44: 1721 – 7. 59. O’Neill BP, et al. Primary central nervous system non-Hodgkin’s lymphoma: survival advantages with combined initial therapy? Int J Radiat Oncol Biol Phys 1995; 33: 663 – 73. 60. Schultz C, et al. Preirradiation chemotherapy with cyclophosphamide, doxorubicin, vincristine, and dexamethasone for primary CNS lymphomas: initial report of Radiation Therapy Oncology Group protocol 88-06. J Clin Oncol 1996; 14: 556 – 64. 61. Brada M, et al. Management of primary cerebral lymphoma with initial chemotherapy: preliminary results and comparison with patients treated with radiotherapy alone. Int J Radiat Oncol Biol Phys 1990; 18: 787 – 92. 62. Glass J, et al. Therapy of primary central nervous system lymphoma with pre-irradiation Methotrexate, Cyclophosphamide, Doxorubicin, Vincristine, and Dexamethasone (MCHOD). J Neurooncol 1996; 30: 257 – 65. 63. Glass J, et al. Preirradiation metho-trexate chemotherapy of primary central nervous system lymphoma: long-term outcome. J Neurosurg 1994; 81: 188 – 95. 64. DeAngelis LM, et al. Primary CNS lymphoma: combined treatment with chemotherapy and radiotherapy. Neurology 1990; 40: 80 – 6. 65. Abrey LE, Yahalom J, DeAngelis LM. Treatment for primary CNS lymphoma: the next step. J Clin Oncol 2000; 18: 3144 – 50. 66. McAllister LD, et al. Cognitive outcomes and long-term follow-up results after enhanced chemotherapy delivery for primary central nervous system lymphoma. Neurosurgery 2000; 46: 51 – 60.
666
NEUROLOGICAL MALIGNANCIES
67. Batchelor T, et al. Treatment of primary CNS lymphoma with methotrexate and deferred radiotherapy: a report of NABTT 96-07. J Clin Oncol 2003; 21(6): 1044 – 9. 68. Herrlinger U, et al. Neuro-Oncology Working Group of the German Society. NOA-03 trial of high-dose methotrexate in primary central nervous system lymphoma: final report. Ann Neurol 2005; 57(6): 843 – 7. 69. Murray K, Kun L, Cox J. Primary malignant lymphoma of the central nervous system. Results of treatment of 11 cases and review of the literature. J Neurosurg 1986; 65: 600 – 7. 70. Berry MP, Simpson WJ. Radiation therapy in the management of primary malignant lymphoma of the brain. Int J Radiat Oncol Biol Phys 1981; 7: 55 – 9. 71. Sagerman RH, Collier CH, King GA. Radiation therapy of microgliomas. Radiology 1983; 149: 567 – 70. 72. Mendenhall NP, et al. Primary lymphoma of the central nervous system: computerized tomography scan characteristics and treatment results for 12 cases. Cancer 1983; 52: 1993 – 2000. 73. Shibamoto Y, et al. Is whole-brain irradiation necessary for primary central nervous system lymphoma? Patterns of recurrence after partialbrain irradiation. Cancer 2003; 97: 128 – 33. 74. DeAngelis LM, et al. Combined modality therapy for primary CNS lymphoma. J Clin Oncol 1992; 10: 635 – 43. 75. DeAngelis LM, Seiferheld W, Schold SC. Combination chemotherapy and radiotherapy for primary central nervous system lymphoma: Radiation Therapy Oncology Group Study 93-10. J Clin Oncol 2002; 20: 4643 – 8. 76. Bessell EM, et al. Importance of radiotherapy in the outcome of patients with primary CNS lymphoma: an analysis of the CHOD/BVAM regimen followed by two different radiotherapy treatments. J Clin Oncol 2002; 20: 231 – 6. 77. Weller M. Glucocorticoid treatment of primary CNS lymphoma. J Neurooncol 1999; 43: 237 – 9. 78. Pirotte B, et al. Glucocorticoid-induced long-term remission in primary cerebral lymphoma: case report and review of the literature. J Neurooncol 1997; 32: 63 – 9. 79. Singh A, et al. Steroid-induced remissions in CNS lymphoma. Neurology 1982; 32: 1267 – 71. 80. Fischer RI, et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin’s lymphoma. N Engl J Med 1993; 328: 1002 – 6. 81. Mead GM, et al. A medical research council randomized trial in patients with primary central non-Hodgkin lymphoma. Cerebral radiotherapy with and without cyclophosphamide, doxorubicin, vincristine, and prednisone chemotherapy. Cancer 2000; 89: 1359 – 70. 82. Ervin T, Canellos GP. Successful treatment of recurrent primary central nervous system lymphoma with high-dose methotrexate. Cancer 1980; 45: 1556 – 7.
83. Abrey LE, DeAngelis LM, Yahalom J. Long-term survival in primary central nervous system lymphoma. J Clin Oncol 1998; 16: 1 – 6. 84. Herrlinger U, et al. German Cancer Society Neuro-Oncology Working Group NOA-03 multicenter trial of single-agent high-dose methotrexate for primary central nervous system lymphoma. Ann Neurol 2002; 51(2): 247 – 52. 85. Abrey LE, et al. Intensive methotrexate and cytarabine followed by high-dose chemotherapy with autologous stem cell rescue in patients with newly diagnosed primary CNS lymphoma: An intent-to-treat analysis. J Clin Oncol 2005; 21: 4151 – 6. 86. Enting RH, et al. Salvage therapy for primary CNS lymphoma with a combination of rituximab and temozolomide. Neurology 2004; 63(5): 901 – 3. 87. Fischer L, et al. Response of relapsed or refractory Primary Central Nervous System Lymphoma (PCNSL) to topotecan. Neurology 2004; 62(10): 1885 – 7. 88. Thiessen B, DeAngelis LM. Hydrocephalus in radiation leukoencephalopathy. Results of Ventriculoperitoneal Shunting. Arch Neurol 1998; 55: 705 – 10. 89. Baumann MA, et al. Treatment of intraocular lymphoma with high dose Ara-C. Cancer 1986; 57: 1273 – 5. 90. Strauchen JA, Dalton J, Friedman AH. Chemotherapy in the management of intraocular lymphoma. Cancer 1989; 63: 1918 – 21. 91. Batchelor TT, et al. High-dose methotrexate for intraocular lymphoma. Clin Cancer Res 2003; 9(2): 711 – 5. 92. Forsyth PA, Yahalom J, DeAngelis LM. Combined-modality therapy in the treatment of primary central nervous system lymphoma in AIDS. Neurology 1994; 44: 1473 – 9. 93. Baumgartner JE, et al. Primary central nervous system lymphomas: natural history and response to radiation therapy in 55 patients with acquired immunodeficiency syndrome. J Neurosurg 1990; 73: 206 – 11. 94. Skiest DJ, Crosby C. Survival is prolonged by highly active antiretroviral therapy in AIDS patients with primary central nervous system lymphoma. AIDS 2003; 17(12): 1787 – 93. 95. Jacomet C, et al. Intravenous methotrexate for primary central nervous system non-Hodgkin’s lymphoma in AIDS. AIDS 1997; 11: 1725 – 30. 96. Chamberlain MC. Long survival of patients with acquired immune deficiency syndrome-related primary central nervous system lymphoma. Cancer 1994; 73: 1728 – 30.
FURTHER READING Kim SK, Chan CC, Wallace DJ. Management of primary intraocular lymphoma. Curr Oncol Rep 2005; 7(1): 74 – 9. Plotkin SR, et al. Treatment of relapsed central nervous system lymphoma with high-dose methotrexate. Clin Cancer Res 2004; 10(17): 5643 – 6.
Section 10 : Neurological Malignancies
61
Choroid Plexus Papilloma and Carcinoma Michael L. Edgeworth and Julie E. Hammack
INTRODUCTION AND HISTORICAL BACKGROUND The first report of a choroid plexus papilloma (CPP) was described in 1833 by Guerard.1 In 1840, Rokitansky described a supposed primary choroid plexus carcinoma (CPC),2 and in 1868, LeBlanc is credited with reporting the first case of distant seeding of a choroid plexus tumor (CPT).3 The first surgical attempt, although unsuccessful due to resulting death, was reported by Bielschowsky and Unger in 1902.4 This was followed by the first successful surgical resection by Perthes in 1919.5 Imaging for suspected CPP/CPC has evolved over the past century from plain radiography of the skull in the early 20th century to pneumoencephalography in the 1920s, and to cerebral angiography in the 1930s. These procedures would dominate neuroscience for the next 40 years until the discovery of computed tomography (CT) and ultimately magnetic resonance imaging (MRI). Because of the rather nonspecific clinical presentation, antemortem diagnosis was often not made prior to the modern imaging era with CT and MRI.6 Treatment recommendations have held fast to the notion that aggressive surgical resection is the treatment of choice for CPP and CPC; however, various pre- and postoperative therapies have been implemented. Van Wagenen, in 1929, reported size reduction of a CPP using radiation therapy (RT) allowing for gross total resection (GTR) at a second stage operation.7 Because of the rarity of this tumor, no study has been randomized using radiotherapy or chemotherapy, although success has been reported in various single and small series reports noted below.
ANATOMY Intuitively, these tumors are almost always intraventricular in location. Most pediatric CPP and CPC arise from the choroid plexus of the lateral ventricle (usually near the trigone). The left lateral ventricle appears to be more likely
involved than the right, and occasionally the tumor may be bilateral.7 – 11 Most adult tumors arise from the fourth ventricle or cerebellopontine angle (CPA). Those tumors arising in the CPA presumably originate from the tuft of choroid within the foramen of Luschka. Origin of the tumor within the third ventricle is rare,12 – 14 but is occasionally seen in both adult and pediatric patients. Bilateral ventricular involvement has been reported,15,16 and there is also one case report of a CPC arising in the left frontal lobe without ventricular system involvement.17 Hydrocephalus is common in both CPP and CPC and is often the cause of clinical signs and symptoms. Mechanical obstruction of the ventricular system, cerebrospinal fluid (CSF) overproduction by the tumor, reduced CSF resorption, or a combination of all these factors are likely causes.18 – 23 These tumors are highly vascular and often associated with hemorrhage and xanthochromic CSF, as well as flow voids on MRI.6,24,25 Dissemination of tumor via ventricular and subarachnoid space is far more common in CPC than in CPP as is invasion of surrounding brain. Metastases have been reported within intraparenchymal tissue.26 – 28 Extraneural metastases have also been reported but are exceedingly rare.26,29
BIOLOGY AND EPIDEMIOLOGY CPP and CPC are rare epithelial-derived neoplasms that account for 0.4–0.6% of all brain tumors.6,30,31 These tumors are more common in the pediatric age-group, in which they account for between 2 and 4% of all intracranial tumors depending on the series.20,32 – 35 The vast majority of choroid plexus neoplasms occur in children less than 2 years of age.32,36 These tumors have even been diagnosed in utero via fetal ultrasound.37 CPPs are considerably more common than CPCs by a factor of about 5 : 1 in most series,38,39 and it has been estimated that fewer than 20 CPCs are diagnosed in the United States each year.40 These tumors are equally distributed among males and females, although a slight male predominance has been reported.9,41
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
668
NEUROLOGICAL MALIGNANCIES
Most cases of CPP and CPC are sporadic and are usually not associated with other neurologic or heritable syndromes. There have been a few case reports of these tumors occurring in patients with Aicardi’s syndrome,42,43 Li–Fraumeni syndrome,44 and Von Hippel-Lindau disease.45 The occurrence of CPC in two siblings has been reported,46 and other factors such as X –chromosome-linked syndromes might be an influence.47 The etiology of some CPT has been linked to SV40 infection.48,49
PATHOLOGY Macroscopically and microscopically, these tumors resemble the choroid epithelial tissue from which they are derived. On gross examination, CPP are typically well circumscribed, globular lesions arising from the choroidal tissue to fill and/or expand the involved ventricle (see Figure 1). CPP are classified as grade I under the World Health Organization (WHO) classification and microscopically display a single layer of columnar epithelium well organized along a
Figure 2 Choroid plexus papilloma. H&E stained section.
Figure 3 Gross specimen of the brainstem demonstrating a choroid plexus carcinoma of the 4th ventricle. Note the invasion of the pons and marked tumor vascularity.
Figure 1 Gross brain specimen demonstrating a choroid plexus papilloma within the left lateral ventricle near the atrium. This patient also has associated hydrocephalus.
continuous basement membrane in a papillary pattern (see Figure 2).50 When cut in cross-section these papillae have a “cobblestone” appearance. Mitoses are rare. Necrosis may be present, but it usually is not extensive. Calcification is common in both CPP and CPC,32 but actual bone formation within these tumors is rare.51 Focal glial differentiation is common.21 CPC are WHO grade III tumors and resemble their benign counterparts but are grossly and microscopically invasive (see Figure 3). Histologically, the epithelial cells are poorly differentiated, pleomorphic, and actively mitotic (see Figure 4). Well-formed papillae are typically not seen and the papillary arrangement may give way altogether to form sheets of undifferentiated cells (see Figure 5). Necrosis may be extensive. In some cases the pathology may be difficult to distinguish from a metastatic epithelial neoplasm, especially if the tumor lacks areas of more organized papillae. Electron microscopy demonstrating a basement membrane and cilia may be helpful in differentiating difficult cases.52 Immunohistochemically, both tumors will demonstrate immunoreactivity for S-100 protein, cytokeratins, and
CHOROID PLEXUS PAPILLOMA AND CARCINOMA
669
CPP, and CPC, but does not reliably differentiate these tumors from metastatic carcinomas.53 Synaptophysin is also immunopositive in normal and neoplastic choroid plexus but negative in metastatic carcinomas with the exception of neuroendocrine carcinomas.54,55 Absent or decreased staining for transthyretin and S-100 with preserved cytokeratin immunoreactivity may indicate a poorer prognosis.56 A high Ki67/MIB-1 labelling index is also indicative of a more aggressive tumor57 with a mean of 1.9% for CPP and 13.8% for CPC.39,58
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS
Figure 4 Choroid plexus carcinoma. H&E stained section.
Figure 5 Choroid plexus carcinoma showing sheets of undifferentiated cells. H&E stained section.
vimentin.52 Glial fibrillary acidic protein (GFAP) immunopositivity is found in 25–55% of CPP and 20% of CPC and is usually restricted to a small subset of cells.39,52 Transthyretin immunoreactivity is seen in normal choroid plexus cells,
The clinical presentation will depend on the age of the patient and the location and size of the tumor. Most commonly, signs and symptoms include those of increased intracranial pressure with or without focal neurological deficit. In infants and very young children, the presentation may be quite nonspecific, with somnolence, apnea, vomiting, weight loss, ataxia, and loss of developmental milestones. A rapidly increasing head circumference may be seen in young children in whom the cranial sutures have not yet fused. Older children and adults usually present with progressive headache and vomiting with or without complaints of a focal neurologic deficit. Seizures may occur. Cerebellar signs may predominate in adults, reflecting the tumor’s propensity for the fourth ventricle in older patients. Papilledema is common. Focal neurologic signs including hemiparesis, aphasia, visual field defect, reduced vertical gaze, and sixth cranial nerve weakness may occur. A few patients have presented with signs and symptoms of acute intracranial hemorrhage.40 CT and MRI typically demonstrate a large, intraventricular, well-delineated, lobulated mass with robust, homogeneous enhancement (see Figures 6 and 7). Both CPC and CPP may have cysts or small foci of necrosis. Calcification is seen in approximately 25% of the cases. T2-weighted MRI sequences may reveal vascular flow voids. Transependymal resorption of CSF is common when hydrocephalus is present, and peritumoral edema may be seen in CPC. Ependymal or leptomeningeal spread of tumor is best appreciated by gadolinium-enhanced MRI and is more common in CPC than CPP (see Figure 8). As most patients have significantly raised intracranial pressure, a CSF examination to look for neoplastic cells is usually contraindicated prior to tumor removal. There is no radiographic feature that reliably distinguishes between CPP and CPC. Moreover, this CT and MRI appearance is not unique to these tumors. Glial-derived tumors (especially ependymomas and subependymomas), primitive neuroectodermal tumors, central neurocytomas, and metastatic carcinomas may arise or appear to arise from within the ventricular system and have similar radiographic characteristics. MR spectroscopy may be helpful in differentiating CPP from CPC.59
TREATMENT Surgery One meta-analysis and numerous small retrospective series suggest that the extent of surgical resection is the single
670
NEUROLOGICAL MALIGNANCIES
Figure 7 Axial T1-weighted MRI of the brain with gadolinium contrast in a child with a large left ventricular choroid plexus carcinoma. Note the intense contrast enhancement and central necrosis within the tumor. Figure 6 Contrasted CT of the brain in an adult patient with a 4th ventricular choroid plexus papilloma. Note as well the enlargement of the temporal horns of the lateral ventricles from associated hydrocephalus.
most important treatment variable predicting disease-free survival, regardless of age or histology.23,36,38,40,41,60 – 64 Unfortunately, a GTR may be difficult due to the often large and highly vascularized nature of CPT. In the absence of leptomeningeal metastases, GTR is often curative for CPP. Most deaths from CPP have been in the early perioperative period, and survival in CPP significantly improved in publications after 1970, likely owing to improved surgical techniques.41 Many patients require pre- or postoperative ventriculoperitoneal shunting of their hydrocephalus. Subdural hematomas and hygromas are a frequent complication as a result of brain collapse following shunting of the hydrocephalus or resection of the large mass.9,61,65 Another series of 41 patients found a 22% incidence of temporary swallowing dysfunction, which often led to placement of a percutaneous endoscopic gastrostomy tube or tracheostomy.63 For CPC, the degree of surgical resection is still the best predictor of outcome; nevertheless, CPC tends to recur even in those that are completely resected.9,29,38,40,60,61 This is due to their propensity to invade the brain and spread to transependymal or leptomeningeal locations. GTR is achieved in only 40 to 57% of patients in the literature.60
Radiotherapy Because of the rarity of these tumors, no prospective randomized trials of radiotherapy exist. The role of RT in the
Figure 8 Sagittal T1-weighted MRI of the brain with gadolinium contrast in a child with left ventricular choroid plexus carcinoma and leptomeningeal spread of tumor. Note the intense leptomeningeal enhancement.
treatment of CPP has not been determined. There appears to be no rationale for its use in patients who have had a GTR of a CPP. Even patients with subtotal resections (STR) have
CHOROID PLEXUS PAPILLOMA AND CARCINOMA
had prolonged survivals without receiving RT.23 A review of 41 patients treated at the Mayo Clinic found that only half of the patients who received an STR required a reoperation for recurrence and that RT after initial STR did not influence outcome.63 Successful treatment of CPP with stereotactic radiosurgery has been reported.13,66 The role of RT in the treatment of CPC remains somewhat controversial. RT is generally avoided in patients younger than 3 years of age due to adverse sequelae. In patients older than 3 years of age, postoperative RT is often recommended, especially in the setting of residual or recurrent disease. One report showed improved survival regardless of the degree of resection.67 Although no systematic comparison of local versus craniospinal axis radiation in CPC has been performed, craniospinal axis radiation would appear to be appropriate in patients with leptomeningeal spread of tumor. There is no report in the literature of the use of radiosurgery for CPC.
Chemotherapy There is no established role of chemotherapy in CPP. Numerous reports of giving chemotherapy for the treatment of CPC exist in the literature; however, the usefulness of this is even less certain.9,20,29,36,38,40,60,68 – 70 These patients have received a variety of chemotherapy agents in various combinations, including vincristine, vinblastine, lomustine, carboplatin, cisplatin, procarbazine, ifosfamide, cyclophosphamide, etoposide, bleomycin, thiotepa, 5-fluorouracil, and methotrexate. The number of patients in each series was small. Most of the series spanned many years and patients within the series usually received different chemotherapy regimens. No study was randomized due to the very small number of patients. It is important to note that there were a number of patients who received no adjuvant chemotherapy or RT following GTR of CPC who had many years of diseasefree survival.9,40,71 In most series, chemotherapy was given at the time of relapse or postoperatively in young children with residual tumor.9,36,40,60,70 Chemotherapy was also used postoperatively as an adjuvant therapy for patients with no radiographically identified residual tumor.38,60,69,70 Other studies have used neoadjuvant chemotherapy to reduce the tumor volume and allow for a more complete surgical resection.29,72
PROGNOSIS In a meta-analysis of 217 publications and 566 patients, the 1, 5 and 10-year projected survival rates were 90, 81, and 77% in CPP compared to 71, 41, and 35% in CPC.41 Metastases at presentation and/or age over 40 were poor prognostic indicators. As noted above, the extent of surgical resection is the best predictor of long-term survival. CPP patients with GTR had an 85% 10-year survival rate compared with 56% in those with STR. The 2-year survival in CPC was 72 and 34% for patients with GTR and STR, respectively. Krishnan et al. reported a series of 41 CPP patients treated between 1974 and 2000.63 The 5-year overall survival was 97% with no significant difference found between patients
671
who received GTR versus those who received STR. McEvoy et al. reported a large series including 25 cases of CPP.64 In the absence of surgical complications, the 5-year survival was 100%. In this series however, morbidity was significant with 39% suffering from developmental delay, 17% from severe behavioral problems and 48% with epilepsy. Berger et al. reported the largest series of CPC patients to receive chemotherapy.60 Seventeen of 22 patients in this series received chemotherapy and no RT. In patients who had had a GTR prior to chemotherapy, five of seven were alive in continuous complete remission at a median of 16 months from diagnosis, while all but one of the patients who had a partial tumor resection were dead of disease at a median of 10 months from diagnosis. The overall 5-year survival rate in CPC was 26% with a median survival of 22 months.60 In the series by Packer et al., the median progressionfree survival was only 6 months in patients receiving an STR versus 46 months in patients who received a GTR.40 Ellenbogen et al. reported a 5-year survival of 50% in their series of 15 patients with CPC dating from 1941 to 1987.9 All patients who underwent an STR were dead within 7 months of surgery. Those patients who underwent GTR of their tumors were alive an average of 9.8 years from surgery.
AUTHORS’ RECOMMENDATIONS In summary, aggressive surgical resection is the treatment of choice and best prognostic indicator for patients with CPP and CPC. Postoperative management of CPP should generally be a conservative “wait and see” approach. CPP patients with atypical features or residual disease should be followed up closely. Given the potential for long-term adverse sequelae (especially in young children), it is probably best to reserve RT for patients with symptomatic disease not amenable to surgical resection. Radiosurgery might be a consideration for symptomatic, residual disease, although very little data exists on its use in this setting. If CPP recurs, consideration should again be made for surgical resection prior to use of chemotherapy or RT. Regarding CPC, the role of adjuvant RT and chemotherapy is not yet defined and requires further investigation. In children less than 3 years of age, chemotherapy would appear to be a reasonable option for residual or recurrent disease following surgical resection. Radiation should be reserved for CPC patients over 3 years of age with residual or recurrent inoperable disease. Radiosurgery might also be an option in this setting; however, there is no data available to support a recommendation. Despite chemotherapy and/or radiotherapy, there are very few long-term survivors among those who had not also received aggressive surgical resections. These data appear to underscore the importance of extent of resection on prognosis in CPC. In addition, MRI evaluation of the total spine should be part of the routine evaluation in patients with CPC.
REFERENCES 1. GuerardM. Tumeur fongueuse dans le ventricule droit du cerveau chez une petite fille de trois ans. Bull Soc Anat 1833; 8: 211.
672
NEUROLOGICAL MALIGNANCIES
2. Rokitansky C. Handbuch der Specieleen Pathologischen Anatomie, Vol. I. Vienna, Austria: Braum¨uller and Seidel Ed., 1844. 3. LeBlanc C, Papilloma myxomatodes, Beitr. z. path. Anat. d. Gehirn Tumoren. Inaug Dissert, Bonn 1868. 4. Bielschowsky M, Unger E. Zur kenntnis der primaren epithelgeschwulste der adergeflechte des gehirns. Arch Klin Chir 1902; 81: 61 – 82. 5. Perthes GCm. Gluckliche entfernung eines tumors des plexus choriodeus an dem seitenventrikel des cerebrum. Munch Med Wochenschr 1919; 66: 677 – 8. 6. Laurence KM, Hoare RD, Till K. The diagnosis of the choroid plexus papilloma of the lateral ventricle. Brain 1961; 84: 628 – 41. 7. Van Wagenen WP. Papillomas of the choroid plexus: report of two cases, one with removal of tumor at operation and one with “seeding” of the tumor in the ventricular system. Arch Surg 1930; 20: 199 – 231. 8. Di Rocco C, Iannelli A. Poor outcome of bilateral congenital choroid plexus papillomas with extreme hydrocephalus. Eur Neurol 1997; 37(1): 33 – 7. 9. Ellenbogen RG, Winston KR, Kupsky WJ. Tumors of the choroid plexus in children. Neurosurgery 1989; 25(3): 327 – 35. 10. Gudeman SK, et al. Surgical removal of bilateral papillomas of the choroid plexus of the lateral ventricles with resolution of hydrocephalus. Case report. J Neurosurg 1979; 50(5): 677 – 81. 11. Yoshino A, et al. Multiple choroid plexus papillomas of the lateral ventricle distinct from villous hypertrophy. Case report. J Neurosurg 1998; 88(3): 581 – 5. 12. Carson BS, et al. Third ventricular choroid plexus papilloma with psychosis. Case report. J Neurosurg 1997; 87(1): 103 – 5. 13. Duke BJ, Kindt GW, Breeze RE. Pineal region choroid plexus papilloma treated with stereotactic radiosurgery: a case study. Comput Aided Surg 1997; 2(2): 135 – 8. 14. Nakano I, Kondo A, Iwasaki K. Choroid plexus papilloma in the posterior third ventricle: case report. Neurosurgery 1997; 40(6): 1279 – 82. 15. Fujimura M, et al. Hydrocephalus due to cerebrospinal fluid overproduction by bilateral choroid plexus papillomas. Childs Nerv Syst 2004; 20(7): 485 – 8. 16. Erman T, et al. Choroid plexus papilloma of bilateral lateral ventricle. Acta Neurochir (Wien) 2003; 145(2): 139 – 43; discussion 43. 17. Carter AB, et al. Choroid plexus carcinoma presenting as an intraparenchymal mass. J Neurosurg 2001; 95(6): 1040 – 4. 18. Milhorat TH, et al. Choroid plexus papilloma. I. Proof of cerebrospinal fluid overproduction. Childs Brain 1976; 2(5): 273 – 89. 19. Smith JF. Hydrocephalus associated with choroid plexus papillomas. J Neuropathol Exp Neurol 1955; 14(4): 442 – 9. 20. Boyd MC, Steinbok P. Choroid plexus tumors: problems in diagnosis and management. J Neurosurg 1987; 66(6): 800 – 5. 21. Buxton N, Punt J. Choroid plexus papilloma producing symptoms by secretion of cerebrospinal fluid. Pediatr Neurosurg 1997; 27(2): 108 – 11. 22. Eisenberg HM, McComb JG, Lorenzo AV. Cerebrospinal fluid overproduction and hydrocephalus associated with choroid plexus papilloma. J Neurosurg 1974; 40(3): 381 – 5. 23. McGirr SJ, et al. Choroid plexus papillomas: long-term follow-up results in a surgically treated series. J Neurosurg 1988; 69(6): 843 – 9. 24. Meyers SP, et al. Choroid plexus carcinomas in children: MRI features and patient outcomes. Neuroradiology 2004; 46(9): 770 – 80. 25. Higano S, et al. Supratentorial primary intra-axial tumors in children. MR and CT evaluation. Acta Radiol 1997; 38(6): 945 – 52. 26. Valladares JB, Perry RH, Kalbag RM. Malignant choroid plexus papilloma with extraneural metastasis. Case report. J Neurosurg 1980; 52(2): 251 – 5. 27. Vaquero J, et al. Primary carcinoma of the choroid plexus with metastatic dissemination within the central nervous system. Acta Neurochir (Wien) 1979; 51(1 – 2): 105 – 11. 28. Leblanc R, et al. Diffuse craniospinal seeding from a benign fourth ventricle choroid plexus papilloma. Case report. J Neurosurg 1998; 88(4): 757 – 60. 29. St Clair SK, et al. Current management of choroid plexus carcinoma in children. Pediatr Neurosurg 1991; 17(5): 225 – 33. 30. Matson DD, Crofton FD. Papilloma of the choroid plexus in childhood. J Neurosurg 1960; 17: 1002 – 27.
31. Peschgens T, et al. Therapy of choroid plexus carcinoma in childhood. Case report and review of the literature. Klin Padiatr 1995; 207(2): 52 – 8. 32. Ho DM, Wong TT, Liu HC. Choroid plexus tumors in childhood. Histopathologic study and clinico-pathological correlation. Childs Nerv Syst 1991; 7(8): 437 – 41. 33. Pollack IF. Brain tumors in children. N Engl J Med 1994; 331(22): 1500 – 7. 34. Rovit RL, Schechter MM. Choroid plexus papillomas. Observations on radiographic diagnosis. Am J Roentgenol Radium Ther Nucl Med 1970; 110(3): 608 – 17. 35. Sharma R, et al. Choroid plexus papillomas. Br J Neurosurg 1994; 8(2): 169 – 77. 36. Allen J, et al. Choroid plexus carcinoma – responses to chemotherapy alone in newly diagnosed young children. J Neurooncol 1992; 12(1): 69 – 74. 37. Romano F, et al. Prenatal diagnosis of choroid plexus papillomas of the lateral ventricle. A report of two cases. Prenat Diagn 1996; 16(6): 567 – 71. 38. Pierga JY, et al. Carcinoma of the choroid plexus: a pediatric experience. Med Pediatr Oncol 1993; 21(7): 480 – 7. 39. Rickert CH, Paulus W. Tumors of the choroid plexus. Microsc Res Tech 2001; 52(1): 104 – 11. 40. Packer RJ, et al. Choroid plexus carcinoma of childhood. Cancer 1992; 69(2): 580 – 5. 41. Wolff JE, et al. Choroid plexus tumours. Br J Cancer 2002; 87(10): 1086 – 91. 42. Trifiletti RR, et al. Aicardi syndrome with multiple tumors: a case report with literature review. Brain Dev 1995; 17(4): 283 – 5. 43. Uchiyama CM, et al. Choroid plexus papilloma and cysts in the Aicardi syndrome: case reports. Pediatr Neurosurg 1997; 27(2): 100 – 4. 44. Yuasa H, Tokito S, Tokunaga M. Primary carcinoma of the choroid plexus in Li-Fraumeni syndrome: case report. Neurosurgery 1993; 32(1): 131 – 3; discussion 3 – 4. 45. Blamires TL, Maher ER. Choroid plexus papilloma. A new presentation of von Hippel-Lindau (VHL) disease. Eye 1992; 6((Pt 1)): 90 – 2. 46. Coons S, et al. Choroid plexus carcinoma in siblings: a study by light and electron microscopy with Ki-67 immunocytochemistry. J Neuropathol Exp Neurol 1989; 48(4): 483 – 93. 47. Steichen-Gersdorf E, et al. Hypomelanosis of Ito in a girl with plexus papilloma and translocation (X;17). Hum Genet 1993; 90(6): 611 – 3. 48. Tabuchi K, Kirsch WM, Van Buskirk JJ. Immunocytochemical evidence of SV 40-related T antigen in two human brain tumours of ependymal origin. Acta Neurochir (Wien) 1978; 43(3 – 4): 239 – 49. 49. Martini F, et al. SV40 early region and large T antigen in human brain tumors, peripheral blood cells, and sperm fluids from healthy individuals. Cancer Res 1996; 56(20): 4820 – 5. 50. Burger PC, Scheithauer BW. Tumors of the Central Nervous System. Washington, District of Columbia: Armed Forces Institute of Pathology, 1994. 51. Doran SE, Blaivas M, Dauser RC. Bone formation within a choroid plexus papilloma. Pediatr Neurosurg 1995; 23(4): 216 – 8. 52. Gaudio RM, Tacconi L, Rossi ML. Pathology of choroid plexus papillomas: a review. Clin Neurol Neurosurg 1998; 100(3): 165 – 86. 53. Albrecht S, et al. Transthyretin immunoreactivity in choroid plexus neoplasms and brain metastases. Mod Pathol 1991; 4(5): 610 – 4. 54. Kepes JJ, Collins J. Choroid plexus epithelium (normal and neoplastic) expresses synaptophysin. A potentially useful aid in differentiating carcinoma of the choroid plexus from metastatic papillary carcinomas. J Neuropathol Exp Neurol 1999; 58(4): 398 – 401. 55. Kohmura E, et al. Usefulness of synaptophysin immunohistochemistry in an adult case of choroid plexus carcinoma. Neurol Res 2000; 22(5): 478 – 80. 56. Paulus W, Janisch W. Clinicopathologic correlations in epithelial choroid plexus neoplasms: a study of 52 cases. Acta Neuropathol (Berl) 1990; 80(6): 635 – 41. 57. Carlotti CG Jr, et al. Evaluation of proliferative index and cell cycle protein expression in choroid plexus tumors in children. Acta Neuropathol (Berl) 2002; 103(1): 1 – 10. 58. Vajtai I, Varga Z, Aguzzi A. MIB-1 immunoreactivity reveals different labelling in low-grade and in malignant epithelial neoplasms of the choroid plexus. Histopathology 1996; 29(2): 147 – 51.
CHOROID PLEXUS PAPILLOMA AND CARCINOMA 59. Horska A, et al. Proton magnetic resonance spectroscopy of choroid plexus tumors in children. J Magn Reson Imaging 2001; 14(1): 78 – 82. 60. Berger C, et al. Choroid plexus carcinomas in childhood: clinical features and prognostic factors. Neurosurgery 1998; 42(3): 470 – 5. 61. Pencalet P, et al. Papillomas and carcinomas of the choroid plexus in children. J Neurosurg 1998; 88(3): 521 – 8. 62. Tomlinson F, et al. Choroid plexus neoplasia: an immunochemical, flow cytometric, and proliferation marker study (abstract). Brain Pathol 1995; 4: 437. 63. Krishnan S, et al. Choroid plexus papillomas: a single institutional experience. J Neurooncol 2004; 68(1): 49 – 55. 64. McEvoy AW, et al. Management of choroid plexus tumours in children: 20 years experience at a single neurosurgical centre. Pediatr Neurosurg 2000; 32(4): 192 – 9. 65. Knierim DS. Choroid plexus tumors in infants. Pediatr Neurosurg 1990; 16(4 – 5): 276 – 80. 66. Eder HG, et al. The role of gamma knife radiosurgery in children. Childs Nerv Syst 2001; 17(6): 341 – 6; discussion 7.
673
67. Wolff JE, et al. Radiation therapy and survival in choroid plexus carcinoma. Lancet 1999; 353(9170): 2126. 68. Arico M, et al. Choroid plexus carcinoma: report of one case with favourable response to treatment. Med Pediatr Oncol 1994; 22(4): 274 – 8. 69. Duffner PK, et al., The Pediatric Oncology Group. Postoperative chemotherapy and delayed radiation in infants and very young children with choroid plexus carcinomas. Pediatr Neurosurg 1995; 22(4): 189 – 96. 70. Gianella-Borradori A, et al. Choroid plexus tumors in childhood. Response to chemotherapy, and immunophenotypic profile using a panel of monoclonal antibodies. Cancer 1992; 69(3): 809 – 16. 71. Fitzpatrick LK, Aronson LJ, Cohen KJ. Is there a requirement for adjuvant therapy for choroid plexus carcinoma that has been completely resected? J Neurooncol 2002; 57(2): 123 – 6. 72. Souweidane MM, Johnson JH Jr, Lis E. Volumetric reduction of a choroid plexus carcinoma using preoperative chemotherapy. J Neurooncol 1999; 43(2): 167 – 71.
Section 10 : Neurological Malignancies
62
Glioma and Other Neuroepithelial Neoplasms
Paul L. Moots, Mahlon D. Johnson, Mark T. Jennings and Anthony T. Cmelak
OVERVIEW Neuroepithelial neoplasms as a group are uncommon but by no means rare cancers. Primary central nervous system (CNS) neoplasms are about one-fifth as common as metastatic neoplasms involving the nervous system. Included in the broad category of primary CNS neoplasms are a variety of histologically distinct types, the majority of which are glial in origin. In recent years, the classification of neuroepithelial neoplasms has evolved considerably.1 Aside from the astrocytic neoplasms, glioblastoma, anaplastic astrocytoma, low-grade astrocytoma, and the oligodendrogliomas, the remainder of these neuroepithelial neoplasms constitute relatively rare entities. The frequency with which various neuroepithelial neoplasms are encountered changes considerably with age. This chapter will focus predominantly on adult gliomas, however, many of the rare forms occur as often or more in children. Most of the treatment plans used for the rare forms of glioma are derived from the knowledge of treatment for the more common forms, and some of the recent developments regarding the treatment of those neoplasms will be reviewed. Many of the primary neuroepithelial neoplasms share similar growth patterns, with a high propensity for local invasion and little likelihood of systemic dissemination. The clinical symptomatology, largely reflecting local cerebral dysfunction, at times combined with altered intracranial pressure, shares many similarities among the various types of CNS neoplasms. The principles of therapy, in regard to both neurological and oncological management, also share many similarities among the various gliomas. Yet, as the knowledge of glioma biology progresses, important differences in tumor biology and differences in response to therapy have been established. In current practice, the histologic distinctions remain the primary piece of information upon which prognosis and treatment planning are based. Increasingly, however, the development of treatment response and prognostic information based on molecular analysis of individual
tumors is having an impact on treatment decisions for CNS malignancies. The development of therapies specific to a given histologic type of glioma is apparent in the evolution of clinical trials over the past 25 years. The tendency to include all highgrade gliomas in the same treatment protocol was common through the 1980s, but has since been supplanted by protocols specific to particular glioma histologies as well as grade in the past 15 years. Analysis of specific genetic markers and identified mechanisms of cellular resistance to radiation and chemotherapy have added refinements in treatment planning for specific tumor types. For example, molecular predictors of high likelihood for response to chemotherapy are now widely used for oligodendrogliomas. It is likely that for glioblastoma and other high-grade gliomas in adults, and for medulloblastoma in children, molecular characterization will augment histological distinctions as the basis for certain therapies and as the basis for eligibility for clinical trials.
Epidemiology Primary malignant neoplasms of the CNS account for about 1.3% of all newly diagnosed cancers. When one subdivides this broad category into the various glial and nonglial neoplasms, each type of neoplasm would reasonably be considered uncommon. Data on the incidence of all types of CNS neoplasms has been collected in the Central Brain Tumor Registry of the United States,2 which publishes annual updates (see Table 1). The overall incidence is 14 cases per 100 000 person-years including those tumors classified as either malignant (57%) or benign (43%). High-grade gliomas as a group are the most common, with an incidence of 4.0 cases per 100 000 person-years. The most common of these, glioblastoma, has a peak incidence of 13.7 cases per 100 000 person-years in the 65–74 age group based on data from 1995 to 1999. By comparison, the peak agespecific incidence rate for glioblastoma in the 1935–1964 time period was six per 100 000 person-years.3 Thus, the incidence of glial neoplasms has been increasing, as has
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS Table 1 The incidence of primary CNS neoplasms of various histologies.
Histology Neuroepithelial Glioblastoma Anaplastic astrocytoma Astrocytoma (including variants) Oligodendroglioma Anaplastic oligodendroglioma Ependymoma Medulloblastoma Benign mixed neuroglial Other Cranial/spinal nerve sheath Meningeal Lymphoma Germ cell tumors Sellar/pituitary
Rate
a
5.9 2.6 0.5 1.6 0.3 0.1 0.2 0.2 0.1 0.3 0.9 3.0 0.4 0.1 1.2
Source: CBTRUS.2 Rates are per 100 000 person-years.
a
the incidence of certain nonglial tumors, particularly primary CNS lymphoma. The prevalence rate for primary brain tumors was about 131 patients per 100 000 in the United States in the year 2000. This amounts to approximately 359 000 persons living with primary brain tumors, of which 81 000 (23%) were classified as malignant, and 267 000 (77%) were classified as benign.2 Epidemiological studies demonstrate that neuroepithelial neoplasms are slightly more common in males than in females (males 8.0, females 5.6 per 100 000 person-years). The peak age of diagnosis is in mid- to late adult life, reflecting the high proportion of glioblastoma. However, by percentage, primary neuroepithelial neoplasms account for a much larger proportion of childhood than adult cancers, being as a group the most common solid tumor in children. The incidence rate for pediatric patients (ages 0–19) is 3.9 cases per 100 000 person-years, with 3110 new cases estimated for the year 2002 in the United States. As in adults, the incidence of CNS neoplasms, such as medulloblastoma and ependymoma, occurring predominantly in children has increased. The prevalence rate for pediatric primary brain tumors was estimated at 9.5 cases per 100 000 person-years or 26 000 patients. Eighty-three percent of the pediatric patients were classified as malignant. Epidemiological studies have also provided some information regarding the causes of primary neuroepithelial neoplasms. A small percentage of patients, perhaps 5% at most, have defined hereditary predispositions to cancer that generally include non-CNS as well as CNS malignancies (see Table 2). In addition, a few families have been described in which multiple individuals have had only neuroepithelial neoplasms, although these are rare.4 A positive family history of cancer (all types) is noted in about 20% of patients with primary CNS neoplasms. However, case-control studies do not indicate an increased risk of brain tumors in patients with a family history of non-CNS cancers. Aside from defined familial cancer syndromes such as neurofibromatosis, a family history of a brain tumor is associated with a modestly increased risk of a primary CNS neoplasm (odds ratio 2.3).5
675
Table 2 Hereditary cancer syndromes associated with primary neuroepithelial neoplasms.
Neurofibromatosis type I Neurofibromatosis type II Tuberous sclerosis Nevoid basal cell syndrome Li-Fraumeni syndrome Turcot’s syndrome Hereditary retinoblastoma Von Hippel-Lindau syndrome Multiple Endocrine Neoplasia type IIa
A few environmental factors have been associated with an increased risk of CNS neoplasms.6 These include exposure to chlorinated aliphatic hydrocarbons (e.g. carbon tetrachloride), polyvinyl chloride, and organic solvents used in synthetic rubber production. Exposure to therapeutic radiation (i.e. for tinea capitis, CNS prophylaxis in leukemia) is also associated with an increased risk. Exposure to electromagnetic fields does not appear to be related to the incidence of CNS tumors in adults. Some studies have suggested a relationship in children, although recent reports have not confirmed this. No association with smoking has been observed. No association with head trauma has been observed for neuroepithelial tumors. An association between glioma and prior CNS infection with tuberculosis or toxoplasmosis has been suggested. An association with viral infections was suggested for medulloblastoma and ependymoma on the basis of studies of children who received SV-40 contaminated polio vaccine in the 1950s. Long-term follow-up has failed to confirm this conclusion. However, analysis of neuroepithelial tumor specimens for SV-40 large T antigen has yielded positive results. Epidemiological evaluations of viral infections among glioma patients suggest a significantly less frequent history of chicken pox and shingles.7 A relationship between exposure, susceptibility, or immunologic response to these infections and the occurrence of gliomas is suggested, although the mechanism underlying this association is unknown.
Biology Our understanding of the pathogenesis of gliomas has advanced greatly in recent years. The developing CNS is composed of a highly cellular, rapidly proliferating population of pluripotential stem cells that give rise to neural and glial precursors. In the first years of life, the subependymal matrix zones that produce these precursors dwindle and nearly disappear. The adult brain is generally considered to be a nonreplicating tissue. However, this is an oversimplification. Even in adults, there remains a population of stem cells committed to glial lineage, but still able to give rise to astrocytic and oligodendroglial elements.8 Neuronal precursors are even more rare, but do exist, a fact that may relate to the relative rarity of neuronal neoplasms in adults. There is some evidence to suggest that the initial genetic changes leading to the development of a neoplastic neuroepithelial cell may occur very early in life. For example, animal models of glioma induced by fetal exposure to nitrosoureas lead to the development of gliomas in adult life. Alternatively, the presence of a small glial stem cell population
676
NEUROLOGICAL MALIGNANCIES
in adults may provide the substrate for the earliest alterations leading toward neoplastic transformation. Evolution of the populations of neuroepithelial cells at risk for neoplastic transformation from a large, rapidly dividing population of subependymal matrix neuroglial precursors in the developing brain to a very small population of glial precursors in the mature brain appears to be reflected in the age-related differences in the types of neuroepithelial neoplasms.9 As in other tissues, neoplastic transformation in neuroepithelial cells occurs as the result of an accumulation of genetic abnormalities that influence cell cycle control, motility, cytoarchitecture, and other features that contribute to the neoplastic phenotype. At least for astrocytic neoplasms, in which a spectrum of histologic grades is relatively well correlated with clinical behavior, it is postulated that the end stage in neoplastic evolution (e.g. glioblastoma) sometimes arises by evolution from low to high grade, corresponding to the sequential accumulation of multiple genetic abnormalities. In other instances, glioblastoma may arise de novo as the result of mutations in those genes specifically associated with the high-grade phenotype. The former type, or secondary glioblastoma, occurs more often in young adults, while the de novo or primary glioblastoma is more typical of older adults. Cytogenetic, molecular genetic, and biochemical analyses have shed light on the genetic events involved in the development of neuroepithelial neoplasms. Most gliomas have a diploid or near diploid karyotype, although subpopulations of hyperdiploid cells are often found. Karyotypic heterogeneity within a given tumor is common, and is thought to contribute to or reflect the development of subclones that differ in sensitivity to various therapies.10,11 Commonly observed cytogenetic abnormalities include gains in chromosome 7, complete or partial loss of chromosomes 10 and 9, loss of the Y chromosome, and double minute chromosomes.12 Loss of material on chromosome 10, observed by molecular analysis in almost all glioblastomas, is correlated with loss of a critical tumor suppressor gene, PTEN.13 A tumor suppressor gene is also suspected to reside on 9p. The presence of double minute chromosomes is commonly associated with amplification of the EGFR gene in these neoplasms. Loss of chromosome 22, which includes the neurofibromatosis type II (NF2 ) gene, is observed occasionally. Low-grade astrocytomas are generally diploid or near diploid, with the trisomy 7 and loss of the Y chromosome being the only abnormalities commonly reported. Loss of chromosome 22q has been observed in some ependymomas, suggesting a role for the NF2 tumor suppressor gene in their development. Molecular genetic studies complement and extend these karyotypic findings and disclose a number of important abnormalities common to neuroepithelial neoplasia. p53 mutations are found in one-third of astrocytomas irrespective of grade, suggesting an early role in transformation. Retinoblastoma susceptibility gene deletion or mutation is observed in up to 50% of astrocytomas and is more commonly associated with high-grade neoplasms, suggesting a role in tumor progression.14 Deletions or mutations of the tumor suppressor gene, PTEN, located on 10q, are observed
in a high proportion (60–90%) of glioblastomas but not lower grades of astrocytoma.13 Mutations in PTEN lead to increased activity in the PI3 kinase pathway, which reduces the apoptotic drive. Overexpression of the EGF receptor, usually by gene amplification, is observed in 10% of grade 3 and 40% of grade 4 astrocytomas. Gene rearrangement and truncation of the receptor protein are common. EGFR amplification often coexists with abnormalities on chromosome 10. This combination of events appears to be fundamental to the genesis of glioblastoma.15,16 The molecular events associated with the development of oligodendroglioma appear to be to some degree, distinct from astrocytic neoplasms. Molecular studies confirm that the losses of genetic material on chromosomes 1p and 19q are characteristic features. The alterations in biological function associated with these changes remain unclear. A number of reports indicate an association with LOH at these sites and a high likelihood of response to chemotherapy. LOH at these sites also predicts a longer survival.17 Abnormalities of the p53 gene are uncommon in oligodendrogliomas. As with astrocytomas, the loss of the PTEN gene on chromosome 10 in oligodendrogliomas may be related to transformation to a high grade. Studies detailing the mechanisms of resistance to chemotherapy in gliomas have provided a lot of information in recent years. Levels of certain DNA repairs enzymes, particularly the O6-methylguanine deoxyribonucleic acid methyltransferase (MGMT), correlate with the sensitivity to alkylating agents such as BCNU and temozolomide that act in part by guanine alkylation.18 The enzyme is a suicide enzyme that is inhibited remarkably by temozolomide as well as by other agents such as O6-benzylguanine. Depletion of the enzyme with these blockers is a therapeutic strategy that has received a lot of attention. It has also been demonstrated that low levels of the enzyme are an inherent property of some glioma. These low levels result from methylation of the MGMT gene promoter. Such tumors are inherently more responsive to alkylating agents and comprise a high percentage of the long-term survivors in recent trials of temozolomide for high-grade gliomas. Thus, stratification based on MGMT activity is being incorporated into current trials for high-grade gliomas. The repair of chemotherapy-related methylation of other DNA sites has also been under investigation. Base excision by the poly(ADP-ribose) polymerases is another important mechanism in gliomas. In an analogous fashion to the MGMT story, inhibitors of this repair mechanism have been developed and are in early phases of clinical testing. In addition to the EGF and the EGFR abnormalities noted previously, a large number of other peptide growth factors and their related receptors have been identified in gliomas. It is likely they contribute both to neoplastic transformation and to the subsequent growth and evolution of gliomas through autocrine and paracrine effects.19 Amplification of the EGFR has been shown to correlate with clinical behavior.20 Likewise, PDGF and PDGFR (platelet-derived growth factor receptor) species are expressed and likely influence growth characteristics. TGF-α and TGF-β are present. VEGF and FGF have also been demonstrated. Among the various effects
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS
attributed to these factors, angiogenesis is an important one in gliomas. The receptor tyrosine kinases related to these factors have become important targets for therapy. Erlotinib, an EGFR TK inhibitor, produces a 15 to 20% objective response rate in recurrent high-grade gliomas.21 Vatalinib, a VEGF TK inhibitor, produced a 45% response rate, but a large percent of patients (65%) had at least stable disease.22 These agents are now in trials combined with temozolomide and radiation for high-grade gliomas. Another fundamental attribute of neoplastic neuroepithelial cells is the ability to infiltrate into the surrounding neuropil. Remarkably, this ability to migrate is generally not accompanied by the invasive characteristics required for metastatic dissemination.23 Although the lack of lymphatic drainage of the CNS has long been postulated as the explanation for the rarity of systemic metastasis by gliomas, recent work indicates a much more complex biological control of invasiveness and metastases. The migration of neoplastic neuroepithelial cells along nerve fibers simulates very closely the migration of neuroglial precursors observed in the early developing nervous system.24 This process is controlled by, and in pathologic situations may be stimulated by, certain extracellular matrix components such as laminin.25,26 The expression of particular cellular adhesion molecules (e.g. integrins, CD44) also appear to influence glial migration.27 – 29 Conversely, the lack of certain surface adhesion molecules (e.g. CD15) may contribute to the inability to disseminate systemically.30 Integrin inhibitors such as Cilengitide are currently in clinical trials for gliomas.
Clinical Features and Natural History The symptoms and signs of primary CNS neoplasms represent a combination of localized or focal CNS dysfunction (i.e. seizures, aphasia, hemiparesis) sometimes with superimposed diffuse cerebral dysfunction resulting from mass effect and elevated intracranial pressure. The majority of patients with CNS neoplasms have a good performance status at diagnosis. A large percentage of patients with low-grade gliomas present with partial seizures without any other evidence of neurological deficit. Even among high-grade glioma patients, 60% present with a Karnofsky performance status of 70% or better.31 Since the majority of adult gliomas are frontal, temporal, or parietal in origin, focal cerebral hemispheric symptoms predominate.32 Seizures are somewhat more common as a presenting feature in low-grade (70–80%) than in high-grade gliomas (20–30%). In the latter, they represent a major, persistent cause of morbidity in about 25% of patients.33 Headaches are not as common as might be expected. In the absence of elevated intracranial pressure, only 36% of patients with supratentorial gliomas complain of headache. The headaches tend to be relatively modest and nondescript except for the recurring, persistent, or progressive nature. There is a tendency for the pain to worsen with bending over. Severe headaches occur when substantial mass effect and elevated intracranial pressure develop. The classical early morning headache ascribed to elevated intracranial pressure is seen only infrequently.34
677
Given the anatomic distribution of gliomas, personality and cognitive changes are often part of the presenting symptomatology. Apathy and blunted or labile effect are strikingly common, and a large number of people are incorrectly treated for depression for months prior to the diagnosis of a CNS neoplasm. Impaired cognition is among the most common, and arguably the most disabling symptom associated with CNS neoplasms. This may arise from localized cerebral dysfunction. Common examples of focal abnormalities resulting in cognitive impairment include apathy due to frontal lobe dysfunction, aphasia due to dominant temporal/frontal dysfunction, and neglect or denial of illness due to nondominant parietal dysfunction. Progressive dementia-like presentations may be seen with high-grade gliomas invading the corpus callosum or deep frontal and midline structures, such as the thalamus. Cognitive impairment may also occur because of elevated intracranial pressure, with either early or chronic herniation syndromes, or hydrocephalus. The rate of symptom progression, over weeks or months for high-grade gliomas, and over a year or two for most lowgrade gliomas, is sufficiently different from that seen with the most common neurodegenerative dementia, Alzheimer’s disease, that they should not be easily confused. Guidelines published by the American Academy of Neurology for the evaluation of dementia include neuroimaging to exclude structural causes such as neoplasms and hydrocephalus.35 Recognizing cognitive dysfunction and assessing the degree to which it impacts on daily life may be difficult because of the often subtle means by which these problems affect routine activities. Family members are often aware of subtle cognitive symptoms and changes in personality based on their persistence and evolution more than on the severity of these symptoms. Routine bedside testing of cognitive abilities is relatively insensitive for these types of deficits, particularly in older adults for whom minor abnormalities on mental status testing might be dismissed. Neurological deficits that alter personality and cognitive functions are particularly important to recognize and explain when educating the patient and caregivers about the illness, in treatment planning, and in the development of long-term care plans. For most adults with primary CNS neoplasms, at some point the decision making process regarding medical care will come to be the responsibility of caregivers, as the patient’s capacity to comprehend, analyze, judge, and make decisions is progressively impaired. Particularly in regard to end-of-life issues, the need for early discussion is highlighted in this patient population.36 Impairments resulting from bilateral involvement of homologous regions of the cerebral hemispheres or the disconnection of homologous regions produce substantially greater functional disability than does unilateral involvement. This caveat holds irrespective of the nature of the underlying process (i.e. direct involvement by tumor, hydrocephalus, treatment-related toxicity, or coincidental CNS pathology unrelated to the neoplasm). As with most types of CNS pathology, the rate of evolution of a CNS neoplasm has a significant impact on the severity of symptoms. A 4-cm glioblastoma with surrounding edema may produce a life-threatening brain herniation, while of 4-cm low-grade
678
NEUROLOGICAL MALIGNANCIES
astrocytoma in the same region may produce only focal seizures and otherwise be asymptomatic. The natural history of primary CNS neoplasms is often viewed from the perspective of high-grade gliomas, as these are the most common and most predictable. Particularly for high-grade astrocytic neoplasms, a fairly predictable and steady evolution of radiographic and clinical features is observed.37 The median survival for glioblastoma without treatment other than surgery and supportive care is approximately 3 months. However, this view of their natural history cannot be generalized to all types of gliomas. Many low-grade gliomas, both of astrocytic and oligodendroglial lineage, can be static clinically and radiographically for long periods. Optic gliomas, for example, may remain unchanged over 5 years or more. Low-grade oligodendrogliomas are also notable for progression-free intervals of many years without treatment. This tendency for long periods of stable disease makes interpretation of reports on small numbers of patients over short intervals (<3–5 years) very difficult. Gliomas, as a group, are neoplasms that fail because of inadequate local control. Leptomeningeal metastases are relatively uncommon, but not rare. These are infrequently observed with gliomas of all histologies and grades, although more commonly among the higher-grade neoplasms.38 Systemic metastases are extremely rare. Thus, in the absence of unexplained spinal symptoms such as radiculopathy or myelopathy, staging of the neuraxis with magnetic resonance imaging (MRI) scanning and cerebrospinal fluid (CSF) cytological evaluation is not warranted. In effect, the clinical history and neurological examination serve as disease staging evaluations. The important exceptions are the
medulloblastoma/primitive neuroectodermal tumor (PNET) and childhood type of ependymoma, which do have a higher incidence of subarachnoid dissemination.39 – 41 Rarely, neoplastic meningitis will be the presenting feature of a medulloblastoma or other neuroepithelial neoplasms.42 If local control for high-grade gliomas improves with advances in therapy, the incidence of leptomeningeal spread will likely increase. In the future, the staging evaluation and treatment design for all gliomas may come to resemble those currently used for medulloblastoma. The prognosis for patients with neuroepithelial neoplasms is strongly dependent on tumor histology (see Table 3). Within a given histological category, tumor grade is important, although the predictive value of grading varies with histological type.1,43 Tumor grading is also fundamental to treatment planning for most gliomas, particularly astrocytomas and oligodendrogliomas. In general, the most important clinical factors predictive of outcome for adult glioma patients are age and performance status (see Table 4). Treatment factors, particularly degree of surgical resection and adequacy of radiation therapy, are also important predictors of outcome for most types of glioma.44 The significance of chemotherapy on outcome varies considerably by tumor histology (see below).
Pathological Anatomy and Neuroimaging With a few exceptions, neuroepithelial neoplasms grow by infiltration along nerve fibers through the brain parenchyma. This sometimes leads to characteristic gross patterns of growth because of the involvement of fiber tracts (i.e. subinsular extension of a temporal glioma, subpial accumulations of tumor cells).45,46 Changes in the affected neuropil may
Table 3 Overall survival for various neuroepithelial neoplasms by histology.
Neoplasm Glioblastoma
MG (grade 3 or 4)a MG (grade 3)a Anaplastic astrocytoma Astrocytoma Anaplastic oligodendroglioma Oligodendroglioma Mixed oligodendroglioma/astrocytoma Ependymoma Childhood Adult Medulloblastoma Standard Risk Poor Risk
Percentage survival
Median survival (years)
3 months
2 years
5 years
10 years
0.9 0.9 0.9 1.1 0.9 2.9 1.7 2.0 7.3 4.6 9.8 10.8
95 95 90 95 – 95 98 85 98 95 100 99
5 22 18 18 – 63 29 55 84 70 92 93
2 10 – 5 8 46 10 – 60 45 73 83
<1 5 – – 0 – 7 – 40 26 49 63
Daumas-Duport, 1988 Levinb , 1990 Shapirob , 1989 Grossmanb , 1997 Shaw, 1997 Levinb , 1990 Daumas-Duport, 1988 Shapirob , 1989 Leighton et al., 1997 Shawc , 1992 Shawc , 1992 Leighton et al., 1997
1.7 >10
85 100
40 85
15 80
– 75
Lyons and Kellyd , 1991
>10 3.5
95 95
70 60
67 45
55 45
Evans et al.c,e , 1990
References
Source: From Moots PL. Pitfalls in the management of patients with malignant gliomas. Seminars in Neurology 1998, 18.2, pp 257 – 265. Reprinted by permission of Thieme. a MG, malignant glioma including multiple histologies, predominantly astrocytic neoplasms. b These studies utilize minimum performance status requirements for entry, that is, Karnofsky >50%. c Excluded patients who did not survive 1 month. d Posterior fossa only. e Data represents event-free survival; the 5-year survival rate was 65%.
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS Table 4 Influence of clinical and treatment factors on median survival of adults with malignant gliomas.
Median survival Age <45 years >65 years
References
24 months 6 months
Shapiro et al., 1989
20 months 13 months 7 months 5 months
Shapiro et al., 1989
Karnofsky status 90 – 100 70 – 80 50 – 60 30 – 40
Residual tumor following surgery (by CT scan) <1 cm2 >4 cm2
21 months 11 months
Wood et al., 1988
3.5 months 4.6 months 9.0 months 8.6 months 13.1 months 14.6 months
Walker et al., 1978
Postsurgical treatment Supportive care BCNU Radiation Radiation and BCNU Radiation and Temozolomide
Walker, 1978 Shapiro, 1989 Stupp et al., 2005
Source: From Moots PL. Pitfalls in the management of patients with malignant gliomas. Seminars in Neurology 1998, 18.2, pp 257 – 265. Reprinted by permission of Thieme.
be very modest for low-grade neoplasms. The cellularity is modestly increased, compared to normal white matter.
679
Some edema, neuronal satellitosis, subpial accumulation of tumors cells, endothelial hyperplasia, microcalcifications, cystic changes, and occasionally perivascular inflammatory infiltrates, are seen. In low-grade neoplasms, the blood–brain barrier is generally relatively well preserved. These features are apparent on computerized tomography (CT) and MRI scanning, although for low-grade gliomas MRI is considerably more sensitive (see Figure 1). Typically there is a region of increased signal on T2-weighted MRI images, indicating increased tissue fluid or loss of myelin content, with relatively indistinct margins. Some evidence of mass effect, demonstrated as displacement of adjacent structures (e.g. compression of sulci, ventricular displacement), is almost always present. This often helps to distinguish low-grade neoplasms from demyelinating and other nonneoplastic processes. Typically, contrast enhancement is absent, indicating preservation of the blood–brain barrier. If contrast enhancement is observed in an otherwise typical low-grade diffuse astrocytoma, the prognosis is worse. However, a few less common variants of low-grade astrocytoma, particularly the pilocytic astrocytoma, characteristically demonstrate contrast enhancement, and yet maintain an excellent prognosis. High-grade gliomas are much more cellular with more extensive edema and mass effect, in part related to impairment of the blood–brain barrier. Barrier disruption results from multiple factors, including inflammation within the
Figure 1 A 45-year-old right-handed male presented with episodes of confusion accompanied by guttural noises diagnosed as complex partial seizures. The precontrast T1-weighted MRI (a) demonstrates an indistinct fullness in the gyri of the anterior temporal lobe on the left. On postcontrast T1-weighted images (b) there is no contrast enhancement. T2-weighted images (c) reveal a poorly demarcated region of increased signal intensity that proved to be a low-grade astrocytoma.
680
NEUROLOGICAL MALIGNANCIES
tumor, release of vasoactive substances by tumors cells, endothelial hyperplasia, and neovascularity. These later features appear to result from the release of angiogenic factors by the glioma cells. The central core of the neoplasm is a region of high cellularity often with necrosis, surrounded by tumor with high cellularity and vascular changes infiltrating the adjacent neuropil. The brain adjacent to this densely packed core is infiltrated by tumor cells often over distances of many centimeters.41,47,48 MRI scanning of high-grade gliomas demonstrates extensive edema and mass effect, often with radiographic evidence of temporal or subfalcine herniation. The tumor core is typically well demonstrated after intravenous contrast as an outer region of intense enhancement with a poorly enhancing center. Delayed views will often reveal further enhancement in the center, indicating that the entire core has a poorly preserved blood–brain barrier, but because of necrosis the central area is less well perfused. The extent of the invasive component is best depicted on T2-weighted views. Again, there is an increased T2-signal abnormality with margins that usually appear indistinct. Correlative pathology studies reveal that the T2 abnormality often approximates but does not perfectly demonstrate the true extent of the neoplasm with microscopic infiltration of tumor cells beyond the radiographically apparent margin (see Figure 2). The MRI FLAIR
sequence provides a depiction similar to the T2 view, but is sometimes easier to interpret because the CSF appears dark in the FLAIR sequence. Thus, despite remarkable advances in the ability to image the CNS, the information available from MRI scanning has important limitations in regard to assessment of most neuroepithelial tumors. The true extent of the tumor is not clearly defined. It is approximated by the T2-weighted abnormality more closely than by the region of contrast enhancement for most gliomas.43,49 Changes in the T2weighted and FLAIR images are multifactorial (i.e. edema, loss of myelin, gliosis, neoplastic infiltration) and often do not improve with treatment. In fact, these changes may worsen as a delayed result of radiation and some chemotherapies. Furthermore, both the degree of contrast enhancement and the extent of the T2-weighted abnormality may improve with escalation of corticosteroid dosage. These limitations are important to consider in relation to treatment planning for surgery and radiation therapy where extent of tumor involvement is a critical element in decision making. The limitations of MRI imaging for assessment of tumor response also have important implications that impact both on the treatment decisions for individual patients (e.g. is radiographically stable disease while on chemotherapy in the months after radiation adequate justification for
Figure 2 A 30-year-old male presented with incoordination of the right leg and headaches. The precontrast T1-weighted MRI (a) demonstrates an area of low signal intensity deep in the left parietal lobe. The postcontrast T1-weighted image (b) demonstrates peripheral enhancement. The T2-weighted images (c) reveal edema and infiltration by tumor without a sharp demarcation throughout the left parietal lobe as well as the corpus callosum. Pathologically, this neoplasm proved to be a glioblastoma.
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS
continuing maintenance chemotherapy?), and on the design of clinical trials for gliomas, which typically assess tumor response based on the dimensions of the contrast-enhancing lesion.50,51 The MRI appearance of a CNS mass lesion can provide a strong impression that a lesion is a primary neoplasm, but it does not provide diagnostic certainty to a degree sufficient to explain the prognosis or to develop treatment plans. Some enhancing lesions prove to be low-grade gliomas. Even multiple enhancing lesions that suggest metastatic disease occasionally prove to be multifocal high-grade gliomas. The distinction between high-grade glioma, CNS lymphoma, solitary metastasis, and abscess remain challenges that have not been resolved by standard neuroimaging. The distinction between oligodendroglial and astrocytic neoplasms also remains very difficult. Positron-emission tomography (PET) scanning and magnetic resonance spectroscopy (MRS) are additional tools that can help determine whether a lesion is neoplastic, and yet many of these difficulties are not settled by PET or MRS. To date, the need for a pathologic diagnosis of a CNS neoplasm remains, and with the development of treatment strategies based on molecular characteristics, the need for a tissue diagnosis is increasing. Two situations in which treatment based on imaging commonly occur are the situation of a brain stem lesion and, secondly, the situation of an enhancing mass suspected to be a high-grade glioma in an elderly or very poor performance status patient. Data on brain stem lesions in children provide some guidance in the former situation. In children with brain stem lesions who have the typical history and imaging features characteristic of a brain stem glioma, a diagnostic biopsy is generally not recommended. This is based on pathological studies showing that such lesions are astrocytoma, generally high grade, in 95% of cases. The risk of biopsy in these children is not warranted. In adults, brain stem gliomas are uncommon, and thus there remains a stronger rational for biopsy, but the risks remain substantial (see below). In the situation of an older adult with advanced symptoms from an enhancing cerebral hemispheric mass, an invasive diagnostic procedure would often want to be avoided. Yet occasionally a process other than glioblastoma that might be more effectively treated, such as CNS lymphoma or abscess, will be missed for lack of a pathologic diagnosis. In our experience, patients who are not well enough to undergo a biopsy either receive only palliative care with steroids or occasionally an abbreviated course of radiation.
NEUROEPITHELIAL NEOPLASMS Astrocytic Neoplasms Glioblastoma
Glioblastoma, or grade 4 astrocytoma, is a highly cellular, cytologically anaplastic neoplasm with frequent mitoses, vascular proliferation, and necrosis, which is the most characteristic feature (see Figure 3). In most instances, glioblastoma is not difficult to diagnose histologically. There are a number of uncommon histological variants such as the small
681
Figure 3 Glioblastoma. Cellular anaplasia, vascular proliferation, and pseudopalisading necrosis are characteristic features. Hematoxylin and eosin; original magnification ×150.
cell predominant glioblastoma and the gliosarcoma. The latter contains transformed sarcomatous elements believed to derive from proliferating vascular elements. These histological variants have a behavior and prognosis that is similar to the typical glioblastoma. The distinction from anaplastic oligodendroglioma may be difficult (see below). The distinction from some low-grade gliomas with unusually anaplastic or pleomorphic cytological features also can present a critical diagnostic challenge (see below). The foundation for current treatment of glioblastoma derives from a series of trials performed by the Brain Tumor Study Group (BTSG) in the 1970s, 1980s, and 1990s investigated the roles of surgery, radiation, and chemotherapy for high-grade gliomas, principally glioblastoma.52 A review of practice guidelines for CNS cancers has recently been published.53 Resections that accomplish removal of all enhancing tumor or of all radiographically abnormal tissue might both be described as “gross total resections” depending on the author’s viewpoint. The term “gross total resection” is an imprecise one, and is often misleading to patients. This degree of resection can be accomplished in 40–60% of patients with high-grade gliomas. The percentage has been increasing with the advent of newer surgical navigation techniques combined with the use of functional mapping of brain adjacent to tumor by either functional MRI, or intraoperative electrophysiological mapping techniques performed in conjunction with awake craniotomy. Neuroanatomic factors are the primary limiting concern for surgical debulking. These include direct involvement of primary language cortex, involvement of deep midline structures (i.e. thalamus, basal ganglia), bilateral cerebral hemisphere involvement through the corpus callosum, and tumors arising in the brainstem or spinal cord. Multifocality, which occurs in about 5% of glioblastomas, and uncommon growth patterns such as meningeal/subarachnoid dissemination or gliomatosis cerebri occasionally provide contraindications to surgical resection.54 Despite a twofold increase in median survival attributable to surgical debulking demonstrated in BTSG data, surgery did not improve the long-term survival (i.e. 5-year survival).44 In a review of the experience at
682
NEUROLOGICAL MALIGNANCIES
MD Andersen, a resection that accomplished removal of 98% of abnormal tissue was required to achieve an increase in media survival.55 Even the most aggressive surgical resection cannot be considered treatment with curative potential. This fact derives from the widely infiltrative nature of glioblastoma, and of most gliomas. However, important exceptions do exist (e.g. pilocytic astrocytoma). The surgical mortality for resection of malignant gliomas (MG) is 3–5%. Transient worsening of neurological deficits is common. Persistent, functionally significant worsening is observed in about 20%.56 A remarkably high incidence of venous thrombosis is also observed. This appears to result from tumor-induced hypercoagulability. Treatment with anticoagulant therapy can be accomplished safely for patients with venous thrombosis, and long-term deep venous thrombosis (DVT) prophylaxis is recommended by some neuro-oncologists.57 Radiation therapy has the most significant impact among all treatment options on survival for patients with gliomas. Compared with surgery and supportive care, the addition of fractionated radiation increased median survival for highgrade glioma patients three- to fourfold in multiple BTSG studies. Although a dose –response curve demonstrating improved survival with increasing radiation dose has been observed, attempts to escalate the radiation dosage are constrained by radiation toxicities affecting normal neurons, glia, and vascular elements.58,59 Current radiation therapy recommendations typically include fractionated doses to a total of 59.4–60.0 Gy over 6 to 7 weeks.60,61 With escalation of the dose to between 65 and 75 Gy, a modest decline in survival has been observed. Radiation port design has evolved considerably in recent years. In the 1950s, treatment of the entire intracranial contents was recommended on the basis of autopsy evidence of diffuse brain infiltration by glioblastoma.62,63 By 1980, the routine use of CT scanning and correlation with postmortem analysis confirmed that 78% of glioblastoma recurrences were within 2 cm of the initial tumor.64 Evaluation of radiation portals using premortem CT scans and autopsy correlation in 15 patients with glioblastoma demonstrated that radiation portals encompassing the contrast-enhancing tumor volume and peritumoral edema with a 3-cm margin covered all histologically identifiable tumor within the specimen. In contrast, portals designed to treat the contrastenhancing tumor and edema with a 1-cm margin would have completely covered only 6 of 11 cases.65 The evolution of conformal, three-dimensional CT-based treatment planning, which minimizes the amount of normal brain irradiated and avoids ports that include homologous regions of both hemispheres, has improved the toxicity profile of radiation therapy for gliomas. Still, when treating a fairly large glioma with 2–3 cm margins, a substantial portion of brain will necessarily be included. Intensity-modulated radiation therapy (IMRT), using beams of variable fluence from numerous angles, has enabled radiation oncologists to further conform the dose in three dimensions around the target volume and off critical adjacent structures. In addition, a simultaneous boost can be delivered and allow dose intensification by contracting overall treatment time.66 It is currently unknown,
however, whether this technique can safely allow dose escalation enough to alter treatment outcome. Hyperfractionated radiation (i.e. twice-daily fractions), theoretically associated with less late toxicity to normal tissues, has been used to escalate dose to over 7000 cGy. Improved outcome has been questionable.67 Two literaturebased meta-analyses also show conflicting benefit from this technique.68,69 Other attempts to provide intensification of local radiation without additional radiation exposure to normal brain have included brachytherapy and radiosurgery.70 – 73 However, these approaches share many of the same limitations as conventional surgery relating to anatomic limitations and the inability to encompass widely infiltrating tumor cells. Radionecrosis within the high-dose region is a common phenomena; therefore, careful selection for patients with good performance status, smaller unilateral tumors in noneloquent brain regions has been a prerequisite. A minority of patients have been eligible for these procedures and has led to criticism that improvement in outcome is related more to selection factors than to actual treatment benefit.74 Preliminary reports from single institutions suggest an improved median survival for patients with glioblastoma who receive radiosurgery in addition to fractionated radiotherapy. However, by limiting eligibility to anatomically and geometrically favorable tumors, significant selection bias is introduced in these preliminary trials. Results from RTOG 93-05, a randomized trial of 203 patients comparing standard therapy with and without radiosurgery, are now available and show no significant survival advantage.75 Another approach to enhancing radiation efficacy is the use of radiation sensitizers. A pure radiation sensitizing agent, unlike temozolomide, provides no cytotoxic effect on its own. A wide variety of agents, including pure and nonpure sensitizers such as hydroxyurea, bromodioxyuridine, misonidazole, and paclitaxel, have been evaluated in phase II trials. Although some enthusiastic reports have been published, as yet no substantial improvements in time to progression or survival have been observed with pure radiation sensitizers in randomized controlled studies.76,77 Localized high-dose radiation induced by a neutron beam that results in radiation emissions from compounds localized to the tumor by virtue of blood–brain barrier disruption, tumor-specific antigen localization or other means, often termed boron-neutron capture, has been used in an attempt to escalate tumor dosage while avoiding radiation to normal tissues. This technically sophisticated technique has resulted in enthusiastic reports, but again with limited availability to patients and without controlled studies. Hyperthermia has been tested as an adjunction to radiation in a small randomized trial with promising results, but is not widely available and patient selection and technique are also important.78 The benefit of chemotherapy for glioblastoma and other high-grade gliomas was a highly debated issue for many years. Randomized trials including the series of BTSG studies evaluating nitrosourea-based chemotherapies demonstrated a four-fold increase in the number of patients surviving at 18 months for patients receiving BCNU. However, no improvement in long-term survival (i.e. >5 years) was demonstrated. A randomized trial of CCNU and vincristine
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS
as adjuvant therapy for children with malignant astrocytomas did demonstrate a statistically significant survival advantage for patients receiving chemotherapy.79 The magnitude of the difference in this population is substantial (i.e. 5-year event-free survival of 46 vs 18% for treatment without chemotherapy). For glioblastoma in adults, no additional benefit was found when single-agent BCNU was compared with the three-drug combination of procarbazine, CCNU, and vincristine (PCV).80 A meta-analysis of randomized trials of chemotherapy for MGs demonstrated a very modest benefit associated with the use of adjuvant nitrosourea-based chemotherapy.81 A recent trial utilizing the oral alkylating agent temozolomide has reinvigorated ideas about the use of cytotoxic chemotherapy for newly diagnosed glioblastoma. In a randomized trail comparing radiation with or without temozolomide for newly diagnosed glioblastoma, a substantial increase in median survival (12.1 vs 14.6 months) was obtained in the group receiving temozolomide.82 This study included patients with Karnofsky performance scale (KPS) = ≥70 and age less than 70 years. These results cannot be directly applied to all types of high-grade gliomas. The ability to predict a response to chemotherapy based on clinical data (e.g. age) and standard histologic observations (e.g. presence of oligodendroglial elements) has helped to individualize treatment planning somewhat. Efforts to establish molecular or biological markers that correlate with chemosensitivity have shed new light on some gliomas. Methylation of the MGMT gene promoter in glioblastoma is the best example of a molecular marker of chemosensitivity for glioblastoma. Other ways to assess the level of DNA alkylation repair activity such as immunohistochemistry for MGMT, or direct biochemical analysis of MGMT activity fresh tumor tissue may also provide indirect data on chemosensitivity. Whether loss of heterozygosity (LOH) on chromosomes 1 and 19 provide information about chemosensitivity in astrocytic lineage gliomas remains unsettled. In vitro assessment of chemosensitivity on freshly cultured surgical specimens has been attempted for over 30 years. In general, lack of sensitivity in vitro has correlated with lack of response, however, the converse has not been true. Furthermore, this type of analysis has proven very difficult to standardize and reproduce. Over the past five years temozolomide has largely supplanted the nitrosoureas as the first choice of chemotherapy for most gliomas. Temozolomide does not have substantially better response rates in phase II trials than nitrosoureas. However, its toxicity profile is somewhat better, particularly because of less hematologic toxicity. As a group, nitrosoureas have been reported to have response rates of 20–40% for recurrent glioma. However, most of the studies evaluating nitrosourea response predated MRI scanning, and the true response rate is probably even lower. A recent trial of BCNU for recurrent glioma showed only a 10% response rate.83 A number of other single agents and multiagent combinations have demonstrated modest response rates in high-grade gliomas. Single agents with reported response rates of 30% or greater include cisplatin, carboplatin, procarbazine, and diaziquone (AZQ).84 – 86 Other single agents
683
with some activity against gliomas include VP-16 given orally, tamoxifen in high dose (e.g. 80–120 mg m−2 d−1 ), retinoic acid, CPT-11. Multiagent regimens with activity include procarbazine/vincristine/CCNU, carboplatin/VP-16, cisplatin/BCNU, and cyclophosphamide/vincristine87 – 89 (see Lesser and Grossman90 for review). Attempts at escalating tumor exposure to chemotherapeutic drugs have taken many forms. These include doseintensification, high-dose myeloablative chemotherapy with bone marrow or stem cell rescue, intra-arterial infusion, blood–brain barrier disruption, and direct application of antineoplastic agents into the tumor bed by biodegradable polymer implants or by continuous infusion. Of these approaches, high-dose myeloablative therapies have demonstrated improved response rates, but again have not improved long-term (i.e. 5-year) survival. High-dose therapy can also be provided by local delivery. Implantable BCNU-containing wafers (Gliadel) have demonstrated a modest improvement in survival for patients with recurrent and newly diagnosed glioblastoma.91 The polymer releases BCNU directly into brain tissue and given the chemical instability of BCNU the extent to which active drug diffuses has been a subject of debate; at least over a centimeter and perhaps farther. This strategy seems best applied when surgical resection of all gross tumor can be accomplished leaving a suitable surface on which to apply the wafers with the aim being to treat microscopic residual disease. Over a period of a few weeks, very high local concentrations are achieved. There is minimal systemic distribution and thus the common hemotologic and other side effects of IV BCNU do not occur. Unfortunately, only a small percentage of patients can receive this treatment because of anatomic and surgical constraints. But the proof of principle has been established and local therapy by direct surgical application or through a catheter continues to be a very active issue. The inherent lack of chemosensitivity observed in the majority of MGs is itself an avenue of therapeutic manipulation. Blocking the repair of DNA damage caused by nitrosoureas with O6-benzylguanine has been investigated.92 Remarkably, temozolomide has MGMT blocking activity, in addition to its alkylating effects. Anaplastic Astrocytoma
Anaplastic astrocytoma, or grade 3 astrocytoma, histologically demonstrates hypercellularity, anaplasia, mitotic figures, and vascular proliferation. These tend to be less prominent than observed in glioblastoma. Unlike glioblastoma, necrosis is absent. Many examples of anaplastic astrocytoma include regions of lower cytological anaplasia and cellularity, or have infrequent mitoses. Thus, the distinction from a grade 2 astrocytoma may be difficult, and somewhat subjective. Immunohistochemical labeling of the actively proliferating cell fraction (e.g. MIB-1 labeling) is helpful in distinguishing more aggressive tumors. The treatment of anaplastic astrocytoma shares many similarities with glioblastoma both in overview and in regard to responsiveness to chemotherapy. However, as a group, these patients are younger than glioblastoma patients, and some evidence suggests that they respond more readily to
684
NEUROLOGICAL MALIGNANCIES
therapeutic interventions. A clear difference in survival is found when compared with glioblastoma (see Table 3). Principles of surgery and radiation are similar to those for glioblastoma. The use of chemotherapy has been evaluated in trials specific to this grade. One widely cited report demonstrated a statistically significant improvement in outcome for patients receiving PCV as opposed to single-agent BCNU.80 However an attempt by the European Organization for Research and Treatment of Cancer (EORTC) to reproduce this finding was negative. Current trials are looking at whether temozolomide is more active. Many neurooncologists continue to favor including chemotherapy in the initial treatment of anaplastic astrocytoma. However, support from a large randomized trial that uses radiation only as the control arm is not available. Astrocytoma
The term astrocytoma is used by pathologists to refer grade 1 or 2 astrocytomas. These are often called lowgrade astrocytoma. The older term, benign astrocytoma, is pathologically imprecise and clinically misleading. Its use should be avoided. The majority of adult low-grade astrocytomas are grade 2 astrocytomas of modest cellularity and little pleomorphism. The tumor cells often have an elongated, bipolar appearance, hence the descriptive term “fibrillary” is often used, although there are other less common histologic subtypes. Mitoses and necrosis are absent. Vascular proliferation is sometimes observed, but generally is less prominent than with highergrade astrocytomas (see Figure 4). As with their more malignant counterparts, the majority of these neoplasms infiltrate extensively. Some of the important histological subtypes tend to be more well circumscribed, and for some of these, aggressive local therapy such as resection may be curative (see below). Multiple series suggest resection of typical low-grade astrocytoma is beneficial, although no randomized trials have been performed to address this question.93 Likewise, a number of reports demonstrate that a survival benefit is obtained with fractionated radiation.94 Considerable debate exists as
to the appropriate dose, and even as to whether radiation therapy at diagnosis provides a better outcome than radiation administered at progression.95 The EORTC has shown equivalent survival and progression-free survival in 343 lowgrade glioma patients randomly assigned to receive 45 or 59.4 Gy postoperatively.96 A follow-up EORTC trial showed equivalent overall survival for patients treated at diagnosis versus those treated when subsequent progression is observed radiographically.97 The increasing use of surgical debulking magnifies this question as more patients are considered for adjuvant therapy with minimal residual disease radiographically. Analysis of these issues is further complicated by the considerable variability in the long-term sequelae of cranial radiation in adults. While severe cognitive impairment may occur, many patients suffer neuropsychological deficits that are subtle and difficult to separate from tumor-related phenomena.98 The cognitive sequelae of cranial irradiation are much more apparent in children. However, since lowgrade astrocytoma has a peak incidence in early to mid adult life and a significant percentage of patients will be 5- and 10-year survivors, the long-term sequelae are important considerations. To date, there is no defined role for chemotherapy in the initial management of low-grade astrocytoma. A randomized trial of radiotherapy versus radiotherapy plus CCNU for incompletely resected low-grade glioma by Southwest Oncology Group (SWOG) showed equivalent median survival.99 The pathological distinction from mixed oligoastrocytoma is a difficult one, and the latter appears to have a better prognosis as well as a better response to chemotherapy (see below). Current trials are comparing radiation with or without temozolomide. In the situation of a young adult with a widely infiltrative unresectable low-grade astrocytoma who would require a very large radiation port, some practitioners advocate a trial of temozolomide alone, hoping to defer radiation. The efficacy of this strategy remains unproven. Its application to patients with LOH on chromosomes 1 and 19 has a stronger rationale, but the significance of that genetic alteration for tumors that histologically appear to be astrocytomas is uncertain. Important Astrocytic Variants
Figure 4 Astrocytoma. These tumors exhibit only modest cellularity and modest nuclear atypia often with microcysts, but without significant vascular proliferation or mitotic activity. Hematoxylin and eosin; original magnification ×240.
In addition to grading, certain histologic and anatomic features distinguish subgroups of astrocytoma for which specific prognostic information or treatment recommendations can be described. These are: (i) gemistocytic astrocytoma, (ii) optic glioma, (iii) juvenile pilocytic astrocytoma, (iv) pleomorphic xanthoastrocytoma, (v) ganglioglioma, and (vi) dysembryoplastic neuroepithelial tumor. Individually, these are relatively rare tumors. Most of these have a predilection to occur in children and young adults. Aside from the gemistocytic astrocytoma, these are neoplasms that display a low potential for malignant progression. However, in all of these subtypes, progression to high-grade histologic features and aggressive clinical behavior sometimes occurs. Additionally, astrocytomas in adults will on rare occasions arise in the brainstem or spinal cord, and treatment of these neoplasms also deserves consideration.
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS
Gemistocytic Astrocytoma Gemistocytes are astrocytes with a large round eosinophilic cytoplasm and eccentric nucleus. Many astrocytomas contain a small number of these, and smaller gemistocytic-appearing cells also may be useful in identifying some oligodendrogliomas. However, a few astrocytomas are composed predominantly (>60%) of gemistocytes. These are essentially all supratentorial in location and predominantly frontal. Despite otherwise lowgrade histologic features, the median survival with treatment is approximately 2.5 years. These results are more typical of anaplastic astrocytoma than low-grade astrocytoma, and many authors favor more aggressive therapy for these patients100 (see Figure 5). A retrospective study of 48 patients with incompletely resected gemistocytic astrocytoma who received postoperative radiation reported a 5-year survival rate of only 30%. Age greater than 35 years was associated with a poorer prognosis.101 Optic Pathway Glioma Optic pathway gliomas (OPGs) are usually pilocytic astrocytomas, and most are characterized as WHO grade 1. A small number of patients have more aggressive tumors including pilomyxoid astrocytomas in young children (see below), and glioblastoma in middleaged adults. Approximately 25% of OPGs are confined to the optic nerve, while 75% arise or extend more posteriorly to involve the optic chiasm, optic tracts, or hypothalamus. Despite the low-grade histological appearance, they usually enhance prominently with gadolinium on MRI scanning. Twenty-five to thirty percent of patients with OPGs have neurofibromatosis type I (NF1). Sixty percent occur in children under 10 years of age, 20% occur between ages 10 and 20 years, 16% occur between 20 and 50 years, and 4% in patients greater than 50 years. The most common presentations are slowly evolving visual complaints, hypothalamic and/or endocrine dysfunction, and hydrocephalus. Hydrocephalus is associated with a poorer outcome, reflecting posterior extension of the tumor. OPGs are the most common CNS neoplasm occurring in association with NF1. About 8–10% of NF1 patients will be found to harbor an optic glioma, and occasionally bilateral optic nerve gliomas arise in this setting. Among
Figure 5 Gemistocytic astrocytoma. These astrocytomas are populated by gemistocytes with round eosinophilic cytoplasm and ovoid eccentric nuclei. Hematoxylin and eosin; original magnification ×240.
685
NF1 patients almost all symptomatic OPGs are diagnosed by age 6 years. The diagnosis of a new OPG or progression of known OPG after this age is unusual. All children with NF1 should be screened with annual visual examinations. Those with abnormal visual exams, whether symptomatic or not, should have MRI scanning performed. OPGs often show long progression-free intervals, measured in many years. Even when visual symptoms progress, the MRI findings are often unchanged. More remarkably, spontaneous regression has been reported in a small number of cases. Thus, many patients are observed with serial MRI scans until evidence of progression is documented. The treatment of tumors that are limited to the prechiasmatic optic nerve is surgical if visual preservation is no longer a goal. In this circumstance, complete resection is usually curative. Ten-year progression-free survival is 95%. Extension into the optic chiasm or beyond to the hypothalamus removes any chance of surgical cure. Many of the anteriorly located OPGs do not eliminate functional vision, and more conservative, function-sparing approaches are warranted. The attendant risk of progression with anatomic extension to structures that preclude curative surgery makes these management decisions very difficult. Both radiation and chemotherapy have been used as more conservative, function-sparing approaches. Radiation therapy is effective at controlling tumor progression.102,103 In a recent retrospective study of fifty patients, those with optic nerve tumors treated with radiation in the range of 42 to 54 Gy had a 72% PFS (progression-free survival) at 10 years. Those with chiasmal-hypothalamic tumors had a 68% PFS at 5 years.104 IMRT and fractionated radiosurgery are now being used frequently in an effort to minimize delayed radiation injury to adjacent structures. Five year PFS was 72% and overall survival 90% in a recent small series using fractionated stereotactic radiotherapy.105 Attempts to defer radiation in young children by initiating treatment with chemotherapy have demonstrated significant activity for carboplatin and vincristine.106 – 108 In 59 children with newly diagnosed or recurrent OPGs treated with this combination, the response rate was 58%, and 3-year PFS was 70%.109 In a smaller trial of 15 patients, the 5-year PFS was 63%.110 This provides an important advantage for very young patients by allowing deferral of radiation therapy. This approach can also be used in adults. Oral VP-16 produced a 36% response rate in a small cohort with recurrent chiasmatic-hypothalamic gliomas.111 In those rare instances of glioblastoma arising in the optic nerve of an adult, by inference from the more common cerebral hemisphere glioblastoma (GBM), combined modality therapy with radiation and temozolomide would be a reasonable recommendation. Survival in these cases is very poor. Pilocytic and Pilomyxoid Astrocytoma In addition to the optic nerve and hypothalamus, predominantly pilocytic lowgrade astrocytomas can arise in the thalamus, cerebellum, and less frequently in the cerebral hemispheres, the brain stem, and spinal cord. The majority are observed in children. These tumors are remarkable for the presence of intense
686
NEUROLOGICAL MALIGNANCIES
“PXA with malignant degeneration”. Recent studies have highlighted the presence of neuronal differentiation in tumor elements such as dysmorphic ganglion cells. They often arise superficially and adjacent cortical dysplasia is not uncommon. Thus, PXAs have become part of the growing number of “glioneuronal” tumors.115 For typical PXAs, surgical resection is the treatment of choice. In a review of 71 patients with a median age of 26 years, the PFS was 72% at 5 years and 61% at 10 years. The respective overall survival rates were 81 and 70%.116 Approximately 15% underwent progressive anaplastic evolution. Optimal treatment for those with malignant degeneration has not been established, but generally includes radiotherapy and often temozolomide (see Figure 7). Figure 6 Pilocytic astrocytoma. Cells with elongated nuclei and bipolar processes accompanied by Rosenthal fibers usually dominate these tumors. However, tumor cells with round nuclei resembling oligodendrocytes may be prominent in some areas. Hematoxylin and eosin; original magnification ×240.
contrast enhancement on MRI, a feature not seen in the more common low-grade fibrillary astrocytoma. Additionally, they are not highly infiltrative and thus are potentially curable by local therapy. In fact, total resection may be curative and even subtotal resection may be followed by long progression-free intervals. The truly localized nature of these neoplasms makes them often amenable to radiosurgery. Conventional fractionated radiation therapy (RT) will also control their growth. As with optic gliomas, chemotherapy with carboplatin and vincristine has demonstrated efficacy (see Figure 6). Pilocytic astrocytomas occur in adults, although much more rarely. As in children, they often have a good prognosis after aggressive local therapy.112 A more aggressive astrocytoma that shares histologic features with pilocytic astrocytoma has been identified in recent years and is called pilomyxoid astrocytoma. This tumor more commonly arises in young children with a mean age of less than 2 years at diagnosis, but a few cases have been reported in adults. It most commonly arises in the hypothalamic region, which is also a common location for pilocytic astrocytoma. However, local recurrence is much more frequent after resection, subarachnoid dissemination is considerably more common (14%), progression-free survival is much shorter than for pilocytic astrocytoma (26 vs 147 months), as is overall survival (60 vs 233 months).113 Pleomorphic Xanthoastrocytoma (PXA) These superficial tumors are usually encountered in the second decade of life, typically presenting with seizures or hemiparesis. Radiographically, the vast majority form a cyst with a tumor nodule. Grossly, these tumors are unusually well circumscribed, although extension into the leptomeninges is common. Histologically, they are characterized by astrocytes with bizarre nuclei with lipidized or foamy cytoplasm. Immunochemical staining for glial fibrillary acid protein demonstrates the glial nature of these cells. These features bestow an anaplastic appearance suggesting malignancy; however, most lack mitoses and necrosis.114 Occasional tumors have multifocal necrosis and numerous mitoses justifying the diagnosis
Ganglion Cell Neoplasms Gangliogliomas are rare lowgrade tumors with differentiated neurons and a variable glial cell component. They arise predominantly in the temporal and frontal lobes, but can arise elsewhere. Most are well circumscribed. Cyst formation and calcification are common. The majority of these tumors fall between the spectrum of gangliocytomas and gangliogliomas. Gangliocytomas are populated predominantly by mature neurons with a minimal glial component. Along this continuum, gangliogliomas contain a variable neuronal component that may be quite limited, and a more prominent glial component. Hence, in some areas gangliogliomas may resemble pure astrocytoma. The distinction from a purely glial tumor with included neurons can be difficult. Notably, cytologic atypia within the astrocytic component does not portend the same poor prognosis that such features predict in pure astrocytomas. Other characteristic features include a fibroblastic component with extensive reticulin deposition, perivascular lymphocytic infiltrates, and telangiectatic blood vessels. These features also contribute to the mistaken impression of a markedly anaplastic neoplasm. The appropriate treatment for gangliogliomas is complete resection. Additional therapy should be reserved
Figure 7 Pleomorphic xanthoastrocytoma. These hypercellular tumors exhibit striking nuclear anaplasia including multinucleated giant cells with lipidized cytoplasm. This histologic appearance may suggest a diagnosis of malignant glioma or malignant fibrous histiocytoma to the unwary. Hematoxylin and eosin; original magnification ×240.
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS
Figure 8 Ganglion cell tumor. Irregular collections of atypical or binucleate ganglion cells may be accompanied by a glial component with varying degrees of atypia. Hematoxylin and eosin; original magnification ×240.
for recurrent tumors, which are fortunately uncommon117,118 (see Figure 8). Dysembryoplastic Neuroepithelial Tumors (DNT) The dysembryoplastic neuroepithelial tumor (DNT) is a recently defined clinicopathologic entity that also demonstrates ganglionic differentiation. It manifests a benign course and a high propensity to be mistaken for more aggressive glial neoplasms, particularly mixed oligoastrocytomas. DNTs are encountered in young adults with long-standing partial seizures, often present since childhood. The majority arise in the temporal lobes or, less commonly, the frontal lobes. Radiographic evidence of a cranial deformity adjacent to the tumor is found in one-third of patients. DNTs involve and expand the cortex. They have a multinodular architecture, occasionally with microcyst formation. The seminal histological feature of DNTs is the “specific glioneural element” characterized by an alveolar or columnar pattern of oligodendrocytes perpendicular to the cortical surface surrounded by variable numbers of neoplastic oligodendroglia, producing a macroscopically nodular appearance. Further distinction between “simple” and “complex” DNTs is made based on the appearance of the adjacent oligodendroglial and astrocytic elements. Surrounding areas of cortical dysplasia are common. Anaplasia, endothelial proliferation, mitotic activity, and necrosis are not features of this neoplasm. However, in a small or poorly representative sample, the distinction from an oligodendroglioma or oligoastrocytoma may be impossible. The ganglionic elements are generally not cytologically bizarre, a feature that helps distinguish DNTs from gangliogliomas119,120 (see Figure 9). Brain Stem Gliomas in Adults Brain stem gliomas in adults are rare. Histologically, most are astrocytic neoplasms and more often are low grade histologically than in children. Oligodendroglioma are rarer still in this location. Adult brain stem gliomas are more often located in the dorsal midbrain (tectal glioma) or laterally in the medulla and extending into the cerebellar peduncles, in contrast to the central pontine location that is so typical of the childhood brain stem glioma. The clinical course in adults is often more indolent than
687
Figure 9 Dysembryoplastic neuroepethelial tumor. These cortically based tumors have nodular “glioneuronal elements” that cytologically may resemble an oligodendroglioma or oligoastrocytoma. Hematoxylin and eosin; original magnification ×240.
that of the childhood brain stem glioma. This fact is critical to point out when discussing this type of tumor with adult patients. Almost every source of information available to the patient about brain stem gliomas will describe the childhood version, which has an extremely poor prognosis. Educating the patient about the differences between adult and childhood brain stem gliomas is very important. A multi-institution retrospective review collecting patients over 14 years identified 48 adult patients (>16 years of age) with brain stem glioma excluding ependymoma.121,122 Median age at diagnosis was 34 years. The median survival for the entire group was 5.4 years. Factors suggesting a poor prognosis included a duration of symptoms of less than 3 months, a Karnofsky score of ≤70, and age greater than 40 years. MRI findings suggestive of necrosis were present in 20%, and also carried a poor prognosis. Many lesions were nonenhancing and appeared infiltrative (50%). Headache, gait disturbance, limb weakness, and diplopia were the most common symptoms. Pathologically childhood pontine gliomas are almost always astrocytomas and usually include high-grade elements. In comparison, informative pathology was obtained in 32 of the 48 cases and demonstrated astrocytoma in 56%, oligodendroglioma or oligoastrocytoma in 25%, and unspecified glioma in 19%. Seventeen of 32 were classified as grade 3 or 4 gliomas. This finding also carried a poor prognosis. The median survival in the high-grade group was approximately 1 year, whereas in the low-grade group was almost 10 years. One subgroup of special note is that of nonenhancing gliomas of the tectal region. These tumors are often stable without treatment for many years. They tend to present with hydrocephalus because of obstruction of the cerebral aqueduct. Ventricular shunt placement is often required, but the neoplasm can be observed with serial MRI scans. There are some proponents of surgical resection of progressive neoplasms in this region. More commonly, progressive glial tumors in this region have been treated with radiation.123
688
NEUROLOGICAL MALIGNANCIES
The combination of clinical, radiographic, and pathologic observations allows a practical scheme for separating distinct prognostic groups of brain stem glioma in adults. A substantial minority are high-grade infiltrating gliomas that are analogous to the very aggressive childhood brain stem glioma. Another substantial minority are those with low-grade infiltrative gliomas of the pons and medulla. These patients have a considerably better prognosis. The third group consists of the localized nonenhancing lesions in the tectal region. Although located in a very critical area, these are remarkably indolent and carry a very good prognosis, often without specific therapy. The primary treatment for adult brain stem gliomas as a group is radiation therapy. Even the low-grade tumors are not benign by any means, and yet long-term survival can be accomplished. The addition of chemotherapy has not been established as part of primary treatment. In the absence of a biopsy, those tumors appearing as high grade based on MRI or PET scan results should be considered to receive temodar as well as radiation therapy in a fashion similar to that done for a glioblastoma. Spinal Cord Gliomas in Adults Primary glial tumors arising in the spinal cord are rare, accounting for less than 5% of all gliomas. The incidence is about 1 per 100 000 personsyears. The most common histologies are ependymoma and astrocytoma. Oligodendrogliomas are exceedingly rare in the spinal cord. About half of the ependymomas arising in the region of the cauda equina/filum terminale display a distinctive histologic pattern, termed myxopapillary, a feature that is essentially limited to ependymomas in this location. The relative frequency of the two common histologies varies by age, with astrocytomas of the spinal cord being more common in children, while in adults ependymomas are more common. Over 90% of these gliomas are histologically low grade and thus the symptomatology is commonly that of a slowly progressive myelopathy with symptoms present for many months, and not uncommonly over a year prior to diagnosis. Sensory symptoms tend to predominate early, followed by weakness. Sixty percent have only minor motor deficit at diagnosis. Modest localized back pain is a common early symptom.124 Multiple ependymomas may occur, usually in the setting of neurofibromatosis. The symptomatology also varies on the basis of the level of involvement of the spinal cord. Astrocytomas most often arise in the cervical and upper thoracic region. Classical myelopathic features are seen. These include upper motor neuron findings in the legs, sometimes asymmetric motor and sensory features such as in the Brown-Sequard syndrome, Lhermitte’s phenomenon, and occasionally superimposed radicular or lower motor neuron findings in the arms. Ependymomas are more equally divided between the cervical and lumbar regions, although in patients older than 50 years, the thoracic region is commonly affected. Lumbar region ependymomas often arise from the filum terminale and are technically outside of the spinal cord proper (i.e. extramedullary). They tend to produce cauda equina symptoms with radicular findings in the legs, such as dermatomal
sensory loss, weakness with lower motor neuron signs (i.e. localized atrophy, fasciculation), absent reflexes and also incontinence. MRI is the best imaging method of imaging for spinal tumors. Most ependymomas will enhance with gadolinium. A few low-grade astrocytomas will also enhance, but most do not. As with intracranial astrocytomas, the high-grade spinal astrocytomas generally do enhance. Contiguous cysts or syringomyelia is seen in at least 30% of intraparenchymal spinal cord tumors. Evidence of subarachnoid seeding should be sought. It is unusually common in high-grade astrocytomas of the spinal cord, and is infrequent but not rare with ependymomas. The histologic diagnosis of intramedullary spinal cord tumors can often be suspected from the patient’s age, the location (e.g. lumbar vs other), and the MRI characteristics. Yet these features are not accurate enough to provide a certain diagnosis. Likewise, CSF studies are rarely helpful in establishing the diagnosis of a spinal cord neoplasm, but occasionally serve to address other diagnostic concerns such as inflammatory or demyelinating lesions that can mimic neoplasms. The diagnosis of a spinal cord neoplasm should be established by surgical biopsy. This is generally performed through a dorsal midline myelotomy. The astrocytomas are generally low grade. Many are infiltrative, which makes surgical resection impossible. Occasionally internal decompression/resection of the tumor and cyst decompression can be accomplished, yet the lack of an established plane around the tumor makes the likelihood of surgical injury to the spinal cord high. Pilocytic astrocytomas are sometimes encountered in the spinal cord. Among the astrocytomas, they are the least infiltrative and the most amenable to resection. Ependymomas tend to be very sharply demarcated. Even though they lack a true capsule, a plane of dissection can sometimes be established allowing for complete resection and potentially a surgical cure. Gross total resection is accomplished in 30–50% of the intramedullary spinal ependymomas and in a higher percentage of the extramedullary filum terminale ependymomas. While symptomatic improvement is sometimes observed after surgery, patients with long-standing and functionally significant neurological deficits are not likely to improve. The goal of surgery in regard to neurological function is to prevent progression of deficits. Postoperatively 10% of patients have a significant decline in functional neurological status mainly judged by ability to walk. One-third of these patients will recover to baseline. However, lesser degrees of neurological deficit such as new paresthesias, incontinence, and pain are common. Resection of tumors in the upper cervical region or cervico-medullary junction can also lead to impairment of breathing. When high risks such as these are weighed against the potential for a curative resection, the decision is a very difficult one. Both the decision making, and the procedure itself, if undertaken, should be done by a neurosurgeon with extensive experience in the resection of spinal neoplasms.125 – 127 Recommendations for treatment after surgery are based largely on retrospective series and by inference from the treatment of intracranial gliomas. Adult patients presenting
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS
with progressive symptoms from a low-grade astrocytoma or ependymoma, and for whom a substantial resection cannot be achieved should receive radiation therapy. Long-term control can be achieved in most patients, although this cannot be considered curative therapy. Patients who achieved a complete or near complete resection of a low-grade ependymoma should be followed with serial MRI scans, usually at 6-month intervals. At the time of progression, a second resection should be considered, but radiation should also be offered at this point. This strategy would also be reasonable for a low-grade pilocytic astrocytoma. Diffusely infiltrating low-grade astrocytomas and all highgrade astrocytomas (i.e. grades 3 and 4) should receive radiation. There is no role for chemotherapy as part of the initial treatment plan for adult low-grade spinal gliomas. In the exceptional situation in which deferring radiation is essential for an adult with a progressive spinal cord, astrocytoma Temozolomide or a nitrosourea might be tried initially. In children, a radiographic response will be achieved in 20–30% with various regimens that generally have included alkylating agents.128 By inference from intracranial tumors, some authors recommend chemotherapy for high-grade astrocytomas of the spine; currently temozolomide in a manner similar to that used for glioblastoma. There is very little data on chemotherapy for adult ependymomas. Multiagent, often very intensive regimens have been used in young children with intracranial ependymomas with the goal of deferring radiation, and responses are achieved in a significant minority. However, the biology of those childhood tumors and their clinical course is so different from that of adult spinal ependymoma that inferences regarding treatment are of little help. Recurrent intracranial ependymomas have generally appeared to respond more frequently to platinum-based therapy than to nitrosoureas and the combination of carboplatin and etoposide is frequently recommended. A retrospective review of 16 patients, predominantly adults, with progressive ependymomas found that the response rate achieved with platinum-based regimens exceeded that of nitrosourea-based regimens (67 vs 25%).129 Long-term outcome for adults with spinal cord gliomas varies on the basis of histology, age, duration of symptoms and neurological function at diagnosis, and tumor location. Patients with low-grade astrocytoma have a much better 5-year survival rate than those with high-grade histology (70–90 vs 30%). Those with pilocytic astrocytoma do the best among this group with a median survival of 98 months, compared with 68 months for grade 2, and 15 months for grades 2 and 4 astrocytomas. For intramedullary ependymomas, the 5- and 10-year survival rates are approximately 85 and 57%, respectively. Complete resection of a filum terminale ependymoma is associated with a 5-year progressionfree survival of nearly 100%.130,131 Young patients have longer survival than older patients. Symptom duration of greater than 6 months is a favorable indicator with a 5-year survival of 71 vs 42% for a shorter duration of symptoms. Patients with spinal card astrocytoma who demonstrated good neurological function pre- and postoperatively have a
689
better 5-year overall survival than those with poor neurological function (73 vs 22%)132 Similar results are reported for Karnofsky performance status. Upper cervical tumors tend to be the most disabling and the worst prognostically, while the extramedullary filum terminale tumors have a particularly good prognosis.
Oligodendroglial Neoplasms Since the late 1980s, the prognostic value of distinguishing low-grade and high-grade oligodendrogliomas has been reproducibly confirmed.133 Yet these neoplasms are not as predictable as their astrocytic counterparts, and within these two grades much more histologic variation is observed than among different grades of astrocytoma. Attempts to characterize them on the basis of histological and gross morphological features demonstrates the wide range of variability of these neoplasms.134,135 The typical oligodendroglioma contains a monotonous population of cells with round nuclei and cytoplasmic halos, the latter representing a retraction artefact of fixation. A delicate interlacing network of vessels is present. Calcification is common. Low-grade oligodendrogliomas may contain areas of higher cellularity, nuclei atypia, and even occasional mitoses (see Figure 10). Most oligodendrogliomas contain cells with astrocytic features, occasionally in sufficient proportions to be termed mixed oligoastrocytomas. Conversely, many predominantly astrocytic tumors contain a proportion of oligodendroglial elements. Considerable debate surrounds the criteria for the diagnosis of these mixed gliomas. Traditionally, the term oligoastrocytoma has been used to describe oligodendrogliomas, generally of low grade, that contain a substantial component of astrocytes (i.e. >25%) or exhibit regions of pure astrocytoma. However, there is growing recognition that the majority of oligodendrogliomas have constituent astrocytes or astrocyte-like cells. The discovery of glial progenitor cells that can differentiate into either oligodendrocytes or astrocytes in vitro depending on tissue culture conditions supports the concept of mixed glial tumors arising from
Figure 10 Oligodendroglioma. Classically, these tumors are populated by cells with uniform nuclei and clear cytoplasm separated by an arcuate vascular pattern. Microcalcification is also common. Hematoxylin and eosin; original magnification ×240.
690
NEUROLOGICAL MALIGNANCIES
a single pluripotent stem cell.136 However, it is also recognized that progressive anaplasia in oligodendrocytic neoplasms is accompanied by the appearance of “microgemistocytes” cells with expanded eosinophilic, GFAP-containing cytoplasm, and round eccentric nuclei that closely resemble gemistocytic astrocytes. Thus, the histological distinction between pure oligodendrogliomas and the highly variable appearing mixed oligoastrocytomas remains controversial. Recent evidence that both oligodendrogliomas and oligoastrocytomas, in contrast to pure astrocytomas, exhibit LOH at loci on chromosomes 1p and 19q further suggests that mixed oligoastrocytomas are better classified as an oligodendroglioma variant.137 Treatment response data also supports this view.138 – 140 Anaplastic oligodendroglioma may exhibit areas of classic or moderately pleomorphic oligodendroglioma with round nuclei, variable halos, and an arcuate vascular pattern. However, most of the tumor is dominated by dense cellularity, nuclear pleomorphism, numerous mitoses, vascular proliferation, and variable necrosis. Microgemistocytes are common, and may comprise a sizable portion of the neoplasm (see Figure 11). Stereotactic biopsies of these areas may lead to the incorrect impression of a gemistocytic or anaplastic astrocytoma. With progressive anaplasia, the histological appearance merges with that of glioblastoma multiforme. This resemblance underscores the fact that similar molecular mechanisms (i.e. LOH on chromosome 10q23) may result in the development of highly anaplastic features in different types of glioma, with a glioblastoma-like neoplasm being the final common endpoint. However, in the case of anaplastic oligodendroglioma, treatment response is an important point of distinction from glioblastoma (see below). Low-grade Oligodendroglioma
As with astrocytomas, therapeutic recommendations for oligodendrogliomas are strongly influenced by tumor grade, with more conservative measures generally recommended for low-grade oligodendrogliomas, given their highly variable natural history. These tumors may evolve to a higher grade
Figure 11 Anaplastic oligodendroglioma. Dedifferentiated oligodendrogliomas contain numerous anaplastic oligodendrocytes including microgemistocytes that sometimes resemble gemistocytic astrocytes. Hematoxylin and eosin; original magnification ×240.
of anaplasia, as do astrocytic neoplasms. Although the mean time to progression is in the range of 5–7 years, the percentage of patients with histological evolution is not well defined. Clinical variables including younger age and higher performance status are predictors of a better outcome, a similarity shared with astrocytic neoplasms. Therapeutic interventions have a significant impact on outcome, although the rationale for various treatment recommendations is currently based on modest, largely retrospective data. Most retrospective studies suggest that a survival benefit is obtained with surgical debulking. Patients with a gross total resection demonstrate a greater than twofold increase in 10-year survival as compared to subtotal resection (59 vs 23%). Conventional fractionated radiation therapy also appears to improve survival. Radiation doses of >50 Gy are associated with median survival about twice that of patients receiving less than 50 Gy (7.9 vs 4.5 years) (see Ref. 133). However, not all reports demonstrate a survival advantage associated with radiation therapy.141 Low-grade oligodendrogliomas do have relatively good response rates to alkylating agents. Initial reports utilizing nitrosourea-based regimens such as PCV, demonstrate response or stable disease in a high percentage (90%) of patients with low-grade oligodendrogliomas.142 These observations followed reports of high response rates (60–90%) achieved in anaplastic oligodendrogliomas with PCV (see below). Recent reports have also demonstrated responses to temozolomide, and many favor its use over the nitrosoureas because of a better toxicity profile. Effective salvage chemotherapy for recurrent oligodendroglioma has also been described, particularly with combinations that include VP-16 and cisplatin or carboplatin.143 The chemoresponsive nature of high-grade oligodendrogliomas has propelled the use of chemotherapy for lowgrade oligodendrogliomas, although the number of patients reported in the available studies is small, and the duration of follow-up remains insufficient to determine the true effect of chemotherapy on outcome. It appears that chemotherapy has a smaller risk of delayed neurological sequelae as compared to radiation, thus making it more favorable as part of initial therapy. Whether the efficacy of radiation varies with time or the results of tumor evolution is unknown. In summary, the initial treatment of low-grade oligodendrogliomas should include resection aimed at removal of radiographically visible tumor or as much as is feasible while maintaining acceptable neurological function. Some form of adjuvant therapy seems appropriate, although lacking controlled data indicating a survival advantage, there are neuro-oncologists who favor observing after surgery with serial MRI scans, and deferring other therapy until evidence of progression. Many still consider fractionated radiation in doses of 50–60 Gy standard therapy. Chemotherapy is now considered as an important alternative to radiation for initial management, although the long-term efficacy of this approach is not as well defined. The need for prospective randomized trials is evident, and yet, like low-grade astrocytomas, a trial with sufficient statistical power to demonstrate a clinically significant difference would likely require greater than 300 patients, with median follow-up duration of at least
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS
5, and perhaps 10, years. The feasibility of such a study using current therapies is low.144 Anaplastic Oligodendroglioma
Anaplastic oligodendroglioma generally presents as an enhancing cerebral hemispheric mass. The enhancement tends to be diffuse and patchy. A well-defined enhancing ring with a central core of lower signal is less commonly seen than with glioblastoma. As with other high-grade gliomas, the initial management must provide stabilization of neurological symptoms. Gross total resection should be performed whenever feasible. Radiation therapy is recommended with treatment plans that are similar to those for glioblastoma, encompassing the tumor and a 2–3 cm margin to a dose of approximately 60 Gy. Anaplastic oligodendrogliomas are highly chemosensitive neoplasms. This observation, first made in the late 1980s, has focused considerable attention, both in terms of pathologic diagnosis and therapeutic approaches, on oligodendrogliomas. Most older series report a very small percentage of oligodendrogliomas with anaplastic features (5–20%), while diagnosing the most histologically anaplastic or poorly differentiated of these neoplasms as glioblastoma. Thus, much of the older literature on glioblastoma is contaminated with such cases in small numbers. PCV has demonstrated remarkably high response rates (i.e. 70–80%) for both recurrent and newly diagnosed anaplastic oligodendrogliomas.145,146 A multigroup, randomized trial comparing four cycles of “intensive” PCV followed by radiation versus radiation alone was opened in 1994. A modest benefit in progression-free survival for the combined treatment arm was observed. However, there was no increase in overall survival, suggesting that second line therapy provided enough benefit to equalize overall survival. More intensive therapy for recurrent disease may be appropriate in this patient group.147
Ependymoma Ependymomas are relatively common in children, with only medulloblastoma and various astrocytomas being more frequent. They are rare in adults. Childhood ependymomas tend
691
to be intraventricular or periventricular masses. They are somewhat more common in the fourth ventricle than the lateral ventricles. The majority are comprised of a monotonous population of ependymal cells, often with regions of high cellularity, modest pleomorphism, perivascular pseudorosettes, and infiltrative margins. A minority are frankly anaplastic with mitotic figures, nuclear atypia, and vascular proliferation. True ependymal rosettes are occasionally seen (see Figure 12). The predictive value of the histological distinction between low-grade and anaplastic ependymoma is not as great as with other types of glioma.148 Whether histologically benign or anaplastic, the childhood ependymomas have a relatively poor outcome, with a 5-year survival of 40%. Tumors with a prominent embryonal component and ependymoblastic rosettes are classified as a form of primitive neuroectodermal tumor, ependymoblastoma, sharing biological, treatment, and outcome characteristics with medulloblastoma and other PNETs.149 The treatment of childhood ependymoma generally includes resection to the fullest extent possible followed by fractionated radiation to the tumor and a surrounding margin. The likelihood of subarachnoid dissemination is higher than with astrocytic neoplasms, but less than PNETs. These patients should undergo complete neuraxis staging including MRI scanning of the spine, and CSF cytological evaluation. However, recurrence is at the primary site in over 90%. Thus, the use of craniospinal radiation is limited to those patients with evidence of subarachnoid dissemination at diagnosis.150 – 152 The efficacy of chemotherapy for childhood ependymoma is less than that observed with medulloblastoma. In young children treated with multiagent chemotherapy (cisplatin, VP-16, vincristine, and cyclophosphamide) following surgery, the response rate (CR + PR) was 48%.153 However, cooperative group trials have not demonstrated a substantial impact by chemotherapy on survival, as has been observed medulloblastoma. Intracranial ependymomas also occur in adults. Some are infiltrative neoplasms that are biologically the equivalent of the childhood ependymoma. In the absence of subarachnoid dissemination, resection and local radiation are appropriate. A small percentage of adults will present with evidence of subarachnoid dissemination, and these should receive craniospinal radiation. In those cases, a rational for chemotherapy can also be made on the basis of the inference that dissemination indicates a more aggressive neoplasm, although the impact of chemotherapy on long-term survival in adults is unknown. Platinum-based regimens appear to have greater activity than nitrosourea-based regimens.154 As noted above, the majority of adult ependymomas are well-circumscribed masses arising in the spinal cord. The tendency for recurrence and dissemination of spinal ependymomas is so entirely different from that of childhood intracranial ependymomas that these represent distinct clinical entities.
ACKNOWLEDGMENT Figure 12 Ependymoma. Perivascular pseudorosettes are a histological hallmark of ependymomas. Hematoxylin and eosin; original magnification ×240.
The authors gratefully acknowledge Elizabeth M. Wright for editorial assistance and help in manuscript preparation.
692
NEUROLOGICAL MALIGNANCIES
REFERENCES 1. Kleihues P, et al. The WHO classification of tumors of the nervous system. J Neuropathol Exp Neurol 2002; 61: 215 – 25. 2. CBTRUS. Statistical Report: Primary Brain Tumors in the United States, 1995 – 1999. Central Brain Tumor Registry of the United States, 2002. 3. Schoenberg BS, Christine BW, Whisnant JP. The descriptive epidemiology of primary intracranial neoplasms: the connecticut experience. Am J Epidemiol 1976; 104: 499 – 510. 4. Moots PL, Rubinstein LJ. Multiple neoplasms of the nervous system. In Stoll BA (ed) Risk Factors and Multiple Cancer. New York: John Wiley and Sons, 1984: 331 – 51. 5. Wrensch M, et al. Familial and personal medical history of cancer and nervous system conditions among adults with glioma and controls. Am J Epidemiol 1997; 145: 581 – 93. 6. Bohnen NI, et al. Descriptive and analytic epidemiology of brain tumors. In Black PM, Loeffler JS (eds) Cancer of the Nervous System. Cambridge, Massachusetts: Blackwell Science, 1997: 3 – 24. 7. Wrensch M, et al. Does prior infection with Varicella – Zoster virus influence risk of adult glioma? Am J Epidemiol 1997; 145: 594 – 7. 8. Linskey ME, Gilbert MR. Glial differentiation: a review with implications for new directions in neuro-oncology. Neurosurgery 1995; 36: 1 – 21. 9. Rubinstein LJ. The correlation of neoplastic vulnerability with central neuroepithelial cytogeny and glioma differentiation. J Neurooncol 1987; 5: 11 – 27. 10. Shapiro JR, Yung WKA, Shapiro WR. Isolation, karyotype, and clonal growth of heterogenous subpopulations of human malignant gliomas. Cancer Res 1981; 41: 2349 – 59. 11. Yung WKA, Shapiro JR, Shapiro WR. Heterogenous chemosensitivities in subpopulations of human glioma cells in culture. Cancer Res 1982; 42: 992 – 8. 12. Bigner SH. Cytogenetic analysis of nervous system tumors. In Black PM, Loeffler JS (eds) Cancer of the Nervous System. Cambridge, Massachusetts: Blackwell Science, 1997: 790 – 801. 13. Li J, et al. PTEN, a putative tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science 1997; 275: 1943 – 7. 14. Henson JW, et al. The retinoblastoma susceptibility (Rb) gene is involved in the malignant progression of astrocytomas. Ann Neurol 1994; 36: 714 – 21. 15. Kirsch M, Zhu J, Cavenee W. Pathogenetic mechanisms of nervous system tumors. In Black PM, Loeffler JS (eds) Cancer of the Nervous System. Cambridge, Massachusetts: Blackwell Science, 1997: 703 – 43. 16. Ohgaki H. Genetic pathways to glioblastoma. Neuropathology 2005; 25: 1 – 7. 17. Cairncross JG, et al. Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendroglioma. J Natl Cancer Inst 1998; 90: 1473 – 9. 18. Hegi ME, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005; 352: 997 – 1003. 19. Jennings MT, et al. Transforming growth factor beta as a potential tumor progression factor among hyperdiploid glioblastoma cultures: evidence for the role of platelet-derived growth factor. J Neurooncol 1997; 31: 233 – 54. 20. Schlegel J, et al. Amplification of the epidermal growth factor receptor gene correlates with different growth behavior in human glioblastoma. Int J Cancer 1994; 56: 72 – 7. 21. Vogelbaum MA, et al. Phase II study of single agent therapy with the EGFR tyrosine kinase inhibitor erlotinib in recurrent glioblastoma multiforme. Ann Oncol 2004; 15(suppl 3): 206. 22. Conrad C, et al. A phase I/II trial of single agent PTK 787/ZK 222584, a novel oral angiogenesis inhibitor in patients with recurrent glioblastoma multiforme. Proc Am Soc Clin Oncol 2004; 23: 110. 23. Pedersen PH, et al. Leptomeningeal tissue: a barrier against brain tumor cell invasion. J Natl Cancer Inst 1994; 86(2): 1593 – 9. 24. Pedersen P-H, et al. Migratory pattern of fetal rat brain cells and human glioma cells in the adult rat brain. Cancer Res 1993; 53: 5158 – 65.
25. Friedlander DR, et al. Migration of brain tumor cells on extracellular matrix proteins in vitro correlates with tumor type and grade and involves alpha-v and beta – 1 integrins. Cancer Res 1996; 56: 1939 – 47. 26. Tysnes BE, et al. Stimulation of glioma-cell migration by laminin and inhibition by anti-alpha3 and anti-beta 1 integrin antibodies. Int J Cancer 1996; 67: 777 – 88. 27. Radotra B, McCormick D, Crockard A. CD44 plays a role in adhesive interactions between glioma cells and extracellular matrix components. Neuropathol Appl Neurobiol 1994; 20: 399 – 405. 28. Tamaki M, et al. Cell adhesion molecules acting between C6 glioma and endothelial cells. J Neurooncol 1995; 24: 181 – 8. 29. Paulus W, et al. Diffuse brain invasion of glioma cells requires beta-1 integrins. Lab Invest 1996; 75: 819 – 26. 30. Martin K, et al. Nonexpression of CD15 by neoplastic glia: a barrier to metastasis? Anticancer Res 1995; 15: 1159 – 66. 31. Shapiro WR et al., Brain Tumor Cooperative Group Trial 8001. Randomized trial of three chemotherapy regimens and two radiotherapy regimens in the post-operative treatment of malignant glioma. J Neurosurg 1989; 71: 1 – 9. 32. Walker MD, et al. Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic gliomas. J Neurosurg 1978; 49: 333 – 43. 33. Moots PL, et al. The course of seizure disorders in adults with malignant gliomas. Arch Neurol 1995; 52: 717 – 24. 34. Forsyth PA, Posner JB. Headaches in patients with brain tumors. A study of 111 patients. Neurology 1993; 43: 1678 – 83. 35. Knopman DS, et al. Practice Parameter: diagnosis of dementia (an evidence-based review). Neurology 2001; 56: 1143 – 53. 36. Moots PL. Pitfalls in the management of patients with malignant gliomas. Semin Neurol 1998; 18: 257 – 65. 37. Curran WJ, et al. Recursive partitioning analysis of prognostic factors in three Radiation Therapy Oncology Group trials. J Natl Cancer Inst 1993; 85: 704 – 10. 38. Moots PL, Jennings MT. Metastases of primary CNS tumors. In Maciunas RJ (ed) Advanced Techniques in Central Nervous System Metastases. American Association of Neurological Surgeons, 1998: 17 – 34. 39. Allen JC, Epstein F. Medulloblastoma and other primary malignant neuroectodermal tumors of the CNS: the effect of patients’ age and extent of disease on prognosis. J Neurosurg 1982; 57: 446 – 51. 40. Packer RJ, et al. Leptomeningeal dissemination of primary central nervous system tumors of childhood. Ann Neurol 1975; 18: 217 – 21. 41. Salazar OM. A better understanding of CNS seeding and a brighter outlook for postoperatively irradiated patients with ependymomas. Int J Radiat Oncol Biol Phys 1983; 9: 1231 – 4. 42. Jennings MT, et al. Neoplastic meningitis as the sole presentation of an occult CNS primitive neuroectodermal tumor. J Child Neurol 1993; 8: 306 – 12. 43. Daumas-Duport C, et al. Grading of astrocytomas: a simple and reproducible method. Cancer 1988; 62: 2152 – 65. 44. Wood JR, et al. The prognostic importance of tumor size in malignant gliomas: a computer tomographic scan study by the Brain Tumor Cooperative Group. J Clin Oncol 1988; 6: 338 – 43. 45. Scherer HJ. The forms of glioma growth and their practical significance. Brain 1994; 63: 1 – 35. 46. Daumas-Duport C, Scheithauer BW, Kelly PJ. A histological and cytological method for the spatial definition of gliomas. Mayo Clin Proc 1987; 62: 435 – 49. 47. Burger PC, et al. Computerized tomographic and pathologic studies of the untreated, quiescent, and recurrent glioblastoma multiforme. J Neurosurg 1983; 58: 159 – 69. 48. Burger PC, et al. Topographic anatomy and CT correlations in the untreated glioblastoma multiforme. J Neurosurg 1988; 68: 698 – 704. 49. Kelly PJ, et al. Stereotactic histologic correlations of computed tomography- and magnetic resonance imaging-defined abnormalities in patients with glial neo-plasms. Mayo Clin Proc 1987; 62: 450 – 9. 50. Friedman HS, et al. Criteria for termination of phase II chemotherapy for patients with progressive or recurrent brain tumor. Neurology 1989; 39: 62 – 6. 51. Macdonald DK, et al. Response criteria for phase II studies of supratentorial malignant glioma. J Clin Oncol 1990; 8: 1277 – 80.
GLIOMA AND OTHER NEUROEPITHELIAL NEOPLASMS 52. Shapiro WR. Therapy of adult malignant brain tumors: what have the clinical trials taught us? Semin Oncol 1986; 13: 38 – 45. 53. Brem SS, et al. Central nervous system cancers: clinical practice guidelines in Oncology. J Natl Compr Canc Netw 2005; 3: 644 – 90. 54. Barnard RO, Geddes JF. The incidence of multifocal cerebral gliomas. Cancer 1987; 60: 1519 – 31. 55. Lacroix M, et al. A multivariate analysis for 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001; 95: 190 – 8. 56. Fadul C, et al. Morbidity and mortality of craniotomies for excision of supratentorial malignant gliomas. Neurology 1988; 38: 1374 – 9. 57. Norris LK, Grossman SA. Treatment of thromboembolic complications in patients with brain tumors. J Neurooncol 1994; 22: 127 – 37. 58. DeAngelis LM, Delattre J-Y, Posner JB. Radiation-induced dementia in patients cured of brain metastases. Neurology 1989; 39: 789 – 96. 59. Jennings MT. Neurological complications of radiotherapy. In Wiley RG (ed) Neurological Complications of Cancer. New York: MarcelDekker, 1995: 219 – 40. 60. Leibel SA, Scott CB, Loeffler JS. Contemporary approaches to the treatment of malignant gliomas with radiation therapy. Semin Oncol 1994; 21: 198 – 219. 61. Shrieve DC, Loeffler JS. Advances in radiation therapy for brain tumors. Neurol Clin 1995; 13: 773 – 93. 62. Bull JND, Rovit RL. The radiographic localization of intracerebral gliomata. J Fac Radiol (Lond) 1957; 8: 147 – 57. 63. Matsukado Y, MacCarthy CS, Kernohan JW. The growth of glioblastoma multiforme (astrocytomas, grade 3 and 4) in neurosurgical practice. J Neurosurg 1961; 18: 636. 64. Hochberg FH, Pruitt A. Assumptions in the radiotherapy of glioblastoma. Neurology 1980; 30: 907 – 11. 65. Halperin EC, Burger PC, Bullard DE. The fallacy of the localized supratentorial malignant glioma. Int J Radiat Oncol Biol Phys 1988; 15: 505 – 9. 66. Cahn MF, et al. Comparison of intensity-modulated radiotherapy with three-dimensional conformal radiation therapy planning for glioblastoma multiforme. Med Dosim 2003; 28: 261 – 5. 67. Shin KH, Muller PJ, Geggie PH. Superfractionation radiation therapy in the treatment of malignant astrocytoma. Cancer 1983; 52: 2040 – 3. 68. Nieder C, et al. Radiotherapy for high-grade gliomas: Does altered fractionation improve outcome? Strahlenther Onkol 2004; 180: 401 – 7. 69. Stuschke M, Thames HD. Hyperfractionated radiotherapy of human tumors: overview of the randomised clinical trials. Int J Radiat Oncol Biol Phys 1997; 37: 259 – 67. 70. Loeffler JS, et al. Radiosurgery as part of the initial management of patients with malignant gliomas. J Clin Oncol 1992; 10: 1379 – 85. 71. Chamberlain MC, et al. Stereotactic radiosurgery for recurrent gliomas. Cancer 1994; 74: 1342 – 7. 72. Selker RG, et al. The Brain Tumor Cooperative Group NIH Trial 8701: a randomised comparison of surgery, external radiation therapy, and carmustine versus surgery, interstitial radiotherapy boost, external radiation therapy, and carmustine. Neurosurgery 2002; 51: 343 – 55; discussion 355 – 7. 73. Laperriere NJ, et al. Randomized study of brachytherapy in the initial management of patients with malignant astrocytoma. Int J Radiat Oncol Biol Phys 1998; 41: 1005 – 11. 74. Lustig RA, Scott CB, Curran WJ. Does stereotactic eligibility for the treatment of glioblastoma cause selection bias in randomised studies? Am J Clin Oncol 2004; 27: 516 – 21. 75. Souhami L, et al. Randomized comparison of stereotactic radiosurgery followed by conventional radiotherapy with carmustine to conventional radiotherapy with carmustine for patients with glioblastoma multiforme: report of Radiation Therapy Oncology Group 93-05 protocol. Int J Radiat Oncol Biol Phys 2004; 60: 853 – 60. 76. Prados MD, et al. Phase III trial of accelerated hyperfractionation with or without difluromethylornithine (DFMO) versus standard fractionated radiotherapy with or without DFMO for newly diagnosed patients with glioblastoma multiforme. Int J Radiat Oncol Biol Phys 2001; 49: 71 – 7. 77. Prados MD, et al. Radiation therapy and hydroxyurea followed but eh combination of 6-thioguanine and BNCU for the treatment of primary malignant brain tumors. Int J Radiat Oncol Biol Phys 1998; 40: 57 – 63.
693
78. Sneed PK, et al. Survival benefit of hyperthermia in a prospective randomised trial of brachytherapy boost +/− hyperthermia for glioblastoma multiforme. Int J Radiat Oncol Biol Phys 1998; 40: 287 – 95. 79. Sposto R, et al. The effectiveness of chemotherapy for treatment of high grade astrocytoma in children: results of a randomized trial. A report from the CCSG. J Neurooncol 1989; 7: 165 – 77. 80. Levin VA, et al. Superiority of post-radiotherapy adjuvant chemotherapy with CCNU, procarbazine, and vincristine (PCV) over BCNU for anaplastic gliomas: NCOC; 6G61 final report. Int J Radiat Oncol Biol Phys 1990; 18: 321 – 4. 81. Fine HA, et al. Meta-analysis of radiation therapy with and without chemotherapy for malignant gliomas in adults. Cancer 1993; 71: 2585 – 97. 82. Stupp R, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352: 987 – 96. 83. Brandes AA, et al. How effective is BCNU in recurrent glioblastoma in the modern era? A phase II trial. Neurology 2004; 63: 1281 – 4. 84. Schold SC, et al. Randomized comparison of diaziquone and carmustine in the treatment of adults with anaplastic glioma. J Clin Oncol 1993; 11: 77 – 83. 85. Newton HE, et al. Comparison between BCNU and procarbazine chemotherapy for treatment of gliomas. J Neurooncol 1993; 15: 257 – 63. 86. Warnick RE, et al. A phase II study of intravenous carboplatin for the treatment of recurrent gliomas. J Neurooncol 1994; 19: 69 – 74. 87. Longee DC, et al. Treatment of patients with recurrent gliomas with cyclophosphamide and vincristine. J Neurosurg 1990; 72: 583 – 8. 88. Jeremic B, et al. Carboplatin and etoposide chemotherapy regimen for recurrent malignant glioma: a phase II trial. J Clin Oncol 1992; 10: 1074 – 7. 89. Grossman SA, et al. Phase II study of continuous infusion carmustine and cisplatin followed by cranial irradiation in adults with newly diagnosed high-grade astrocytoma. J Clin Oncol 1997; 15: 2596 – 603. 90. Lesser GJ, Grossman SA. Chemotherapy of high grade astrocytomas. Semin Oncol 1994; 21: 220 – 35. 91. Olivi A, Brem H. Interstitial chemotherapy with sustained-release polymer systems for the treatment of malignant gliomas. Recent Results Cancer Res 1994; 135: 149. 92. Friedman HS, et al. Enhancement of nitrosourea activity in medulloblastoma and glioblastoma. J Natl Cancer Inst 1992; 84: 1926 – 31. 93. Berger MS, et al. The effect of extent of resection on recurrence in patients with low grade cerebral hemisphere gliomas. Cancer 1994; 74: 1784 – 91. 94. Leighton C, et al. Supratentorial low-grade glioma in adults: an analysis of prognostic factors and timing of radiation. J Clin Oncol 1997; 15: 1294 – 301. 95. Recht LD, Lew R, Smith TW. Suspected low-grade glioma: is deferring treatment safe? Ann Neurol 1992; 31: 431 – 6. 96. Karim ABMF, et al. A randomized trial of dose – response in radiation therapy of low-grade cerebral glioma: European Organization for Research and Treatment of Cancer (EORTC 22844). Int J Radiat Oncol Biol Phys 1996; 36: 549 – 56. 97. Van den Bent MJ, et al. Long-term efficacy of early versus delayed radiotherapy for low-grade astrocytoma and oligodendroglioma in adults: the EORTC 22845 randomized trial. Lancet 2005; 366: 985 – 90. 98. Taphoorn MJB, et al. Cognitive functions and quality of life in patients with low-grade gliomas: the impact of radiotherapy. Ann Neurol 1994; 36: 48 – 54. 99. Eyre HJ, et al. A randomized trial of radiotherapy versus radiotherapy plus CCNU for incompletely resected low grade glioma: a Southwest Oncology Group study. J Neurosurg 1993; 78: 909 – 14. 100. Krouwer HGJ, et al. Gemistocytic astrocytomas: a reappraisal. J Neurosurg 1991; 74: 399 – 406. 101. Nowak-Sadzikowska J, et al. Postoperative irradiation of incompletely excised gemistocytic astrocytomas. Clinic outcome and prognostic factors. Strahlenther Onkol 2005; 181: 246 – 50. 102. Wong JYC, et al. Optic gliomas. A reanalysis of the University of California, San Francisco experience. Cancer 1987; 60: 1847 – 55.
694
NEUROLOGICAL MALIGNANCIES
103. Flickinger JC, Torres C, Deutsch M. Management of low-grade gliomas of the optic nerve and chiasm. Cancer 1988; 61: 635 – 42. 104. Khafaga Y, et al. Optics gliomas: a retrospective analysis of 50 cases. Int J Radiat Oncol Biol Phys 2003; 56: 807 – 12. 105. Combs SE, et al. Fractionated stereotactic radiotherapy of optic pathway gliomas: tolerance and long-term outcome. Int J Radiat Oncol Biol Phys 2005; 62: 814 – 9. 106. Packer RJ, et al. Treatment of chiasmatic/hypothalamic gliomas of childhood with chemotherapy: an update. Ann Neurol 1988; 23: 79 – 85. 107. Packer RJ, et al. Carboplatin and vincristine for progressive low grade gliomas of childhood. J Clin Oncol 1992; 11: 850 – 74. 108. Listernick R, et al. Optic pathway gliomas in children with neurofibromatosis 1: Consensus statement from the NF1 Optic pathway glioma task force. Ann Neurol 1997; 41: 143 – 9. 109. Friedman HS, et al. Treatment of children with progressive or recurrent brain tumors with carboplatin or iproplatin: A Pediatric Oncology Group randomised phase II trial. J Clin Oncol 1992; 10: 249 – 56. 110. Silva MM, et al. Optic pathway gliomas in children under three years of age: the role of chemotherapy. Pediatr Neurosurg 2000; 33: 151 – 8. 111. Chamberlain MC, Grafe MR. Recurrent chiasmatic-hypothalamic glioma treated with oral etoposide. J Clin Oncol 1995; 13: 2072 – 6. 112. Bell D, et al. Pilocytic astrocytoma of the adult – clinical features, radiological features and management. Br J Neurosurg 2004; 18: 613 – 6. 113. Komotar RJ, et al. Pilocytic and pilomyxoid hypothalamic/chiasmatic astrocytomas. Neurosurg 2004; 54: 72 – 80. 114. Kepes JJ, Rubinstein LJ, Eng LF. Pleomorphic xanthoastrocytoma: a distinctive meningocerebral glioma of young subjects with relatively favorable prognosis. A study of 12 cases. Cancer 1979; 44: 323 – 8. 115. Im S-H, et al. Pleomorphic xanthoastrocytoma: a developmental glioneuronal tumor with prominent glioproliferative changes. J Neurooncol 2004; 66: 17 – 27. 116. Giannini C, et al. Pleomorphic xanthoastrocytoma: what do we really know about it? Cancer 1999; 85: 2033 – 45. 117. Hakim R, et al. Gangliogliomas in adults. Cancer 1997; 79: 127 – 31. 118. Hirose T, et al. Ganglioglioma. Cancer 1997; 79: 989 – 1003. 119. Daumas-Duport C, et al. Dysembryoplastic neuroepithelial tumor: a surgically curable tumor of young patients with intractable partial seizures. Neurosurgery 1988; 23: 545 – 56. 120. Daumas-Duport C. Dysembryoplastic neuroepithelial tumor. Brain Pathol 1993; 3: 283 – 95. 121. Guillamo J-S, Doz F, Delattre J-Y. Brain stem gliomas. Curr Opin Neurol 2001; 14: 711 – 5. 122. Guillamo J-S, et al. Brainstem gliomas in adults: prognostic factors and classification. Brain 2001; 124: 2528 – 39. 123. Yeh DD, Warnick RE, Ernst RJ. Management strategies for adult patients with dorsal midbrain gliomas. Neurosurg 2002; 50: 735 – 40. 124. Shrivastava RK, et al. Intramedullary spinal cord tumors in patients older than 50 years of age: management and outcome analysis. J Neurosurg Spine 2005; 2: 249 – 55. 125. Jallo GI, Kothbauer KF, Epstein FJ. Intrinsic spinal cord tumor resection. Neurosurg 2001; 49: 1124 – 8. 126. Parsa AT, et al. Spinal cord and intradural-extraparenchymal spinal tumors: current best care practices and strategies. J Neurooncol 2004; 69: 291 – 318. 127. Sandacioglu IE, et al. Functional outcome after surgical treatment of intramedullary spinal cord tumors: experience with 78 patients. Spinal Cord 2005; 43: 34 – 41. 128. Balmaceda C. Chemotherapy for intramedullary spinal cord tumors. J Neurooncol 2000; 47: 293 – 307. 129. Gornet MK, et al. Chemotherapy for advanced CNS ependymoma. J Neurooncol 1999; 45: 61 – 7. 130. Henson JW. Spinal cord gliomas. Curr Opin Neurol 2001; 14: 679 – 82.
131. Robinson CG, et al. Long-term survival and functional status of patients with low grade astrocytoma of the spinal cord. Int J Radiat Oncol Biol Phys 2005; 63: 91 – 100. 132. Lee HK, et al. The prognostic value of neurological function in astrocytic spinal cord glioma. Neurooncol 2003; 5: 208 – 13. 133. Shaw EG, et al. Oligodendrogliomas: the Mayo Clinic experience. J Neurosurg 1992; 76: 428 – 34. 134. Daumas-Duport C, et al. Oligodendrogliomas Part 1: patterns of growth, histological diagnosis, clinical and imaging correlations: a study of 153 cases. J Neurooncol 1997; 34: 37 – 59. 135. Daumas-Duport C, et al. Oligodendrogliomas Part II: A new grading system based on morphological and imaging criteria. J Neurooncol 1997; 34: 61 – 78. 136. Lillien LE, Raff CM. Differentiation signals in the CNS: type 2 astrocyte development in vitro as a model system. Neuron 1990; 5: 111 – 19. 137. Kraus JA, et al. Shared allelic losses on chromosomes 1p and 19q suggest a common origin of oligodendroglioma and oligoastrocytoma. J Neuropathol Exp Neurol 1995; 54: 91 – 5. 138. Glass J, et al. The treatment of oligodendrogliomas and mixed oligodendroglioma-astrocytomas with PCV chemotherapy. J Neurosurg 1992; 76: 741 – 5. 139. Kyritsis AP, et al. The treatment of anaplastic oligodendrogliomas and mixed gliomas. Neurosurgery 1993; 32: 365 – 71. 140. Kim L, et al. Procarbazine, lomustine, and vincristine (PCV) chemotherapy for grade III and grade IV oligoastrocytomas. J Neurosurg 1996; 85: 602 – 7. 141. Bullard DE, et al. Oligodendroglioma: an analysis of the value of radiation therapy. Cancer 1987; 60: 2179 – 88. 142. Mason WP, Krol GS, DeAngelis LM. Low-grade oligodendroglioma responds to chemotherapy. Neurology 1996; 46: 203 – 7. 143. Peterson K, et al. Salvage chemotherapy for oligodendrogliomas. J Neurosurg 1996; 85: 597 – 601. 144. Levin VA. Controversies in the management of low-grade astrocytomas and oligodendrogliomas. Curr Opin Oncol 1996; 8: 175 – 7. 145. Cairncross JG, MacDonald DR, Ramsay DA. Aggressive oligodendroglioma: a chemosensitive tumor. Neurosurgery 1992; 31: 78 – 82. 146. Cairncross JG, et al. Chemotherapy for anaplastic oligodendroglioma. J Clin Oncol 1994; 12: 2013 – 21. 147. Cairncross JG, et al. An intergroup randomised controlled clinical trial or chemotherapy plus radiation versus RT alone for pure and mixed anaplastic oligodendrogliomas: Initial report of RTOG 94-02. ASCO Proc 2004; 23: 107. 148. Ross GW, Rubinstein LJ. Lack of histopathological correlation of malignant ependymomas with postoperative survival. J Neurosurg 1989; 70: 31 – 6. 149. Mork SJ, Rubinstein LJ. Ependymoblastoma. A reappraisal of a rare embryonal tumor. Cancer 1985; 55: 1536 – 42. 150. Wallner KE, et al. Intracranial ependymomas: results of treatment with partial or whole brain irradiation without spinal irradiation. Int J Radiat Oncol Biol Phys 1986; 12: 1937 – 41. 151. Shaw EG, et al. Postoperative radiotherapy of intracranial ependymoma in pediatric and adult patients. Int J Radiat Oncol Biol Phys 1987; 13: 1457 – 62. 152. Lyons MK, Kelly PJ. Posterior fossa ependymomas: report of 30 cases and review of the literature. Neurosurgery 1991; 28: 659 – 65. 153. Duffner PK, et al. Postoperative chemotherapy and delayed radiation in children less than three years of age with malignant brain tumors. N Engl J Med 1993; 328: 1725 – 31. 154. Brandes AA, et al. A multicenter retrospective study of chemotherapy for recurrent intracranial ependymal tumors in adults by the Gruppa Italiano Cooperative di Neuro-Oncologia. Cancer 2005; 104: 143 – 8.
Section 10 : Neurological Malignancies
63
Medulloblastoma and CNS Primitive Neuroectodermal Tumors Paul L. Moots and Mark T. Jennings
INTRODUCTION Medulloblastoma/primitive neuroectodermal tumor (PNET) accounts for approximately 25% of primary central nervous system (CNS) neoplasms in the pediatric age range, but only 1% of adult primary CNS neoplasms. The peak incidence is in the midportion of the first decade. The incidence declines rapidly after the midteens. Its occurrence is exceptionally rare after age 40, and there is not a second peak in incidence affecting older adults. Yet, 10–20% of patients with medulloblastoma are above than 16 years of age. The understanding of the biology of medulloblastoma has advanced dramatically in recent years. This neoplasm evolves from rapidly proliferating neural precursors in the primordium of the cerebellum. Analogous populations at other sites in the CNS likely give rise to other types of embryonal CNS neoplasms such as supratentorial PNETs. The association of medulloblastoma/PNET with genetic syndromes such as Turcot’s and Gorlin’s syndromes has long raised the suspicion that specific mutations were etiologic. Elegant studies on genetically engineered mice with mutations in pathways controlling proliferation and differentiation have established some of the critical alterations in the developmental program that lead to neogenesis. Some of these alterations may differ in adults compared with children. There are also important clinical distinctions between childhood and later arising medulloblastoma/PNET. Histological differences, including a higher likelihood of desmoplastic medulloblastoma, are seen in adults. The risk of subarachnoid metastases appears highest in young children and somewhat lower in adults. Treatment planning for adults has generally been inferred from results of randomized studies in children that have distinguished between patients at “standard risk” and those at “poor risk” for recurrence when treated with craniospinal radiation alone. The “poor-risk” children appear to have a substantial improvement in survival
with the addition of chemotherapy. It is unclear whether that advantage is also achieved in adults. In recent pediatric trials, chemotherapy has been added for standard-risk patients, with the aim of reducing the craniospinal radiation dose and thus reducing the delayed sequelae of radiation. Whether that strategy would be effective in adults is unknown. Although cognitive sequelae are not as severe in adults as in young children, they are still an important risk. Other late sequelae include endocrine dysfunction, fertility problems, vasculopathy, second cancers, and in some adolescents linear growth may still be an important concern. Despite these important concerns, medulloblastoma is one of the few curable primary CNS neoplasms. Overall survival (OS) in adults is similar to that of older children and teenagers.
BIOLOGY AND EPIDEMIOLOGY The incidence of medulloblastoma/PNET in children and teens is 0.62 per 100 000 person-years. In the 20- to 34-year age range, it is 0.21 per 100 000 and decreases by half in each of the subsequent two decades.1 A small percent of medulloblastomas/PNET arise in patients with well-defined genetic predispositions to cancer. These causes include Turcot’s syndrome, Gorlin’s syndrome, and hereditary retinoblastoma. Taken together, these genetic causes account for perhaps five percent of cases. Remarkably, neurofibromatosis type I and neurofibromatosis type II, which are strongly associated with the development of astrocytoma and ependymoma respectively, are not associated with the development of medulloblastoma/PNET. The developing brain has a relatively high susceptibility and/or propensity to undergo neoplastic transformation. Brain tumors are the most common solid tumor type in children, but this category includes a wide variety of distinct types. The brain has an extended period of histogenesis compared with other solid organs. The immaturity of the blood–brain
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
696
NEUROLOGICAL MALIGNANCIES
barrier during fetal and early postnatal life may also be a contributing factor in the high relative incidence of brain tumor in children. The populations of neuroepithelial precursors that are vulnerable to carcinogenic effects vary in location and temporal sequence during the development of the brain. It is likely that the various discrete populations of neuro-progenitor cells have at least partially distinct molecular mechanisms of neoplastic transformation, and differ in their susceptibility to various carcinogens. These factors may explain the important differences between medulloblastoma, other types of CNS PNET, and the glial (i.e. astrocytic, oligodendroglial, and ependymal) neoplasms common in children. These differential susceptibilities are also reflected in the marked differences in the types of brain tumors arising in children as compared with adults.2 Medulloblastoma derives from neural precursors in the rhombic lip, the primordium of the cerebellum, that are destined to form the external granular layer of the cerebellar cortex. These highly cellular tumors display histologic features typical of embryonal neoplasms (see Figure 1). Molecular investigations in mice have elucidated some of the critical factors in the normal development of the cerebellar cortex. These studies have also provided a very sophisticated understanding of the alterations in neurogenesis in this cell population that lead to neoplasia.3,4 Alterations that favor differentiation and suppress proliferation are at the core of these developmental mechanisms. Signaling through the Sonic Hedgehog – Patched (PTCH) pathway regulates mitogenic activity in these cells, stimulating proliferation through N-myc effects on cyclin D1 and cyclin D2 expression. Activity of the retinoblastoma gene product also contributes significantly in this pathway. Abnormal activity in this pathway, either by oversupply of the stimulatory ligand or by mutations that constitutively activate the downstream components of the pathway, can promote neoplasia. This type of animal model has a very strong resemblance to human medulloblastoma, and many of the genes that are implicated in medulloblastoma formation in mouse models are occasionally found to be abnormal in human medulloblastoma. Activation of the Wnt pathway by the binding of Wnt to its cell surface receptor, Frizzled, or by alterations in the activity of the intracellular components of this pathway, including
the adenomatous polyposis coli (APC) gene product, leads to an increase in beta-catenin in the nucleus. These pathways, as well as the Notched signaling pathway, are also important in cerebellar development. Activating mutations in the components of these pathways can also lead to medulloblastoma formation in mice. The occurrence of medulloblastoma in some patients with germline APC and PTCH mutations are explained by these controls on cell proliferation and differentiation in the developing cerebellum. Sporadic mutations in Wnt and Sonic-hedgehog pathways have also been identified in medulloblastoma at a frequency up to 10%. This knowledge has fostered a great interest in targeted therapy. Inhibitors of the Sonic-hedgehog pathway, such as cyclopamine, are under consideration as therapeutic agents.5,6 Some of the important goals of the molecular understanding of neoplasia include the use of molecular markers as prognostic indicators, as criteria for stratification into different treatment groups and as indicators of likely response to a particular therapy. However, the development of patient subgroup stratification based on molecular diagnosis for prognostic purposes and for targeted therapies in individual patients lags behind the knowledge base.7 Today, criteria for risk stratification in medulloblastoma still rely almost entirely on the clinical features of residual disease after surgery and the presence or absence of metastases (see below). Epidemiologic studies of childhood brain tumors provide some very interesting suggestions about environmental factors.8 As with adult gliomas, exposures to nitroso compounds are associated with an increased relative risk. These may come from dietary sources, medications, and other chemical exposures. Exposure during gestation as well as in postnatal life may be important. Exposure to ionizing radiation also increases the risk of brain tumors, and again with a long latency. There is some evidence to suggest that exposures to pesticides during pregnancy or soon after birth or parental exposure to some pesticides is associated with an increased risk of childhood brain tumors. Tobacco use in adults has not been associated with brain tumor development. Epidemiologic studies in children of parents who smoke have given mixed results. Some studies indicate paternal smoking is associated with an increased risk of brain tumors, while other studies have been negative. Exposure to common viral infections such as influenza, as well as exposure to SV-40 virus, has been associated with an increased risk in some studies. Remarkably, prenatal vitamin use has been associated with a decreased risk of childhood brain tumors, perhaps related to direct effects on CNS development analogous to the reduced risk of spinal dysraphism associated with folate use or due to antioxidant effects that limit prenatal exposure to potential carcinogens.
CLINICAL PRESENTATION
Figure 1 Medulloblastoma. These highly cellular lesions are populated by cells with hyperchromatic nuclei and little cytoplasm. Mitoses and focal necrosis are common. Hematoxylin and eosin; original magnification ×240).
The most common presenting symptoms for adults with medulloblastoma are headaches (83%), nausea/vomiting (43%), gait imbalance (40%), and dizziness (23%). Cranial neuropathies including diplopia are less common. Presentations with spinal symptomatology such as back or radicular
MEDULLOBLASTOMA AND CNS PRIMITIVE NEUROECTODERMAL TUMORS
pain, or with signs of myelopathy due to subarachnoid “drop metastases” are uncommon in adults, but when an apparently isolated intradural-extramedullary spinal lesion is found, medulloblastoma must be considered in the differential diagnosis. The duration of symptoms prior to diagnosis is typically 2–3 months but can be more than 6 months.9,10 Alternatively, a few patients will present with a dramatic decline after a very short duration of symptoms. Intratumoral hemorrhage or obstruction of cerebrospinal fluid (CSF) flow through the cerebral aqueduct and fourth ventricle can lead to acute hydrocephalus and cerebellar herniation through the foramen magnum. This life-threatening emergency requires emergency measures for neurological stabilization. Medical interventions such as high-dose corticosteroids (e.g. dexamethasone 40 to 100 mg i.v. bolus), mannitol, and hyperventilation may provide a few hours for definitive management. Intubation can sometimes precipitate such a decline. Definitive treatment of this neurological emergency is surgical decompression often accompanied by ventriculostomy. Magnetic resonance imaging (MRI) scanning is currently the best method of assessment for the types of symptoms related to intracranial tumors. This is particularly true for posterior fossa lesions because MRI scanning provides a much better image of the posterior fossa with fewer artifacts than does computed tomography (CT) scanning. Medulloblastoma is generally an intensely enhancing, relatively discrete mass in the cerebellar midline or hemisphere.11 A lateral hemispheric location is more common in adults than in children. Occasionally, the mass may appear to arise in the cerebellopontine angle or the middle cerebellar peduncle. In a very small percentage, enhancement is absent or minimal. Rarely, the mass appears cystic. Evidence of invasion of the floor of the fourth ventricle carries a poor prognosis in some adult medulloblastoma series. The scan should be carefully reviewed for evidence of subarachnoid or intraventricular subependymal spread, as well as for hydrocephalus. The differential diagnosis arising from the posterior fossa location and MRI characteristics includes ependymoma, and cerebellar forms of astrocytoma such as pilocytic astrocytoma. The latter often, but not always, have a distinctive cystic appearance. Other types of glioma such as glioblastoma and anaplastic oligodendroglioma are exceedingly rare in this location. In the adult population, metastasis is an important part of the differential. The fact that most adult medulloblastomas present as solitary lesions, and the age being rarely over forty, tends to favor a primary neoplasm rather than a metastasis. A few other mass lesions arising in the posterior fossa, such as meningioma, choroid plexus papilloma, and acoustic schwannoma of the cerebellopontine angle, will on occasion have an atypical appearance that will be difficult to distinguish from medulloblastoma radiographically. In children, the importance of identifying subarachnoid metastases is well established. This has important prognostic implications and greatly influences treatment planning. Detailed staging of the neuraxis with MRI imaging of the entire spine and CSF cytology studies are warranted if medulloblastoma is suspected. There is an advantage to obtaining these studies prior to surgery. Postoperative artifacts, including meningeal enhancement at the operative site, meningeal
697
enhancement more widely intracranially, spinal meningeal enhancement, and blood tracking along the spinal epidural space, may be difficult to distinguish from meningeal metastases. CSF cytology is also an important part of neuraxis staging but is often contraindicated until after surgical decompression. It is widely held that neuraxis staging is also essential in adults with medulloblastoma/PNET.12 The presence of subarachnoid metastases or positive CSF cytology at diagnosis is less common in adults (10–30%) than in children. Many, but not all, adult studies identify subarachnoid metastases as a poor prognostic factor. The demonstration of subarachnoid metastases will alter the radiation planning, often including boosts to regions of bulky metastases. Chemotherapy decisions also are heavily influenced by staging results. Hydrocephalus is another problem that contributes to the presenting symptomatology of headaches, nausea, and vomiting. About 25% of adults have hydrocephalus that requires placement of a ventriculoperitoneal (VP) shunt. Many neurosurgeons prefer to place a temporary ventriculostomy drain prior to or at the time of surgery. After resection of the tumor, CSF flow is often reestablished to an adequate degree. At that point, the ventriculostomy can be clamped and then removed after a period of observation. This avoids the need for a permanent VP shunt for some patients. Shunt malfunction, infection, and the relatively uncommon occurrence of tumor dissemination with peritoneal seeding via the shunt are considerations leading away from VP shunting when possible.13
TREATMENT The prognosis and treatment plans for medulloblastoma are strongly influenced by age. For infants and young children (<3 years), surgery is generally followed by chemotherapy, primarily attempting to avoid or defer craniospinal radiation with the attendant severe neuropsychological and growthretarding sequelae. The efficacy of chemotherapy as the primary treatment modality has received extensive evaluation in this setting, which in essence represents a neoadjuvant approach. A 5-year progression-free survival (PFS) of 32% with few patients requiring radiation was achieved in a recent trial.14 For older children and teenagers, surgery is generally followed by craniospinal radiation. In the subgroup of patients with poor prognostic features, such as subtotal resection, subarachnoid metastases, or systemic metastases, the addition of chemotherapy improves survival. Neoadjuvant chemotherapy has been used in this age-group to improve the completeness of cytoreduction achieved prior to the institution of radiation. However, some reports indicate that a delay in the initiation of radiation therapy may be associated with a worse outcome. Chemotherapy has also been used to allow a reduction in the craniospinal radiation dose used for good prognosis patients, again as a means of limiting radiation toxicities. The treatment of medulloblastoma in adults has generally been patterned after that of older children, although there are some important biological distinctions. Craniospinal radiation with a boost to the primary site is generally considered standard therapy. This is not infrequently accompanied by
698
NEUROLOGICAL MALIGNANCIES
chemotherapy. The toxicities of craniospinal radiation in this age-group differ substantially from those of children. While effects on bone growth are no longer an issue, and the neuropsychological impact of usual craniospinal radiation doses for adults is less than for children, irradiation of the axial skeleton does affect the bone marrow sufficiently to limit the intensity of subsequent chemotherapy. Thus, in the adult with medulloblastoma and other primary brain tumors that require craniospinal radiation, a new rationale emerges for the use of chemotherapy in a neoadjuvant design.
Medulloblastoma in Children The prognosis and treatment of childhood medulloblastoma has been the subject of a number of cooperative group trials.15 – 17 The most important prognostic variables are (i) tumor size (T stage), (ii) extent of surgical resection, (iii) presence or absence of metastases (M stage – see Table 1), and (iv) age. In most recent studies, postoperative assessment of residual disease has replaced the preoperative T stage for assessment of risk.18 – 20 Residual disease postoperatively of greater than 1–1.5 cm carries a poor prognosis. In addition, failure to achieve a complete response with subsequent therapy is also demonstrated in some studies to be a predictor of early progression, although a relationship between the extent of response and OS has not been demonstrated in some neoadjuvant chemotherapy studies.21 A variety of additional prognostic factors have been established. Histologic evidence of neuronal differentiation has been reported as valuable for prognostic purposes in some reports.22 Desmoplastic histologic appearance is also associated with a better outcome in some studies. Large cell variants and anaplastic variants have a worse prognosis. Molecular analysis is also proving valuable. Expression of differentiation promoting factors such as Trk C are associated with a favorable outcome. Conversely, overexpression of the c-myc oncogenes is associated with a poor outcome. Gene expression profiling of medulloblastoma demonstrated by DNA microarray appears to have a powerful ability to predict outcome independent of clinical parameters.23 Stratification of prognosis that goes beyond basic clinical factors and includes biological characteristics is rapidly evolving and is increasingly included in pediatric cooperative group trials.24,25 Another important issue that combines histologic and molecular assessment in the differential diagnosis of medulloblastoma is the distinction from the CNS atypical teratoid (AT)/rhabdoid tumor (RT). The histological appearance of Table 1 Metastasis stage for medulloblastoma, PNET.
M0 M1 M2
M3 M4
No evidence of subarachnoid or hematogenous metastasis. Tumor cells demonstrated in CSF. Nodular seeding in the cerebellar, cerebral subarachnoid space, or in the third or lateral ventricles. (Note: M2 may be confirmed by pathologic demonstration of meningeal involvement adjacent to the primary tumor site in the cerebellum.) Nodular seeding in the spinal subarachnoid space. Metastasis outside the cerebrospinal axis.
Source: Derived from Chang et al.,18 with permission of the Radiological Society of North America, 1969.
AT/RT can be difficult to distinguish from medulloblastoma. The diagnosis is established by mutation analysis of the SMARCB1 gene (also called INI1 ).26,27 AT/RT are predominantly found in children under 3 years of age, but in that age-group comprise up to 20% of malignant CNS tumors. AT/RT has an extremely poor prognosis with a 15% 2-year survival for children under 3 years of age. Older children have a somewhat better outcome.28 Mutational analysis is becoming a standard part of the pathological evaluation of malignant CNS tumors in young children. Assessment of SMARCB1/INI1 expression can also be done by immunohistochemistry.29 The current clinical schema for risk stratification is based on the presence of residual tumor after surgery, and the presence or absence of metastatic disease. Children are categorized as “standard risk” (no or minimal residual tumor, M0) or “high risk” (>1–1.5 cm2 residual tumor, M1–M4) for recurrence when treated with radiation therapy alone. The respective 5-year event-free survival rates for these groups are approximately 65 and 0%.15 In addition to the completeness of surgical resection, other treatment factors have a substantial impact on outcome for patients with medulloblastoma. The adequacy of radiation therapy in dose and distribution is critical, as demonstrated by the improved survival observed, following the institution of craniospinal radiation for medulloblastoma many years ago.30 – 32 Decreased radiation dose to the craniospinal axis without the addition of chemotherapy is associated with a decline in event-free survival.33 For children with “poorrisk” characteristics, chemotherapy following radiation has been demonstrated to produce a substantial increase in 5-year event-free survival compared to radiation alone (48 vs 0%)15,34 Children less than 4 years of age have a worse prognosis than older patients, with a 5-year event-free survival of only 32%. This has been attributed in part to a higher incidence of subarachnoid dissemination at the time of diagnosis in young children (M1–M3, 34%). For children over 4 years, the incidence of metastases at diagnosis is 14–25%.35 However, in comparison, a recent report of 40 adult patients with medulloblastoma identified metastases in 33%.12 In infants, the response rates and long-term efficacy of chemotherapy have been studied utilizing a variety of agents and combinations. The combination of cisplatin, cyclophosphamide, VP-16, and vincristine produced a response rate (CR + PR) of 48% with a 2-year event-free survival of 34%.36 Substitution of ifosfamide for cyclophosphamide produced equivalent results. This multiagent combination has also been used as an intensive induction regimen prior to myeloablative chemotherapy with autologous bone marrow rescue. This achieved a 40% event-free survival at 24 months.37,38 A number of other combinations have demonstrated similar results in this age-group. The “8-in-1” combination produced a 43% response rate after induction, with a 3-year event-free survival of 22%.39 Radiation therapy remains an efficacious salvage therapy for infants and young children who have recurrence after chemotherapy. A complete response to radiation therapy was achieved in 6 of 11 patients (55%) who progressed during
MEDULLOBLASTOMA AND CNS PRIMITIVE NEUROECTODERMAL TUMORS
chemotherapy.40 High-dose chemotherapy with autologous bone marrow rescue may also serve as effective salvage therapy for these children.41 The treatment of medulloblastoma in children from midto late childhood through the early teen years forms the bulk of our knowledge about this neoplasm. Within this age range, a comparison of event-free survival rates for older children (i.e. 4–7 and 8–13 years) is not statistically different from that for teens (>14 years) in most trials. However, some studies focusing specifically on teenagers (ages 10–20 years) demonstrate a positive linear relationship between age and time to relapse.42 Standard treatment for these children includes craniospinal radiation. Postradiation chemotherapy with CCNU and vincristine or CCNU, vincristine, and cisplatin has been demonstrated to improve outcome for “poor-risk” patients. Packer et al.43 describe a 5-year event-free survival of 85% using the latter combination in a cohort of 63 “poor-risk” patients, results that equal the outcome for “standard-risk” patients. The use of adjuvant chemotherapy for “standardrisk” patients has been investigated as a means of allowing a reduction in the craniospinal radiation dose.44,45 A number of reports describe the use of neoadjuvant chemotherapy followed by craniospinal radiation in children with “poor-risk” medulloblastoma/PNET. We have observed a major cytoreductive response in 60% of medulloblastoma/PNET patients using an intensive regimen of Cisplatin, cyclophosphamide, VP-16, and vincristine prior to hyperfractionated radiotherapy.46 This chemotherapy regimen has been adapted for use in adults with good results, and has served as the basis for an Eastern Cooperative Oncology Group (ECOG) trial in adult medulloblastoma (see below). Others have made similar observations. A radiographically documented response rate of 43% was achieved in 30 patients using preradiation cyclophosphamide, cisplatin, and vincristine. The 2-year event-free survival was 40%.21 An 85% response rate was observed in 11 children with the combination of cisplatin and VP-16 in a neoadjuvant setting, with a 2-year event-free survival of 76%.47 The combination of carboplatin and VP-16 produced a 53% response rate prior to radiation. Progression during chemotherapy was more frequent with carboplatin than had been observed with regimens containing cisplatin.48 The “8-in-1” combination produced a 50% response in two cycles administered prior to radiation.49 A single randomized trial comparing treatment without or with preradiation chemotherapy using vincristine, procarbazine, and methotrexate for six weeks failed to demonstrate a benefit for the neoadjuvant therapy.50 Topotecan has been effective in a neoadjuvant setting, with a response rate of 25% and an additional 47% with stable disease after two cycles.51 The rationale for neoadjuvant chemotherapy for “poorrisk” medulloblastoma in this age-group is fourfold: (i) the survival benefit obtained by surgical cytoreduction may be increased by the added cytoreduction due to chemotherapy given prior to radiation, (ii) responsiveness to chemotherapy may be greater prior to radiation, (iii) the efficacy of radiation may be improved as the volume of residual
699
tumor is decreased, and (iv) the ability to tolerate intensive chemotherapy declines following craniospinal radiation. Overall, the results of neoadjuvant chemotherapy in children are similar to those achieved with postradiation chemotherapy.22 While maximal cytoreduction prior to radiation therapy may be advantageous, a number of pediatric studies raise concerns about the potential detrimental effect of delaying the institution of radiation therapy. For example, in one Children’s Cancer Group (CCG) study, only 64% of 82 children with medulloblastoma/PNET completed five planned cycles of intensive chemotherapy – vincristine, VP-16, cisplatin, cyclophosphamide alternating with carboplatin, and VP-16 – prior to radiation.52 The most common cause of failure to complete therapy was progression of disease. Of those children who completed the planned chemotherapy, all but three completed craniospinal hyperfractionated radiation. However, using relatively high craniospinal radiation doses, substantial myelotoxicity and gastrointestinal toxicity were encountered. Other reports also indicate a small but significant rate of disease progression during chemotherapy, and the potential to interfere with the planned radiation.53
Medulloblastoma in Adults Medulloblastoma in adults shares many similarities with that of children in regard to pathology, prognostic characteristics, treatment response, and outcome. Some distinctions have been observed, notably a higher incidence of laterally placed tumors and a higher incidence of desmoplastic histologic features in adults (see Figure 2). Some reports suggest that these features are associated with a better prognosis. MRI cannot reliably distinguish desmoplastic from classical medulloblastoma in adults.11 The value of clinical staging for prognosis and treatment decision has not been prospectively confirmed in adults, and yet the approach defined in children is widely used.54 In a group of patients consistently staged for the extent of postoperative residual disease and evidence of metastases, 59% of 44 adult patients were characterized as “poor risk”.12
Figure 2 Desmoplastic medulloblastoma. Growth into the leptomeninges produces a whorled or glomeruloid appearance. Hematoxylin and eosin; original magnification ×240.
700
NEUROLOGICAL MALIGNANCIES
Thirteen of these patients (30%) had metastatic disease. This study and others suggest that the percentage of adults with “poor-risk” features is somewhat lower than children, and yet the percentage of M1–M4 patients is similar. The study by Prados et al. also demonstrates that distinguishing “standardrisk” and “poor-risk” adult patients is statistically significant in regard to outcome. The 5- and 10-year survival rates for adults with medulloblastoma range from 48 to 78% and from 38 to 55%, respectively, in multiple studies.55 – 57 Thus, despite somewhat better prognostic features at diagnosis, the overall outcome for adults with medulloblastoma is similar to that of older children. The approach to treatment of adults with medulloblastoma has largely been inferred from the randomized trials performed with children. Reports focusing on adults have generally confirmed the benefit of surgical resection over biopsy. However, Carrie et al. report that poor postoperative performance status (ECOG performance status 3) is significantly associated with a worse prognosis, a factor that may limit the benefit of surgical debulking.56 As with children, radiation therapy utilizing craniospinal radiation with a boost to the primary tumor site is central to achieving long-term disease control. The standard craniospinal dose is 36 Gy, and care must be taken to include the full extent of the neuraxis. Special attention must be paid to the cribiform plate and the lateral and caudal projections of the spinal theca. Reduction of the craniospinal dose in conjunction with adjuvant chemotherapy for “standard-risk” children is a strategy that has not been evaluated in adults. The standard tumor dose is 54 Gy, with lower doses being less effective.58 Traditionally, the radiation port for the primary site is defined anatomically as the entire posterior fossa. Less commonly, the port is defined as the enhancing mass plus a 2-cm margin. The use of newer radiation planning methods such as threedimensional conformal planning appears to provide effective local control while sparing radiation to the temporal bones and cochlea.59 This reduces the likelihood of auditory nerve damage. Most adults complete craniospinal radiation without interruption. However, radiation-related neutropenia was seen in 17% of adults.12 Nadir white blood counts (WBCs) tend to be in the 2000–2500 m−3 range. In adults treated with nitrosourea-based chemotherapy following radiation, significant myelosuppression was observed in 33–45%, usually requiring dose reduction. Even in pediatric trials (CCSG/RTOG) involving postradiation CCNU (100 mg m−2 ) and vincristine, only 60% of patients received 75% or more of the planned chemotherapy.15 Packer et al., using CCNU (68 mg m−2 ), cisplatin (75 mg m−2 ), and vincristine after radiation, observed only modest hematological toxicity in children.60 Fifteen percent of patients required CCNU dose reductions. However, only 44% received the prescribed cisplatin dose because of various toxicities including audiologic, neurologic, renal, and hematologic. Audiologic toxicity from cisplatin is worse following cranial radiation than when given prior to radiation, and is a dose-limiting toxicity.61,62
The overall 5-year survival for “standard-risk” adults treated with radiation therapy has been 70–80% in most studies, a figure very similar to that for older children. Despite a somewhat lower event-free 5-year survival of 58% for their “standard-risk” adult patients, Prados et al. conclude that at present radiation therapy alone is the best recommendation for those patients.12 The likelihood of additional benefit from chemotherapy for “standard-risk” adult patients is very debatable. The use of chemotherapy for poor-risk adults with medulloblastoma is unsettled but has tended to follow from that for children. Studies of patients with recurrent medulloblastoma have demonstrated relatively high response rates, occasionally achieving good long-term disease control (i.e. greater than 12 months), but not cure. Responses have been achieved with the same spectrum of agents that are used in children, most notably cisplatin, carboplatin, cyclophosphamide, VP-16, and vincristine. Nitrosoureas are also active, but less so, and have a greater likelihood of prolonged myelosuppression.63 Topotecan and temodar are also drugs with some demonstrated activity. The use of pediatric chemotherapy regimens for adults with medulloblastoma has been detailed in a number of reports. Among the best results in children with medulloblastoma are the combination of weekly vincristine during radiation and postradiation chemotherapy with eight planned cycles of cisplatin, CCNU, and vincristine developed by Packer et al..43 A 5-year PFS of 85% was achieved in poor-risk children. A report of 10 adults treated with this regimen resulted in a median PFS of 26 months, and an OS of 36 months. Seven patients completed the planned eight cycles, and 9 of 10 required dose reductions of all three drugs. The toxicities appeared greater in adults than had been reported in children.64 In seven patients treated on a Pediatric Oncology Group trial utilizing alternating cycles of cisplatin/VP-16 and cyclophosphamide/vincristine followed by radiation, the PFS was 48 months and the OS was 57 months. Even among patients aged 10 to 20 years who are treated on pediatric medulloblastoma protocols, the incidence of toxicities and the number of required modifications is greater than in younger children.65 Spreafico et al. report results of 26 adult patients treated on pediatric medulloblastoma protocols.66 Two regimens were used: (i) preradiation high-dose methotrexate or (ii) sequential doses of methotrexate/vincristine, VP-16, cyclophosphamide, and carboplatin/vincristine. Both regimens included postradiation CCNU and vincristine. The median age was 26 years (range 18–41). Eighteen patients (69%) were standard risk and 8 (31%) were high risk. Radiation doses were similar to those previously described, although some patients received a twice daily hyperfractionated schedule. Eight patients had recurrence at a median time of 34 months. As is typical in most studies, the majority of patients recurred in the posterior fossa (75%), but simultaneous activity at other sites was common. The 5-year PFS was 65%, and OS was 73%. The limitations to intensive chemotherapy, resulting from decreased bone marrow reserve following craniospinal
MEDULLOBLASTOMA AND CNS PRIMITIVE NEUROECTODERMAL TUMORS
radiation in adults, provide a different rational for neoadjuvant treatment in adult medulloblastoma patients. The limited data on neoadjuvant chemotherapy for adults suggest that high response rates and excellent cytoreduction can be achieved in this setting. We have treated six “poorrisk” adult medulloblastoma patients with three cycles of cisplatin, etoposide, cyclophosphamide, and vincristine followed by craniospinal radiation.46 Radiographic responses were observed in all six (three CR, three PR) prior to radiation. All six patients completed radiation therapy as planned, although two had neutropenia sufficient to require G-CSF. None of these six patients experienced disease progression prior to radiation therapy. One patient died with progressive disease at 24 months, and a second died of an acute pneumonitis of uncertain etiology after completion of all therapy.67 This protocol, which also included patients with disseminated ependymoma, was opened for enrollment in the ECOG. Evaluation of long-term follow-up data is under way.68 One prospective trial in adult medulloblastoma has been reported in detail by Brandes et al.69 This multicenter trial enrolled 36 patients over 12 years. The median age was 26 years (range 18–57). The median Karnofsky score was 78 (range 40–90). Standardized neuraxis staging with MRI and CSF studies was performed. Positive CSF cytology was found in 30%. Nodular seeding of the meninges was found in 22%, some of whom also had positive cytology. The tumor histology was desmoplastic in 34%. Low-risk patients (n = 10) received 36 Gy to the craniospinal axis, followed by a boost to the entire posterior fossa to a total of 54.8 Gy. High-risk patients (n = 26) received two cycles of preradiation chemotherapy, initially using a MOPP-like regimen (n = 6), and in latter patients a combination of cisplatin, cyclophosphamide, and VP-16 (n = 16). Four highrisk patients refused chemotherapy. The median PFS for the entire group was 6.7 years. For low-risk patients, PFS at 5 years was 76%, and for high-risk patients it was 61%. OS for the entire group was 8.15 years and was also substantially better in the low-risk group with 89 versus 69% OS at 5 years. M0 patients did substantially better than M+ patients in PFS (75 vs 45%) and in OS (87 vs 52%) at 5 years. M status was the strongest predictor of outcome. Both the Brandes et al. and Spreafico et al. reports, as well as our own experience, have shown that intensive preradiation chemotherapy for adults with medulloblastoma can be accomplished without interfering with the planned radiation. Both the total dose and the duration of time required to deliver craniospinal radiation did not deviate substantially from protocol. Progression during preradiation chemotherapy is uncommon. It was not seen in either of these published studies, although we have seen it twice in more recently treated patients. Collin’s law describing the period of risk of recurrence for embryonal neoplasms has been applied to medulloblastoma, but even in children exceptions occur. In adults, this is not applicable. Medulloblastoma is notable for late recurrences. PFS continues to decline after 5 years for high-risk patients in particular. Some studies show no additional recurrence after 5 years in standard-risk patients.69 Long-term surveillance
701
imaging by MRI scanning is the primary means of followup. Typically, scans for monitoring in the absence of new symptoms would be obtained every 3 to 4 months for the first year after completion of therapy, and then at 6-month intervals. After 5 years, it would be reasonable to scan yearly for standard-risk patients. Since recurrence in the cerebellum is by far the most common pattern, it is usually not necessary to scan the entire neuraxis each time unless there is known spinal subarachnoid seeding. It is reasonable to scan the spine for high-risk patients at least yearly even if they did not have visible metastases. Spinal fluid cytology should be monitored routinely in patients with prior positive results. Evaluation for systemic metastases is generally only done when warranted by symptoms. Identification of asymptomatic recurrence by surveillance imaging is associated with a better survival than when recurrence presents as symptomatic disease.70 The treatment of relapse is largely approached by chemotherapy. In selected patients with localized recurrence, usually in the posterior fossa, surgical and radiation therapies may be appropriate. Even for these patients, chemotherapy would be recommended. The same spectrum of agents noted previously would be considered. Retrial of a regimen that produced a response and was associated with a very long progression-free duration is sometimes chosen, but more often an alternative is used. Many regimens have been demonstrated to produce responses in recurrent medulloblastoma for many months, and not infrequently good durable responses are achieved. In 17 adults with recurrent medulloblastoma, the median survival from time of recurrence was 21 months (range 0–67+).71 In a cohort of twenty adolescents with recurrent medulloblastoma, largely treated with chemotherapy, PFS at 1 year was 25% and OS at 1 and 2 years was 55 and 10%.42 Myeloablative chemotherapy has been investigated in this setting as well. In pediatric trials, long-term recurrence-free survival is sometimes achieved. In 30 children with recurrent medulloblastoma treated with high-dose chemotherapy and autologous stem cell rescue, 30% were recurrence free at a median of 54 months.72 In 11 adults with recurrent medulloblastoma treated with high-dose chemotherapy and autologous stem cell support, the 2-year PFS and OS were 54 and 72%. However, seven patients had recurred at a median time of 33 months. The treatment-related mortality was 18%.73 This approach remains limited to investigational studies. Long-term sequelae are another major concern after treatment for medulloblastoma. Chronic or delayed sequelae have been associated with all aspects of therapy. Impaired cognitive development is one of the most obvious. This can result from hydrocephalus complicating a posterior fossa mass. Another cause is postoperative cerebellar mutism.74 This peculiar syndrome is seen in about 25% of children and a smaller number of adults after resection of cerebellar tumors. The major features resolve over many weeks, but some long-term sequelae in speech and motor function have been observed. Craniospinal radiation is a major cause of impaired cognitive development in these patients. Age and dose are the strongest determinants of subsequent cognitive impairment,
702
NEUROLOGICAL MALIGNANCIES
and hence the efforts to delay or decrease the radiation dose in children. Neuropsychological assessment of 16 longterm survivors of childhood medulloblastoma (mean age at diagnosis 7.6 years; mean age at testing 22 years) demonstrated marked impairments in many areas, including memory, executive functioning, and motor functions. Language was relatively well preserved. The estimated IQ for these 10-year survivors was 75, with verbal IQ being better than performance IQ. Most of these patients required special education, did not drive or maintain steady employment, and did not live independently.75 Very remarkably, self-reporting of overall quality of life and multiple subcategories related to functional, social, psychological, and family well-being by these patients and their caregivers were identical to the general population. This result has been observed in other groups of patients with chronic conditions. Even with risk-adapted therapy using lower craniospinal radiation doses, these problems still occur.76 Cognitive sequelae in adults are, in general, less severe than in children who have received craniospinal radiation. Many maintain the ability to live independently, maintain employment even at high-functioning jobs, maintain driving skills and function well as family members and parents. However, careful testing demonstrates a similar pattern of impairment across multiple cognitive spheres. In 10 adult medulloblastoma patients (mean age at diagnosis 30 years; mean age at testing 36.5 years; mean interval from treatment 6.6 years), the mean IQ score was 90 (range 67–103).77 In children, the other major treatment sequelae include endocrinopathies, growth retardation, infertility, and second malignancies. Hypothyroidism and hypogonadism are the most common endocrinopathies in adults. Panhypopituitarism is less common. Hypothyroidism occurs in at least 25% within a few years after treatment and should be screened for annually. These contribute to infertility, although for female patients radiation of the sacral spinal canal may affect ovarian function. Surgical repositioning of the ovaries before radiation has been advocated by some. Second malignancies occur at a 5.4-fold increased risk in children previously treated for medulloblastoma. Most of these occur within the radiation field. Meningiomas and sarcomas, particularly of the skull, are the most common. Other second neoplasms include other primary CNS tumors, salivary gland tumors, thyroid cancer, and acute leukemia.78
Primitive Neuroectodermal Tumor (PNET) in Adults PNETs of the nervous system are a collection of rare neoplasms that share some histological and biological characteristics with medulloblastoma, particularly in regard to embryonal histological features. In some instances, the appearance of a PNET is indistinguishable from a medulloblastoma; however, the neoplasm is located outside of the cerebellum (i.e. supratentorial PNET). In other instances, the neoplasm may demonstrate distinctive histologic features of differentiation such as ependymal rosettes, neural rosettes, or evidence of pineal differentiation. These various neoplasms, S-PNET, ependymoblastoma, cerebral neuroblastoma, and pineoblastoma, as a group have a risk of subarachnoid spread that is similar to medulloblastoma. As with medulloblastoma, this
results in a high risk of recurrence at other sites within the nervous system, and infrequently outside the nervous system. Thus, craniospinal radiation is a standard part of treatment. The response to chemotherapy appears to be less frequent than that observed for medulloblastoma. In infants, the response (CR + PR) to multiagent chemotherapy using cisplatin, etoposide, cyclophosphamide, and vincristine administered with the intent of deferring radiation was 29% for S-PNET.36 The response rate to the “8-in-1” regimen was 33% for S-PNET and 0% for pineoblastoma.39 In six patients with newly diagnosed pineoblastoma treated with cyclophosphamide as a single agent, three achieved responses and three had stable disease.79 High-dose chemotherapy with stem cell rescue has been reported with some encouraging results in this relatively refractory population. A study of 17 PNET patients, all but two of whom were pediatric (median age 3.6 years), pineoblastomas (n = 8) did extremely poorly. However, supratentorial cortical PNETs (n = 7) did remarkably well, with five remaining disease free at a median of 8 years after treatment. However, four of these received radiation following the chemotherapy.80 Another trial of high-dose chemotherapy and radiation for newly diagnosed pineoblastoma, including six children and six adults, achieved a median PFS of 62 months for nine patients.81 Whether chemotherapy adds additional benefit to radiotherapy for adults with PNET is unknown, although these patients are usually considered “high risk” on the basis of their poorer prognosis and often receive combined modality therapy.
AUTHORS’ RECOMMENDATIONS The rarity of medulloblastoma/PNET in adults makes it unlikely that a large prospective, randomized trial will be accomplished in this age-group. Pediatric trials should encourage enrollment of teenagers and young adults, up to the age of 25 or 30 years, even though this necessitates some modification of treatment. This becomes increasingly important, as sophisticated molecular analysis of medulloblastoma samples holds such high potential for improving risk stratification and individualization of treatment. Currently, the clinical stratification of risk remains a useful guide for managing adults. Standard-risk patients are best managed with craniospinal radiation. The concept of reduced neuraxis dosing with adjuvant chemotherapy will not be testable in the adult population unless they can be included in pediatric trials. We have favored preradiation chemotherapy and craniospinal radiation for high-risk adults. The benefit of this approach is that it allows for more intensive chemotherapy dosing. It is not clear that better long-term control is achieved. In pediatric trials, preradiation chemotherapy does not produce a clearly better outcome in the high-risk group. As treatments with better toxicity profiles become available, particularly agents with less hematologic and ototoxicity, the rationale for preradiation therapy becomes less attractive.
MEDULLOBLASTOMA AND CNS PRIMITIVE NEUROECTODERMAL TUMORS
ACKNOWLEDGMENTS The authors gratefully acknowledge Elizabeth M. Wright for editorial assistance and help in manuscript preparation.
REFERENCES 1. CBTRUS. Statistical Report: Primary Brain Tumors in the United States 1995 – 1999: The Central Brain Tumor registry of the United States, 2002. 2. Rubinstein LJ. The correlation of neoplastic vulnerability with central neuroepithelial cytogeny and glioma differentiation. J Neuro-oncol 1987; 5: 11 – 27. 3. Marina S. Medulloblastoma: developmental mechanisms out of control. Trends Mol Med 2005; 11: 11 – 22. 4. Rice J. Causation of nervous system tumors in children: insights from traditional and genetically engineered animal models. Toxicol Appl Pharmacol 2004; 199: 175 – 91. 5. Pomeroy SL, Sturla LM. Molecular biology of medulloblastoma therapy. Pediatr Neurosurg 2003; 39: 299 – 304. 6. Romer J, Curran T. Targeting medulloblastoma: small-molecule inhibitors of the sonic hedgehog pathway as potential cancer therapeutics. Cancer Res 2005; 65: 4975 – 8. 7. Gilbertson RJ, Gajjar A. Molecular biology of medulloblastoma: will it ever make a difference to clinical management. J Neuro-oncol 2005; 75: 273 – 8. 8. Baldwin RT, Preston-Martin S. Epidemiology of brain tumors in children - a review. Toxicol Appl Pharmacol 2004; 199: 118 – 31. 9. Kunschner LJ, et al. Survival and recurrence factors in adult medulloblastoma: the MD Anderson Cancer Center experience from 1978 – 1998. Neuro-oncol 2001; 3: 176 – 3. 10. Peterson K, Walker RW. Medulloblastoma/primitive neuroectodermal tumor in 45 adults. Neurology 1995; 45: 440 – 2. 11. Malheiros SM, et al. MRI of medulloblastoma in adults. Neuroradiology 2003; 45: 463 – 7. 12. Prados MD, et al. Medulloblastoma in adults. Int J Radiat Oncol Biol Phys 1995; 32: 1145 – 52. 13. Magtibay PM, et al. Unusual presentation of adult metastatic peritoneal medulloblastoma associated with a ventricular shunt: a case study and review of the literature. Neuro-oncol 2003; 5: 217 – 20. 14. Geyer JR, et al. Multiagent chemotherapy and deferred radiotherapy in infants with malignant brain tumors. J Clin Oncol 2005; 23: 7621 – 31. 15. Evans AE, et al. The treatment of medulloblastoma. Results of a prospective randomized trial of radiation therapy with and without CCNU, vincristine and prednisone. J Neurosurg 1990; 72: 572 – 82. 16. Tait DM, et al. Adjuvant chemotherapy for medulloblastoma: the first multi-centre control trial of the International Society of Pediatric Oncology (SIOP I). Eur J Cancer 1990; 26: 464 – 9. 17. Zeltzer PM, et al. Metastasis stage, adjuvant treatment and residual tumor are prognostic factors for medulloblastoma in children: conclusions from the Children’s Cancer Group 921 randomized phase III trial. J Clin Oncol 1999; 17: 832 – 45. 18. Chang CH, Housepian EM, Herbert C. An operative staging system and megavoltage radiotherapeutic technique for cerebellar medulloblastoma. Radiology 1969; 93: 1351 – 9. 19. Laurent JP, Chang CM, Cohen ME. A classification system for primitive neuroectodermal tumors (medulloblastoma) of the posterior fossa. Cancer 1985; 56: 1807 – 9. 20. Jenkin D, et al. Posterior fossa medulloblastoma in childhood: treatment results and proposal for a new staging system. Int J Radiat Oncol Biol Phys 1990; 19: 265 – 74. 21. Mosijczuk AD, et al. Preradiation chemotherapy in advanced medulloblastoma. A Pediatric Oncology Group pilot study. Cancer 1993; 72: 2755 – 62. 22. Cohen BH, Packer RJ. Chemotherapy for medulloblastoma and primitive neuroectodermal tumors. J Neurooncol 1996; 29: 55 – 68. 23. Fernandez-Teijeiro A, et al. Combining gene expression profiles and clinical parameters for risk stratification in medulloblastomas. J Clin Oncol 2004; 22: 971 – 4.
703
24. Gajjar A, et al. Clinical, histopathologic, and molecular markers of prognosis: toward a new disease risk stratification system for medulloblastoma. J Clin Oncol 2004; 22: 984 – 93. 25. Fisher PG, Burger PC, Eberhart CG. Biological risk stratification of medulloblastoma: The real time is now. J Clin Oncol 2004; 22: 971 – 4. 26. Biegel JA, et al. Germline and acquired mutations of INI1 in atypical and rhabdoid tumors. Cancer Res 1999; 59: 74 – 9. 27. Hilden JM, et al. Central nervous system atypical teratoid/rhabdoid tumor: results of therapy in children enrolled in a registry. J Clin Oncol 2004; 22: 2877 – 84. 28. Tekautz TM, et al. Atypical teratoid/rhabdoid tumors (ATRT): Improved survival in children 3 years of age and older with radiation and high-dose alkylator-based chemotherapy. J Clin Oncol 2005; 23: 1491 – 9. 29. Judkins AR, et al. Immunohistochemical analysis of hSNF%/INI1 in pediatric CNS neoplasms. Am J Surg Pathol 2004; 28: 644 – 50. 30. Bloom HJG. Medulloblastoma: prognosis and prospects. Int J Radiat Oncol Biol Phys 1977; 2: 1031 – 3. 31. Kun LE, Constantine LS. Medulloblastoma – caution regarding new treatment approaches. Int J Radiat Oncol Biol Phys 1991; 20: 897 – 9. 32. Grabenauer GG, et al. Postoperative radiotherapy of medulloblastoma. Impact of radiation quality on treatment outcome. Am J Clin Oncol 1996; 19: 73 – 7. 33. Thomas PR, et al. Low-stage medulloblastoma: final analysis of a trial comparing standard-dose with reduced dose neuraxis radiation. J Clin Oncol 2000; 18: 3004 – 11. 34. Tait DM, et al. Adjuvant chemotherapy for medulloblastoma: the first multi-centre control trial of the International Society of Pediatric Oncology (SIOP I). Eur J Cancer 1990; 26: 464 – 9. 35. Packer RJ, et al. Leptomeningeal dissemination of primary central nervous system tumors of childhood. Ann Neurol 1975; 18: 217 – 21. 36. Duffner PK, et al. Postoperative chemotherapy and delayed radiation in children less than three years of age with malignant brain tumors. N Engl J Med 1993; 328: 1725 – 31. 37. Finlay JL, et al. The ‘Head Start’ regimen for children less than six years of age with newly-diagnosed malignant brain tumors. Med Pediatr Oncol 1995; 25: 250. 38. Dunkel IJ, Finlay JL. High dose chemotherapy with autologous stem cell rescue for patients with medulloblastoma. J Neurooncol 1996; 29: 69 – 74. 39. Geyer JR, et al. Survival of infants with primitive neuroectodermal tumors or malignant ependymomas of the CNS treated with eight drugs in one day: a report from the Children’s Cancer Group. J Clin Oncol 1994; 12: 1607 – 15. 40. Gajjar A, et al. Medulloblastoma in very young children. Outcome of definitive craniospinal irradiation following incomplete response to chemotherapy. J Clin Oncol 1994; 12: 1212 – 6. 41. Dupuis-Girod S, et al. Will high dose chemotherapy followed by autologous bone marrow transplantation supplant craniospinal irradiation in young children treated for medulloblastoma? J Neurooncol 1996; 27: 87 – 98. 42. Tabori U, et al. Distinctive clinical course and pattern of relapse in adolescents with medulloblastoma. Int J Radiat Oncol Biol Phys 2006; 64: 402 – 7. 43. Packer RJ, et al. Outcome for children with medulloblastoma treated with radiation and cisplatin, CCNU, and vincristine chemotherapy. J Neurosurg 1994; 81: 690 – 8. 44. Packer RJ, et al. Treatment of medulloblastomas with reduced dose craniospinal radiation therapy and adjuvant chemotherapy: A Children’s Cancer Group Study. J Clin Oncol 1999; 17: 2127 – 36. 45. Oyharcabal-Bourden V, et al. Standard-risk medulloblastoma treated by adjuvant chemotherapy followed by reduced-dose craniospinal radiation therapy: A French Society of Pediatric Oncology Study. J Clin Oncol 2005; 23: 4726 – 34. 46. Jennings MT, et al. Differential responsiveness among “high risk” pediatric brain tumors in a pilot study of dose-intensive induction chemotherapy. Pediatr Blood Cancer 2004; 43: 46 – 54. 47. Kovnar EH, et al. Preirradiation cisplatin and etoposide in the treatment of high-risk medulloblastoma and other malignant embryonal tumors
704
48.
49.
50.
51.
52.
53. 54.
55.
56. 57.
58.
59.
60.
61. 62. 63. 64.
NEUROLOGICAL MALIGNANCIES
of the central nervous system. A phase II study. J Clin Oncol 1990; 8: 330 – 6. Heideman RL, et al. Preirradiation chemotherapy with carboplatin and etoposide in newly diagnosed embryonal pediatric CNS tumors. J Clin Oncol 1995; 13: 2247 – 54. Pendergass TW, et al. Eight drugs in one day chemotherapy for brain tumors: experience in 107 children and rationale for preradiation chemotherapy. J Clin Oncol 1987; 5: 1221 – 31. Bailey CC, et al. Prospective randomized trial of chemotherapy given before radiotherapy in childhood medulloblastoma. International Society of Paediatric Oncology (SIOP) and the (German) Society of Paediatric Oncology (GPO): SIOP II. Med Pediatr Oncol 1995; 25: 166 – 78. Stewart CF, et al. Results of a phase II upfront window of pharmacologically guided topotecan in high-risk medulloblastoma and supratentorial primitive neuroectodermal tumor. J Clin Oncol 2004; 22: 3357 – 65. Boyd T, et al. Feasibility and acute toxicities of craniospinal hyperfractionated radiotherapy (CHFRT) for high risk intracranial primitive neuroectodermal tumors: CCG – 9931; A groupwide phase II study of intensive chemotherapy and CHFRT. Int J Radiat Oncol Biol Phys 1997; 39(Suppl 1): 144. Attard-Montalto S, et al. Is there a danger in delaying radiotherapy in childhood medulloblastoma? Br J Radiol 1993; 66: 807 – 13. Brandes AA, Paris MK. Review of the prognostic factors in medulloblastoma of children and adults. Crit Rev Oncol Hematol 2004; 50: 121 – 8. Bloom HJG, Bessell EM. Medulloblastoma in adults: a review of 47 patients treated between 1952 and 1981. Int J Radiat Oncol Biol Phys 1990; 18: 763 – 72. Hazuka MB, et al. Survival results in adult patients treated for medulloblastoma. Cancer 1992; 69: 2143 – 8. Carrie C, et al. Multivariate analysis of prognostic factors in adult patients with medulloblastoma. Retrospective study of 156 patients. Cancer 1994; 74: 2352 – 60. Moody AM, Norman AR, Tait D. Paediatric tumors in the adult population: the experience of the Royal Marsden Hospital 1974 – 1990. Med Pediatr Oncol 1996; 26: 153 – 9. Breen SL, et al. A comparison of conventional, conformal and intensity-modulated coplanar radiotherapy plans for posterior fossa treatment. Br J Radiol Suppl 2004; 77: 768 – 74. Packer RJ, et al. Outcome for children with medulloblastoma treated with radiation, and Cisplatin, CCNU, and vincristine chemotherapy. J Neurosurg 1994; 81: 690 – 8. Haie C, et al. Results of radiation treatment of medulloblastoma in adults. Int J Radiat Oncol Biol Phys 1985; 11: 2051 – 6. Schell MJ, et al. Hearing loss in children and young adults with or without prior cranial irradiation. J Clin Oncol 1989; 7: 754 – 60. Galanis E, et al. Effective chemotherapy for advanced CNS embryonal tumors in adults. J Clin Oncol 1997; 15: 2939 – 44. Greenberg HS, et al. Adult medulloblastoma: multiagent chemotherapy. Neuro-oncol 2001; 3: 29 – 34.
65. Tabori U, et al. Medulloblastoma in the second decade of life: a specific group with respect to toxicity and management. Cancer 2005; 103: 1874 – 80. 66. Spreafico F, et al. Survival of adults treated for medulloblastoma using pediatric protocols. Eur J Can 2005; 41: 1304 – 10. 67. Moots PL, et al. Multiagent chemotherapy followed by craniospinal radiation for adults with poor risk medulloblastoma and ependymoma with subarachnoid dissemination. Neurology 1998; 50: A380. 68. Moots PL, et al. Toxicities associated with chemotherapy followed by craniospinal radiation for adults with poor-risk medulloblastoma/PNET and disseminated ependymoma: a preliminary report of ECOG 4397. Proc Am Soc Clin Oncol 2004; 125: A1573. 69. Brandes AA, et al. The treatment of adults with medulloblastoma: a prospective study. Int J Radiat Oncol Biol Phys 2003; 57: 755 – 61. 70. Saunders DE, et al. Surveillance neuroimaging of intracranial medulloblastoma: how effective, how often, and for how long? J Neurosurg 2003; 99: 280 – 6. 71. Herrlinger U, et al. Adult medulloblastoma: prognostic factors and response to therapy at diagnosis and at relapse. J Neurol 2005; 252: 291 – 9. 72. Dunkel IJ, et al. High-dose carboplatin, thiotepa, and etoposide with autologous stem-cell rescue for patients with recurrent medulloblastoma. Children’s Cancer Group. J Clin Oncol 1998; 16: 222 – 8. 73. Abrey LE, et al. High dose chemotherapy with autologous stem cell rescue in adults with malignant primary brain tumors. J Neuro-oncol 1999; 44: 147 – 53. 74. Huber JF, et al. Long-term effects of transient cerebellar mutism after cerebellar astrocytoma or medulloblastoma tumor resection in childhood. Childs Nerv Syst 2005; (Epub; in press). 75. Maddrey AM, et al. Neuropsychological performance and quality of life of 10 year survivors of childhood medulloblastoma. J Neuro-oncol 2005; 72: 245 – 53. 76. Mulhern RK, et al. Neurocognitive consequences of risk-adapted therapy for childhood medulloblastoma. J Clin Oncol 2005; 23: 5511 – 9. 77. Kramer JH, et al. Neuropsychological sequelae of medulloblastoma in adults. Int J Radiat Oncol Biol Phys 1997; 38: 21 – 6. 78. Goldstein AM, Yuen J, Tucker MA. Second cancers after medulloblastoma: population-based results from the United States and Sweden. Cancer Causes Control 1997; 8: 865 – 71. 79. Ashley DM, et al. Treatment of patients with pineoblastoma with high dose cyclophosphamide. Med Pediatr Oncol 1996; 26: 387 – 92. 80. Broniscer A, et al. High-dose chemotherapy with autologous stem cell rescue in the treatment of patients with recurrent non-cerebellar primitive neuroectodermal tumors. Pediatr Blood Cancer 2004; 42: 261 – 7. 81. Gururangan S, et al. High-dose chemotherapy with autologous stem cell rescue in children and adults with newly diagnosed pineoblastomas. J Clin Oncol 2003; 21: 2187 – 91.
Section 10 : Neurological Malignancies
64
Craniopharyngiomas Gene H. Barnett and John Park
HISTORICAL BACKGROUND Craniopharyngiomas (CPs) are slow-growing, histologically benign, sellar, and suprasellar tumors whose location and adherence to surrounding critical structures often result in a malignant course of endocrinological, behavioral, and visual abnormalities leading ultimately to premature death. They were first described in the late nineteenth century and postulated to arise from the hypophyseal duct or Rathke’s pouch,1 a theory that remains popular today. Complete microsurgical resection is the traditional goal of treatment and may lead to cure.2 – 5 However, due to their critical location and adherent nature, such an aggressive approach can be fraught with hazards and lead to debilitating outcomes. A multitude of therapeutic options have therefore been developed6 – 12 to manage these lesions, and controversy still exists with regard to optimal therapy. In this chapter, we will review the nature of these rare tumors, consider various management options, discuss their use alone and in combination, and make recommendations regarding contemporary management.
ANATOMY CPs are tumors of the sellar, suprasellar, and/or anterior third ventricle regions along the path of development of Rathke’s pouch.1,2,13 Most CPs arise near the infundibular stalk, in intimate connection to the hypothalamus and optic apparatus (see Figure 1), as well as the major arteries in the region such as the internal carotid arteries. While sellar tumors are supplied by small vessels arising from the adjacent dura of the cavernous sinus, suprasellar CPs are directly supplied by small branches of the arteries of the anterior diencephalon. These include the anterior cerebral artery, the anterior communicating artery, and the posterior communicating artery. They generally are not supplied by the posterior circulation (i.e. posterior cerebral arteries and vertebrobasilar distribution), a fact that has important surgical implications. Surgeons typically categorize CPs as sellar, prechiasmatic, and retrochiasmatic because of implications regarding surgical approach. Radiographically, about three quarters of CPs
are suprasellar, 20% are sellar and suprasellar, while only about 5% are entirely sellar in nature. They may also occur in the optic chiasm or third ventricle, and even in the nasopharynx, pineal regions, or sphenoid bone.14
BIOLOGY AND EPIDEMIOLOGY At the third to fourth week of gestation, the ectoderm of the stomodeum (roof of the oral cavity) invaginates toward the diencephalon to meet the downwardly projecting infundibular bud, thus forming the primitive craniopharyngeal duct. Rathke’s pouch is then formed as the development of the sphenoid bone pinches off the invagination from the pharyngeal epithelium. The pouch then surrounds the infundibulum and subsequently goes on to form the adenohypophysis. It is believed by many that embryonic rests of cells that would otherwise be destined to become tooth buds or oral epithelium may be deposited during this development along the craniopharyngeal duct and are responsible for the development of CPs. CPs are true cystic tumors and should not be confused with Rathke’s cleft cysts, although these may also have embryonic origins. Some adult tumors with squamous papillary pathology (see below) may arise by different mechanisms.2,4,5,13 CPs may be diagnosed at any age, although there is a bimodal distribution (see Figure 2)15 with peaks in childhood (aged 5–14 years) and later years (aged 50–74 years). The overall incidence is about 0.5 to 2.5 per 100 000 persons yearly and is independent of gender or race,15 although there is some data suggesting male predominance in childhood16 and an increased incidence in Japanese children. Approximately 338 cases of CPs are expected to occur annually in the United States, with 96 occurring in children from 0 to 14 years of age. They constitute about 5–10% of all childhood tumors but less than one percent of adult intracranial tumors.3 Of parasellar tumors, they represent 50% in children and 20% in adults. Although there have been a few reports of occurrences in siblings, there is no known genetic predisposition to CPs. Reports on the molecular characteristics of these tumors have been largely unrevealing to date; however, mutations in the
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
706
NEUROLOGICAL MALIGNANCIES
Figure 1 Craniopharyngioma in 46-year-old man (adamantinomatous). Note intimate relationship to hypothalamus and optic pathways.
Age-specific incidence CBTRUS 1990–1993
Incidence rate per 100 000
0.25
0.2
0.15
0.1
0.05
0 0–4
5–9 10–14 15–19 20–34 35–44 45–54 55–64 65–74 75–84 Age
Figure 2 Age distribution of craniopharyngioma in 135 cases as compiled by the Central Brain Tumor Registry of the United States (CBTRUS). (From Moore KD et al.Craniopharyngioma. In: Bernstein M and Berger MS. 2000. Neuro-Oncology: The Essentials, pp 409 – 418. Reprinted by permission of Thieme.)
β-catenin gene have been associated with adamantinomatous CPs.
PATHOLOGY CPs are typically comprised of a combination of solid portions with a benign histology and cystic components to a greater or lesser degree.2,13 Their size at diagnosis is variable, and it is not uncommon for tumors to be larger than 5 cm when discovered. They are World Health Organization grade I tumors indicating that they are potentially curable with complete surgical resection. The tumor elicits an intense glial
reaction from the adjacent hypothalamus, optic pathways, and infundibulum and may be particularly adherent to these and nearby vessels. These points of attachment often limit the role of surgery or contribute to postoperative morbidity. Two distinct histopathologies are found – the adamantinomatous and the papillary varieties. The former is found in childhood CPs and in about half of adult onset tumors, and is comprised of ribbons and bands of palisaded, columnar, or polygonal epithelial cells around the periphery of these structures and internally more loosely arranged with spongy architecture. The cysts may be lined with flattened epithelium (see Figure 3). This pattern resembles the adamantinomatous
CRANIOPHARYNGIOMAS
707
Figure 3 Histology of adamantinomatous craniopharyngioma. Left image shows typical palisading columnar squamous cells with spongy center. Right image shows flattened epithelium lining a cyst with calcification at bottom right.
pattern of tooth buds and leads to the initial postulates that the tumors are derived from Rathke’s duct. “Wet” keratin, calcifications, and cholesterol deposits are also common. The cysts are laden with oily brown, cholesterol-rich fluid, and microscopic tumor projections into the adjacent brain occur frequently. About half of adult tumors are of the papillary variety, which is composed of sheets of squamous epithelium forming pseudopapillae and papillae. Cysts are typically lined with simple squamous epithelium and lack the machine oil quality. Calcifications are rare. These tumors may derive from cell rests attempting to differentiate into oral mucosa or from squamous cells present in the adult sellar region.2 Labeling indices such as MIB-1 are typically <2% and of limited predictive value regarding behavior except when >7% where recurrence is considered likely. Spontaneous malignant transformation and metastases are said not to occur; however, seeding along biopsy or surgical tracts have been reported, as well as a report of distant leptomeningeal metastasis after surgical resection.17 Radiation-induced conversion to squamous carcinoma has also been reported.18
CLINICAL PRESENTATION AND DIAGNOSTIC CONSIDERATIONS Presenting symptoms of CP include visual, endocrinological, and behavioral disorders as well as headache.2 – 5 Adults are more likely to present with symptoms of visual impairment or endocrinological abnormalities than children. Visual disturbances are particularly prevalent in adults, including bitemporal hemianopsia, homonymous hemianopsia, or unilateral and bilateral decreased acuity or blindness. In one series, formal visual field testing showed visual compromise in 74.5% of patients at the time of diagnosis. Visual impairment was particularly noted in patients scheduled for transcranial surgery as opposed to those scheduled for transsphenoidal surgery (84.3 vs 42.9%). Papilledema is found in 29% of adults at presentation.
Endocrinological symptoms are the presenting complaints in about 30% of adults, although testing may reveal hormonal abnormalities in 80 to 90% of individuals. Gonadal insufficiency causes loss of libido and reduced masculine hair growth in adult males and secondary dysmenorrhea in women. Hyperprolactinemia occurs in 20% of patients due to compression of the hypothalamus or pituitary stalk and the decrease in the normally inhibitory effect on pituitary prolactin release. Diabetes insipidus is seen in 9 to 17% of patients prior to surgery. Other endocrinological abnormalities include hypothyroidism and adrenal insufficiency. Hypothalamic disturbances can include central hyperphagia with obesity, disturbances of thirst, and alterations in sleep cycles, but tend to affect adults less frequently than children. On the other hand, mental disturbance is more commonly seen in adults than in children, with more than 30% of patients older than 45 years of age suffering from dementia or intermittent confusion, apathy, depression, or psychomotor slowing. Increased intracranial pressure and obstructive hydrocephalus secondary to upward tumor growth into the third ventricle are seen more frequently in children than in adults. In one series, only 29% of adult patients had evidence of hydrocephalus at presentation, compared to over 50% in pediatric patients. In addition, headache (80 vs 30%), nausea or vomiting (60 vs 20%), and short stature (30 vs 15%) are more common in children than adults.2 The combination of progressive visual disturbance with endocrinological disorder should suggest a possible lesion in the sellar or suprasellar region such as CP.
Imaging Contemporary neuroimaging studies such as computerized tomography (CT) and magnetic resonance imaging (MRI) have greatly facilitated the radiographic diagnosis of these lesions compared to the era limited to pneumoencephalography, angiograms, and skull X rays where displacement of
708
NEUROLOGICAL MALIGNANCIES
Table 1 Radiographic differential diagnosis of craniopharyngiomas.
Lesion
Distinguishing feature(s)
Rathke’s cleft cyst
Pituitary adenoma
Arachnoid cyst Epidermoid/dermoid cysts
Pilocytic/infiltrating astrocytoma of chiasm or hypothalamus Xanthoastrocytoma Thrombosed aneurysm
Doesn’t enhance, no solid component, less heterogenous Confluent enhancement (unless hemorrhagic and cystic), sellar origin Doesn’t enhance, no calcifications Nonenhancing, low CT density, MRI diffusion changes Solid or microcystic, robust enhancement, not calcified, possible necrosis Adolescent age-group MRI flow changes, blood products
the ventricles and vessels, as well as suprasellar calcifications or sellar erosions often suggested the diagnosis of CP. The radiographic hallmark of CP is an enhancing solid/cystic lesion in the sellar, suprasellar, and/or anterior third ventricle. Adamantinomatous lesions are typically mixed solid/cystic and calcified on CT scan, whereas papillary lesions are often isointense, solid, and rarely show calcifications. Most all CPs enhance after administration of intravenous contrast. On MRI, the tumors are usually isointense with cystic components of various signal intensity on T1-weighted imaging. T2 weighting demonstrates high signal in the cysts, variable signal in the solid component, and high signal in the adjacent brain, which may indicate gliosis, edema from compression of the optic chiasm or tracts, tumor invasion, and irritation from leaking fluid.19 Magnetic resonance spectroscopy demonstrates high-lipid content in the cystic components. Conventional CT or MR angiography is of little benefit in the diagnosis other than for showing displaced and/or encased vessels. The radiographic differential diagnosis is shown in Table 1.
Medical and Endocrinological Evaluation These tests are not diagnostic of CP, per se, but when suggestive of pituitary and/or hypothalamic dysfunction, they can help localize the lesion and direct preoperative management. Common abnormalities include evidence of one or more insufficiencies of adenohypophyseal products (ACTH, TSH, LH, FSH, GH/IGF-1) and/or mild overproduction of prolactin due to disruption of physiologic infundibular inhibition of prolactin production. Patients may also exhibit diabetes insipidus due to posterior pituitary dysfunction and an inability to concentrate urine. The impact on preoperative management is discussed below.
TREATMENT Controversy continues to exist as to the best way to treat CPs. Multiple therapeutic strategies exist, which can
be combined to create a multimodality approach individualized to each patient. Aggressive and complete surgical resection would be ideal for achieving complete cure, but it must be weighed against the potential significant complications.
Pretreatment Management Prior to beginning therapy, an evaluation should be performed to address the spectrum of symptoms brought on by these lesions.2 – 5 The prevalence of visual impairment and the potential for damage to the optic apparatus make a full visual acuity and visual field examination mandatory. Extensive endocrinological workup is also necessary for documentation and correction of underlying hormonal deficiencies. If a patient is found to have pituitary dysfunction, steroid administration will be required perioperatively for those undergoing surgery. Hydrocephalus may require shunting for cerebrospinal fluid diversion, although normalization of ventricular size is often seen following efficacious treatment of the tumor.
Surgery The traditional management of these tumors is aggressive microsurgical resection. Numerous surgical approaches have been described for the resection of these lesions, detailed descriptions of which are beyond the scope of this chapter. Choice of an approach is dictated primarily by the location of the tumor.2 – 5 The subfrontal approach (below the frontal lobes) is the most commonly employed and allows good visualization of both optic nerves and the internal carotid artery. The pterional approach provides a shorter trajectory to the parasellar space along the Sylvian fissure and lateral sphenoid wing. The subtemporal approach is rarely used for these lesions, and is primarily for unilateral retrochiasmatic tumors. The transsphenoidal approach is ideal for tumors confined to the sella.8 Although the potential for cure is maximized by aggressive initial surgical resection, complete resection is not a guarantee for cure (so-called “false cures”) and often cannot be attained without considerable morbidity. Overall surgical mortality ranges between 0 and 4% in recent large surgical series, with a notable rate of 16.7% in a series by Yasargil et al. where aggressive total resection (ATR) was performed in all patients.20 Aggressive surgery also appears to increase the risk of serious morbidities including visual, other neurologic, behavioral, and endocrinologic (particularly diabetes insipidus).21 – 23 Location may also contribute to the morbidity of ATR as lesions in the third ventricle may also have a mortality of about 17%.24 Subtotal resection (STR) is often dictated by tumor adherence to vital neurovascular structures and the desire to minimize morbidity.25 In this setting, ATRs are achieved in about 50–60% of cases by the transcranial routes.26 More minimally invasive approaches have become increasingly popular in an effort to reduce surgical discomfort, morbidity, and recovery time. These include “keyhole” approaches through eyebrow incisions,27,28 endoscopic procedures for sellar29 and even suprasellar lesions,30 – 32
CRANIOPHARYNGIOMAS
709
stereotactic or endoscopic placement of drainage or delivery catheters for the management of CP cysts33 and stereotactic biopsy for the purposes of diagnosis and cyst drainage.34 There is a growing trend of combining minimally invasive stereotactic surgical procedures with drug or radiation (interstitial, fractionated radiotherapy, stereotactic radiosurgery (SRS)) treatments for a multimodality approach of maintaining lesion control at minimum risk.6,9,35
aspiration (to reduce the size of the tumor) or interstitial radiotherapy (to better control cysts).9,10,47 – 52 Some advocate using SRS only for sellar CPs.47 Regardless, tumor doses should be at least 6 if not 9.5 Gy10,52 with dose to the optic nerves and chiasm no more than 8–10 Gy (the less the better). Risk of postoperative visual or endocrinological deficits are typically less than 4% with extended (12-year) tumor control of about 85% when adhering to these parameters.10,52
Radiotherapy
Chemotherapy
Radiation for CPs is traditionally delivered via external fractionated delivery in total doses ranging from 50 to 60 Gy.11,36,37 Such radiotherapy is rarely used alone for the treatment of these lesions38 but is used most commonly after STR at the time of tumor recurrence39 with results that are typically superior to reoperation. There appears to be no clear benefit in terms of local control or survival for using radiation after gross total resection; however, there is good evidence to support its use after STR with 5and 10-year survivals of about 70 and 65%, respectively in one series.11 Less clear is the timing of treatment after STR – empirically if incomplete, or waiting until progression/recurrence. There appears to be no difference in survival between either of these approaches, so many advocate the latter delayed approach to minimize the risk of late radiation toxicities such as cognitive impairment, malignant transformation,18 or induction of benign40 or malignant secondary tumors.41 An advance in the external delivery of fractionated radiation is the use of stereotactic techniques – so-called stereotactic radiotherapy which allows more conformal delivery, limiting collateral damage to surrounding structures.42 Even with control of the solid tumor by surgery or radiotherapy, growth of existing or development of new cysts can lead to progressive neurologic dysfunction. Intracavitary use of radioactive solutions delivered using stereotactically directed catheters and, at times, subcutaneous Ommaya reservoirs, appears to be an important tool in the management of tumor cysts.9,12,34,43 – 46 Such therapeutic agents include phosphorous-32, yttrium-90, iridium-192, gold-198 colloid, and iodine-125.
The principal uses of chemotherapy for CPs is for control of tumoral cysts or for surgical/radiation failures of the lesion as a whole. Intratumoral bleomycin (typically intracystic delivery, 3 mg every other day) may help control otherwise intractable cyst expansion.53,54 Extracavitary extravasation of the drug, however, can cause blindness or even death, tempering enthusiasm for this approach.53,55 Systemic chemotherapy may be of benefit in some otherwise treatment refractory cases of CP. Reported successes have been achieved with a combination of vincristine, BCNU and procarbazine,7 doxorubicin and lomustine,56 and a cisplatin derived compound.57 The identification of estrogenreceptors in some of these tumors has not yet translated into successful treatments exploiting this finding.58
Radiosurgery More controversial than the issues pertaining to radiotherapy and CPs is the use of SRS in the management of new or recurrent CPs. SRS is a combined surgical/radiation oncology technique where high-dose radiation is delivered to a small volume in a single session with the intention of ablating or inactivating the target using the precision of image-guided stereotactic techniques to direct this energy and sparing nearby structures. There are certain radiobiologic advantages of SRS over fractionated radiotherapy for benign lesions;9,32 however, there is also concern about possible injury to the optic nerves and hypothalamus for certain CPs. Nonetheless, there is a blossoming literature supporting the use of SRS for newly diagnosed, incompletely resected, or recurrent tumors. SRS may be used alone or in combination with other techniques such as preoperative cyst
RECOMMENDATIONS Because of the heterogeneous locations, histologies, and age populations, no single strategy is appropriate for all CPs. Some generalities are probably valid, despite the lack of any class I evidence in this disorder. Patients with sellar lesions can be well managed with transsphenoidal microsurgery or endoscopic surgery, or high precision SRS (such as with the Gamma Knife, Elekta Medical Instruments, Stockholm, Sweden). Microsurgical resection (tempered with the fallback of STR if too adherent to critical structures) should be considered for suprasellar lesions, with or without endoscopic techniques. Third ventricular tumors, particularly if they show adjacent brain changes, may best be approached by a multimodality approach of stereotactic aspiration, high precision SRS, and intracavitary treatment for progressive cysts. Radiotherapy should probably be reserved for progressive or progressive subtotally resected lesions. It is likely that the role of multimodality minimally invasive stereotactic techniques will become even more prevalent as patients demand lower risk, less invasive approaches to management of intracranial disease.
PROGNOSIS In a review by Buddin et al. of the National Cancer Data Base (NCDB) data on 285 CP patients, the overall 5-year survival rate was 80%. However, age was a significant negative prognostic factor with 5-year survivals of 99, 79, and 37% for individuals diagnosed at ages of less than 20, 20–64, and 65 years or older, respectively. Survival in children appears to have improved over time.
710
NEUROLOGICAL MALIGNANCIES
REFERENCES 1. Mott FW, Barrett JOW. Three cases of tumor of the third ventricle. Arch Neurol (Lond) 1989; 11: 417 – 40. 2. Carmel PW. Craniopharyngiomas. In Wilkins RH, Rengachary SS (eds) Neurosurgery. New York: McGraw-Hill, 1985: 905 – 916. 3. Matson DD, Crigler JF Jr. Management of craniopharyngiomas in childhood. J Neurosurg 1969; 30: 377 – 90. 4. Mehta V, Black PM. Craniopharyngioma in the Adult. In Winn RH (ed) Youmans Neurological Surgery. 5th ed. W.B. Saunders, 2003: 1207 – 1221. 5. Moore KD, Couldwell WT. Craniopharyngioma. In Bernstein M, Berger MS (eds) Neuro-Oncology: The Essentials. New York: Thieme Medical Publishers, 2000: 409 – 418. 6. Barajas MA, et al. Multimodal management of craniopharyngiomas: neuroendoscopy, microsurgery, and radiosurgery. J Neurosurg 2002; 97(Suppl 5): 607 – 9. 7. Bremer AM, Nguyen TQ, Balsys R. Therapeutic benefits of combination chemotherapy with vincristine, BCNU and procarbazine on recurrent, cystic craniopharyngioma. A case report. J Neurooncol 1984; 2: 47 – 51. 8. Cappabianca P, et al. Endoscopic endonasal transsphenoidal approach: outcome analysis of 100 consecutive procedures. Minim Invasive Neurosurg 2002; 45: 193 – 200. 9. Coffey RJ, Lunsford LD. The role of stereotactic techniques in the management of craniopharyngiomas. Neurosurg Clin N Am 1990; 1: 161 – 72. 10. Chung WY, et al. Gamma Knife radiosurgery for craniopharyngiomas. J Neurosurg 2000; 93(Suppl 3): 47 – 56. 11. Danoff BF, Cowchock FS, Kramer S. Childhood craniopharyngioma: survival, local control, endocrine and neurologic function following radiotherapy. Int J Radiat Oncol Biol Phys 1983; 9: 171 – 175. 12. Hasegawa T, et al. Management of cystic craniopharyngiomas with phosphorus-32 intracavitary irradiation. Neurosurgery 2004; 54: 813 – 822. 13. Janzer RC. Craniopharyngioma. In Kleihues P, Cavenee WK (eds) World Health Organization Classification of Tumours: Pathology & Genetics of Tumors of the Nervous System. Lyon, France: IQRC Press, 2000: 244 – 246. 14. Fujimoto Y, et al. Craniopharyngioma involving the infrasellar region: a case report and review of the literature. Pediatr Neurosurg 2002; 37: 210 – 216. 15. Bunin GR, et al. The descriptive epidemiology of craniopharyngioma. J Neurosurg 1998; 89: 547 – 551. 16. Stewart AM, Lennox EL, Sanders BM. Group characteristics of children with cerebral and spinal cord tumors. Br J Cancer 1973; 28: 568 – 574. 17. Gupta K, et al. Metastatic craniopharyngioma. AJNR Am J Neuroradiol 1999; 20: 1059 – 60. 18. Thomas C, et al. Malignant craniopharyngioma. J Neuropathol Exp Neurol 1999; 58: 567. 19. Saeki N, et al. MR imaging study of edema-like change along the optic tract in patients with pituitary region tumors. AJNR Am J Neuroradiol 2003; 24: 336 – 42. 20. Yasargil MG, et al. Total removal of craniopharyngiomas: approaches and long-term results in 144 patients. J Neurosurg 1990; 73: 3 – 11. 21. Bin-Abbas B, et al. Endocrine sequelae of childhood craniopharyngioma. J Pediatr Endocrinol Metab 2001; 14: 869 – 74. 22. Honegger J, Buchfelder M, Fahlbusch R. Surgical treatment of craniopharyngiomas: endocrinological results. J Neurosurg 1999; 90: 251 – 7. 23. Merchant TE, et al. Craniopharyngioma: the St. Jude Children’s Research Hospital experience 1984 – 2001. Int J Radiat Oncol Biol Phys 2002; 53: 533 – 42. 24. Behari S, et al. Intrinsic third ventricular craniopharyngiomas: report on six cases and a review of the literature. Surg Neurol 2003; 60: 245 – 52. 25. Fahlbusch R, et al. Surgical treatment of craniopharyngiomas: experience with 168 patients. J Neurosurg 1999; 90: 237 – 50. 26. Van Effenterre R, Boch AL. Craniopharyngioma in adults and children: a study of 122 surgical cases. J Neurosurg 2002; 97: 3 – 11.
27. Jho HD. Orbital roof craniotomy via an eyebrow incision: a simplified anterior skull base approach. Minim Invasive Neurosurg 1997; 40: 91 – 7. 28. Wiedemayer H, et al. The supraorbital keyhole approach via an eyebrow incision for resection of tumors around the sella and the anterior skull base. Minim Invasive Neurosurg 2004; 47: 221 – 5. 29. Laws ER, Laws ER Jr. Transsphenoidal microsurgery in the management of craniopharyngioma. J Neurosurg 1980; 52: 661 – 6. 30. Jho HD, Carrau RL. Endoscopic endonasal transsphenoidal surgery: experience with 50 patients. J Neurosurg 1997; 87: 44 – 51. 31. Locatelli D, et al. Endoscopic approach for the treatment of relapses in cystic craniopharyngiomas. Childs Nerv Syst 2004; 20: 863 – 7. 32. Lunsford LD, et al. Stereotactic options in the management of craniopharyngioma. Pediatr Neurosurg 1994; 21(Suppl 1): 90 – 7. 33. Joki T, et al. Neuroendoscopic placement of Ommaya reservoir into a cystic craniopharyngioma. Childs Nerv Syst 2002; 18: 629 – 33. 34. Mundinger F, et al. Stereotactic treatment of brain lesions. Biopsy, interstitial radiotherapy (iridium-192 and iodine-125) and drainage procedures. Appl Neurophysiol 1980; 43: 198 – 204. 35. Nicolato A, et al. Multimodality stereotactic approach to the treatment of cystic craniopharyngiomas. Minim Invasive Neurosurg 2004; 47: 32 – 40. 36. Gurkaynak M, et al. Results of radiotherapy in craniopharyngiomas analysed by the linear quadratic model. Acta Oncol 1994; 33: 941 – 3. 37. Jephcott CR, Sugden EM, Foord T. Radiotherapy for craniopharyngioma in children: a national audit. Clin Oncol (R Coll Radiol) 2003; 15: 10 – 3. 38. Honegger J, et al. Regression of a large solid papillary carniopharyngioma following fractionated external radiotherapy. J Neurooncol 1999; 41: 261 – 6. 39. Jose CC, et al. Radiotherapy for the treatment of recurrent craniopharyngioma. Clin Oncol (R Coll Radiol) 1992; 4: 287 – 9. 40. Kano T, et al. A juvenile case of radiation-induced meningioma two years after radiation for craniopharyngioma. No Shinkei Geka 1994; 22: 367 – 70. 41. Ushio Y, et al. Glioblastoma after radiotherapy for craniopharyngioma: case report. Neurosurgery 1987; 21: 33 – 38. 42. Schulz-Ertner D, et al. Fractionated stereotactic radiotherapy for craniopharyngiomas. Int J Radiat Oncol Biol Phys 2002; 54: 1114 – 20. 43. Huk WJ, Mahlstedt J. Intracystic radiotherapy (90Y) of craniopharyngiomas: CT-guided stereotaxic implantation of indwelling drainage system. AJNR Am J Neuroradiol 1983; 4: 803 – 6. 44. Kobayashi T, Kageyama N, Ohara KT. Internal irradiation for cystic craniopharyngioma. J Neurosurg 1981; 55: 896 – 903. 45. Schefter JK, et al. The utility of external beam radiation and intracystic 32P radiation in the treatment of craniopharyngiomas. J Neurooncol 2002; 56: 69 – 78. 46. Yu X, Liu Z, Li S. Combined treatment with stereotactic intracavitary irradiation and gamma knife surgery for craniopharyngiomas. Stereotact Funct Neurosurg 2000; 75: 117 – 22. 47. Jackson AS, et al. Stereotactic radiosurgery. XVII: recurrent intrasellar craniopharyngioma. Br J Neurosurg 2003; 17: 138 – 43. 48. Laws ER, Vance ML, Laws ER Jr. Radiosurgery for pituitary tumors and craniopharyngiomas. Neurosurg Clin N Am 1999; 10: 327 – 36. 49. Mokry M. Craniopharyngiomas: a six year experience with Gamma Knife radiosurgery. Stereotact Funct Neurosurg 1999; 72(Suppl 1): 140 – 9. 50. Prasad D, Steiner M, Steiner L. Gamma Knife surgery for craniopharyngioma. Acta Neurochir (Wien) 1995; 134: 167 – 76. 51. Suh J, Barnett GH. Stereotactic radiosurgery for brain tumors in pediatric patients. Technol Cancer Res Treat 2003; 2: 141 – 6. 52. Ulfarsson E, et al. Gamma knife radiosurgery for craniopharyngiomas: long-term results in the first Swedish patients. J Neurosurg 2002; 97: 613 – 22. 53. Mottolese C, et al. Intracystic chemotherapy with bleomycin in the treatment of craniopharyngioma. Childs Nerv Syst 2001; 17: 724 – 30. 54. Savas A, et al. Intracavitary chemotherapy of polycystic craniopharyngioma with bleomycin. Acta Neurochir (Wien) 1999; 141: 547 – 8. 55. Savas A, et al. Fatal toxic effect of bleomycin on brain tissue after intracystic chemotherapy for a craniopharyngioma: case report. Neurosurgery 2000; 46: 213 – 6.
CRANIOPHARYNGIOMAS 56. Lippens RJ, et al. Chemotherapy with Adriamycin (doxorubicin) and CCNU (lomustine) in four children with recurrent craniopharyngioma. Eur J Paediatr Neurol 1998; 2: 263 – 8. 57. Plowman PN, et al. Dramatic response of malignant craniopharyngioma to cis-platin-based chemotherapy. Should craniopharyngioma be
711
considered as a suprasellar ‘germ cell’ tumor? Br J Neurosurg 2004; 18: 500 – 5. 58. Izumoto S, et al. Immunohistochemical detection of female sex hormone receptors in craniopharyngiomas: correlation with clinical and histologic features. Surg Neurol 2005; 63: 520 – 5.
Section 10 : Neurological Malignancies
65
Ophthalmic Cancers
Arun D. Singh, William J. Dupps, Jr and Sophie Bakri
INTRODUCTION Ophthalmic tumors, although uncommon, arise from various ophthalmic and adnexal structures. In general, they can be anatomically classified as eyelid tumors, conjunctival/corneal tumors, uveal tumors, retinal tumors, and orbital/adnexal tumors (see Table 1). Ophthalmic tumors, similar to neoplasia elsewhere, can be benign, premalignant, and malignant with presentation both in children (retinoblastoma) and adults (uveal melanoma). Symptomatology and clinical presentation are varied depending upon the anatomical location of the tumor. In general, treatment involves surgery, radiation, and chemotherapy. In addition, specialized ophthalmic procedures such as thermotherapy and brachytherapy are used for local ophthalmic treatment. In this chapter, we limit our discussion to general aspects of the most common types of ophthalmic cancers from each region: eyelid (sebaceous carcinoma), conjunctiva (squamous cell carcinoma [SCC]), uvea (melanoma), and retina (retinoblastoma).
SEBACEOUS CARCINOMA OF THE EYELID Historical Background Current understanding of sebaceous carcinoma was initiated by a review by Straatsma in 1956, which established the origin of this neoplasm.1 Historical developments are summarized elsewhere.2 The terms sebaceous gland carcinoma, sebaceous cell carcinoma, and sebaceous carcinoma are interchangeably used in the literature.3
Anatomy The tarsal plate of the eyelid offers structural support to the eyelid. It is a fibrous tissue, which contains meibomian glands within its substance. Meibomian glands are modified sebaceous glands contributing to the oily layer of the tear film. Zeis glands, which are also sebaceous in nature, are located closer to the eyelid margin in association with eyelashes.
Biology and Epidemiology Sebaceous carcinoma can occur in a variety of sites but the ocular and orbital involvement accounts for about 75% of
all cases. Overall, sebaceous carcinoma represents about 1% of all eyelid tumors with an incidence of 0.5 per million white population in the United States.4 However, sebaceous carcinoma represents a greater proportion of eyelid tumors (28–33%) in the Asian population.5,6 The etiology of sebaceous carcinoma remains obscure. In a small minority of cases, genetic predisposition such as seen in retinoblastoma and Muir-Torres syndrome may play a contributory role.7 – 9 External beam radiotherapy, especially in children, is a known risk factor.10 Immunosuppression as seen in human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) may also be associated with sebaceous carcinoma.11
Pathology Four histopathologic variants are known: lobular, comedocarcinoma, papillary, and mixed, with variable amount of differentiation. The lobular pattern is most common, mimicking the structure of a normal sebaceous gland.12 The sebaceous carcinoma contains lipids, which is readily demonstrated by oil-red-O stain. Immunohistochemically, the sebaceous carcinoma expresses human milk fat globule-1 and epithelial membrane antigen but not cytokeratins.13 Sebaceous carcinoma tends to spread locally to the lacrimal gland and to the draining (preauricular and submandibular) lymph nodes. The risk of lymph node spread may be lower now a days as compared to the historic reports of 30%.14 – 16 Distant hematogenous metastasis to the lung, liver, bone, and brain is also becoming uncommon.16
Clinical Presentation and Diagnostic Considerations Because of its rarity, sebaceous carcinoma is often misdiagnosed both clinically and histopathologically. It typically affects older females. In the periocular region, the neoplasm arises from meibomian glands, Zeis glands, caruncle, conjunctiva, and eyebrows.17 Sebaceous carcinoma can appear as a solitary nodule when arising from the Zeis glands or present as diffuse thickening of the eyelid when arising from meibomian glands (see Figure 1). The upper eyelid (63%) is more often involved than the lower eyelid (27%) with involvement of both eyelids (5%) in a minority of cases.2
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
OPHTHALMIC CANCERS Table 1 Common ophthalmic tumors classified by anatomical layers.
Structure
Benign
Malignant
Eyelid
Papilloma Nevus Hemangioma Nevus Retinocytoma Nevus Hemangioma Dermoid cysts Hemangioma Lymphangioma
Basal cell carcinoma Squamous cell carcinoma Meibomian gland carcinoma Squamous cell carcinoma Retinoblastoma Uveal melanoma Uveal metastasis Rhabdomyosarcoma Lymphoma Orbital metastasis
Conjunctiva/cornea Retina Uvea Orbital/adnexal
713
Sentinel lymph node biopsy offers the possibility of early detection of regional spread.25
Authors’ Recommendations Excision with mapping biopsy as outlined above is generally performed. In the presence of extensive orbital involvement, exenteration is recommended.
Prognosis The prognosis with sebaceous carcinoma is rather poor because of delayed diagnosis, its tendency for multicentric origin, and the risk of regional/distant spread. Nevertheless, it appears that survival with sebaceous carcinoma may be improving as recent studies have indicated mortality rates of less than 10%.14 – 16
CONJUNCTIVAL SQUAMOUS CELL CARCINOMA Historical Background SCC of the conjunctiva was described as early as 1860 by von Graefe.26 He used the term “epithelioma” to describe gelatinous elevations of the conjunctiva typified by variable amounts of vascularity and keratinization. Although SCC of the conjunctiva is often considered a low-grade carcinoma, timely diagnosis and appropriate management is essential to prevent visual disability, disfigurement, and life-threatening invasive disease.
Anatomy Figure 1 Photograph showing a typical external appearance of sebaceous carcinoma.
The localized form is often misdiagnosed as chalazion and the diffuse form as blepharitis.2 A high index of suspicion is a must for early diagnosis. Excisional or incisional biopsy must include full thickness of the eyelid so that the tarsus can be examined histopathologically. Because of its tendency to show pagetoid spread along the conjunctival epithelium and have multicentric origin, biopsies from multiple sites, especially of the conjunctiva (mapping biopsy), may be necessary to establish the full extent of the disease.18
Treatment Surgical excision with about 5 mm of clear margins is considered the standard of care.19 Clear margins may be obtained either with frozen section control or Mohs technique with comparable results.19 Mapping biopsy of the conjunctiva is also indicated to determine the extent of the disease.18 Supplemental cryotherapy is advocated, but its benefit remains to be established.20 Topical mitomycin therapy is useful for treatment of conjunctival involvement.21 Radiotherapy offers only palliative treatment and is recommended when a patient declines or is unfit for surgery.19,22,23 Exenteration is performed when there is extensive orbital extension.19,24
The conjunctiva is a membranous lining composed of nonkeratinized squamous epithelium and columnar epithelium with a deeper substantia propria, which extends from the eyelid margin to the corneal limbus.27 SCC is considered within the clinical spectrum of ocular surface squamous neoplasia (OSSN), which includes all epithelial tumors, whether dysplastic or carcinomatous, that affect the conjunctiva or the cornea.28
Biology and Epidemiology Although SCC of the conjunctiva is an uncommon ophthalmic tumor, it is the most common malignancy of the conjunctiva in the United States.29 Incidence ranges from 0.03 to 2.8 per 100 000 people per year.30,31 SCC is most commonly observed in older male Caucasians (mean 56 years).28 At the limbus, a transition from the columnar conjunctival epithelium of the fornices to the stratified squamous epithelium of the cornea occurs. This region is biologically akin to other transitional tissues such as the uterine cervix that are also prone to dysplasia.32 Most squamous tumors originate from the interpalpebral limbus and involve both the conjunctiva and the cornea. Ultraviolet-B (UV-B) light exposure has consistently emerged as a major etiologic factor.28 Chronic environmental exposure to wind and dust, petroleum, and cigarette smoke are also believed to be significant risk factors. Human papilloma virus (subtypes 16 and 18) has been associated with
714
NEUROLOGICAL MALIGNANCIES
SCC but is probably not an independent cause of conjunctival neoplasia.33 The mechanism of squamous transformation after UV-B radiation may be related to point mutations of the p53 tumor suppressor gene.34 Patients with xeroderma pigmentosa have a much higher incidence and account for most SCC in young patients.32 Systemic immunosuppression, as seen in AIDS may increase the risk of developing OSSN by as much as 13-folds in patients with AIDS.35
Pathology Conjunctival epithelial neoplasia is broadly divided into conjunctival intraepithelial neoplasia (CIN) and invasive SCC. Lesions that are confined to the epithelium are preferentially termed CIN32 or corneal intraepithelial neoplasia,36 respectively. Tumors that infiltrate the substantia propria are referred to as invasive SCC of the conjunctiva or cornea.37 Within the heterogeneous category of CIN, dysplasia may range from mild (less than one-third thickness occupied by atypical cells) to severe (near full-thickness involvement), and full-thickness obliteration of normal cells without penetration of the basement membrane is referred to as carcinoma in situ. The predominant cell type is the spindle variety, characterized by small elliptical cells without prominent nucleoli, a weakly basophilic cytoplasm, and frequent mitotic figures. The epidermoid variety with larger polyhedral cells is less frequent (5%).28 Mucoepidermal carcinoma is a rare variant characterized by mucous-filled cysts and a greater propensity to local invasion.38
Clinical Presentation and Diagnostic Considerations Unilateral red eye and ocular irritation are the most common presenting symptoms, and diffuse tumors can be misdiagnosed as chronic conjunctivitis. On examination, nodular thickening of the interpalpebral limbal conjunctiva with a gelatinous, papillomatous, or leukoplakic appearance is typically observed (see Figure 2).39 The lesion may mimic or even coexist with benign conjunctival degenerations such as pingueculae and pterygia.29 Clinical differentiation of CIN from invasive SCC of the conjunctiva or cornea is unreliable.40 Rose bengal stain may be useful in delineating abnormalities of the mucin layer associated with OSSN. While features such as size and extensive involvement of the limbal circumference have been associated with invasive disease,41 the diagnosis is ultimately histopathological and relies on careful evaluation for violation of the epithelial basement membrane. Nonsurgical ascertainment of OSSN may be enhanced with exfoliative or impression cytology, but the ability to distinguish carcinoma in situ from minimally invasive SCC is limited.39
Treatment Management is influenced primarily by the extent of the lesion. Surgical excision with 2 to 3 mm margins has been the preferred approach for OSSN involving less than 4 clock hours of limbal conjunctiva.42 Alcohol epitheliectomy of any corneal component is generally followed by removal of the conjunctival component via lamellar scleral resection.
Figure 2 Slit lamp photograph of the left eye. Note nodular thickening of the interpalpebral limbal conjunctiva with gelatinous, papillomatous, and leukoplakic changes.
Application of intraoperative supplemental cryotherapy (double freeze –thaw cycles) to conjunctival margins reduces the risk of tumor recurrence.43 Mohs technique has also been described in the management of conjunctival SCC.44 Surgical management is more complicated with larger lesions, where attempts at total surgical excision carries a risk of limbal stem cell depletion, corneal scarring, and visual loss. Extensive excisions for locally invasive disease can require limited or en masse removal of affected ocular and orbital tissues followed by reconstruction with conjunctival autografts, amniotic membrane grafts, or keratolimbal stem cell grafts. Topical chemotherapy has been advocated both intraoperatively and postoperatively as an adjuvant for incomplete excision of large and diffuse primary or recurrent tumors.45,46 More recently, topical therapy has been explored as an alternative to excision in selected cases.47 It has also been reported to be effective as a neoadjuvant therapy.48 While mitomycin C (MCC) is the best studied of the chemotherapeutic agents (see Table 2), other topical agents including 5-fluorouracil47 and interferon α-2b49 have been investigated. Protocols for topical therapy typically involve temporary punctual occlusion followed by alternating weeks of 0.04% MCC drops four times daily and rest until resolution occurs.45 A reversible chemical keratoconjunctivitis is common with all agents, although interferon α-2b produces less epithelial toxicity in vitro.50 Radiation is rarely indicated.28
Authors’ Recommendations In general, complete excision, where feasible, should be attempted. Supplemental cryotherapy is always performed. In patients with larger, diffuse, or recurrent involvement, incisional biopsy followed by topical MMC therapy is effective.51
Prognosis Most CIN lesions do not progress to invasive SCC.41 When conjunctival SCC does occur, it is rarely metastatic, but the risk of regional and distant metastasis is much greater in the
OPHTHALMIC CANCERS
715
Table 2 Published reports of treatment of conjunctival and corneal intraepithelial neoplasia with mitomycin.a
Mitomycin First author
Year
Method of treatment
Concentration (%)
Frequency (per day)
Duration
Response rate (%)
Frucht-Pery Frucht-Pery Heigle Wilson Akpek Rozenman Frucht-Pery Khokar Siganos
1994 1997 1997 1997 1999 2000 2002 2002 2002
Primary Primary Primary Primary Postoperative Primary Postoperative Intraoperative Intraoperative
0.02 0.02 0.02 – 0.04 0.04 0.02 0.02 – 0.04 0.02 – 0.04 0.02 0.02
4 4 3–6 4 3 4 4 – –
10 – 22 days 7 – 28 days 3 – 5 cyclesb 1 – 5 cyclesc 14 days 14 – 28 days 3 cyclesc 5 minutes 5 minutes
100 13/17 100 6 0f 7 100 7/8 100 100 7 of 8
a b c
Excluding single case reports. One cycle = 1 week of treatment followed by 1 week without treatment. One cycle = 2 weeks of treatment followed by 4 weeks without treatment.
face of immunosuppression and HIV infection.37 Recurrence occurs at rates between 16 and 52% in long-term studies and is predicted by larger lesion size, positive surgical margins, lack of supplemental cryotherapy, increased patient age, and elevated proliferation index by Ki67 immunostaining.39,41,43 Currently, recurrence rates of 5–10% are expected with a median time to recurrence of about 18 months.51 Intraocular (13%) or orbital (11%) involvement requiring enucleation or orbital exenteration is uncommon.52 Fortunately, metastatic conjunctival SCC is extremely rare. The most important factor in cases resulting in visual loss or death is a delay in diagnosis.28
UVEAL MELANOMA Historical Background Until the 1980s, the optimal management for uveal melanomas, small, medium, or large was debatable53,54 because the majority of data was derived from retrospective studies based only on a small number of patients treated at a given center.55,56 Moreover, the benefit of enucleation was being questioned.57,58 In 1984, funded by the National Eye Institute (Bethesda, Maryland), Collaborative Ocular Melanoma Study (COMS) trials were designed. Over the last 20 years, a wealth of reliable data derived from COMS and other studies, has offered clearer guidelines for the management of patients with uveal melanoma.
Anatomy The uvea is the middle coat of the eyeball with the sclera external to it and the retina being the innermost layer. The uvea extends anteriorly from the pupil to the optic disc posteriorly. The uvea is divided into three parts: iris, ciliary body, and choroid (from anterior to posterior). The uvea is a highly vascular layer with abundant melanocytes of neuroectodermal origin.
Biology and Epidemiology Uveal melanoma is the most common primary intraocular malignant tumor.59 Even so, melanomas of the ocular
and adnexal structures comprise approximately 5% of all melanomas.60 The majority (85%) of ocular melanomas are uveal in origin whereas primary eyelid, conjunctival, and orbital melanomas are very rare.60,61 Uveal melanoma is a primary malignant tumor of the uvea arising from uveal melanocytes. On the basis of anatomic location, it is classified into three types: iris melanoma, ciliary body melanoma, and choroidal melanoma. Although uveal melanocytes and cutaneous melanocytes share common embryologic origin and some morphologic properties, several significant differences exist between cutaneous and uveal melanomas in terms of the prognostic factors, site of distant metastasis, and response of metastatic disease to chemotherapy. Cell type, an important prognostic factor in uveal melanoma, is of minor significance in cutaneous melanoma.62 Cutaneous melanoma tends to show nonvisceral metastasis involving skin, subcutaneous tissues, and distant lymph nodes more often than visceral involvement. In contrast, uveal melanoma predominantly metastasizes to liver. The chemotherapy regimen currently used in the treatment of metastatic cutaneous melanoma is ineffective against metastatic uveal melanoma.63 Incidence
The global incidence of uveal melanoma ranges from 5.3 to 10.9 cases per million.61 The incidence of uveal melanoma in the United States and European countries is similar to that in Australia64 and New Zealand,65 where the population is exposed to a higher intensity of ultraviolet light. The overall mean incidence of uveal melanoma is 4.3 per million, with a higher rate in males (4.9 per million) as compared to females (3.7 per million).59 Unlike trends for cutaneous melanoma, the incidence of uveal melanoma has remained stable in the United States for the last 50 years.59,66 Host Factors
Among the host factors, race seems to be the most significant, as uveal melanoma is about 150 times more common in whites than in blacks, and is less common in Asians.61,67 Light skin color, blond hair, and blue eyes are specific host risk factors. Genetic factors such as family history and other
716
NEUROLOGICAL MALIGNANCIES
syndromic associations, except oculo(dermal) melanocytosis, play only a minor role in predisposition to uveal melanoma.68 Environmental Factors
The evidence with regards to the contributory role of sunlight exposure in the etiopathogenesis of uveal melanoma is at best weak and contradictory.61,67 In addition, there is no consistent evidence indicating a specific occupational exposure to ultraviolet light or a chemical agent as a risk factor for uveal melanoma.61,67
Pathology Uveal melanoma is comprised of melanocytes with variable amounts of intrinsic vascularity. Areas of hemorrhage, necrosis, and lymphocytic infiltration are also observed. The tumors are composed of spindle cells, epithelioid cells, or most commonly a mixture of both types of cells (mixed cell type).69
(a)
Clinical Presentation Iris melanoma presents as an iris pigmented mass. A ciliary body melanoma can cause blurred vision due to lenticular astigmatism. Loss of vision, flashing light sensation, and exudative retinal detachment are some of the common presentations of a choroidal melanoma. The diagnosis is essentially clinical based on indirect ophthalmoscopy, angiographic studies, and the ultrasonographic pattern (see Figure 3). The diagnostic accuracy with such techniques is 99% and therefore, biopsy is performed only in very atypical cases.
Treatment The treatment of uveal melanoma is surgical or radiation therapy. Visual acuity, potential for retaining vision, vision in the unaffected eye, tumor size and location, and patient’s preference are important factors when considering the treatment options. For smaller tumors surgical resection (iridectomy, cyclectomy, choroidectomy) is performed. Larger tumors and those associated with significant loss of vision or those that have no potential for maintaining vision are managed by enucleation. Medium-sized tumors can be treated either with proton beam radiotherapy, brachytherapy (Iodine-125, Ruthenium 106), or with enucleation. Smaller choroidal tumors are also amenable to transpupillary thermotherapy (TTT). The COMS divided choroidal melanoma according to size into small, medium, and large tumors on the basis of the largest basal diameter (LBD) and height (see Table 3).70 COMS comprises of two randomized trials, one each for medium71 and large tumors,72 and one observational study for small tumors.73 The small choroidal melanoma observational study identified clinical features associated with time to tumor growth. The Kaplan-Meier estimate of the probability of growth was 11 and 31% by 1 and 5 years respectively. It must be realized that about two-thirds of tumors did not show evidence of growth, even though these tumors were classified as small choroidal melanoma. Patients with medium
(b) Figure 3 Fundus photograph showing a pigmented choroidal mass (a). Ultrasonography demonstrates a dome-shaped choroidal mass with retinal detachment (b).
choroidal melanoma were randomized to receive enucleation or iodine-125 brachytherapy.71,74,75 The Kaplan-Meier estimates of 5-year all-cause and melanoma-related mortality rates were comparable for both treatment arms. Patients with a large choroidal melanoma were randomly assigned to enucleation with and without pre-enucleation radiation (20 Gy was delivered in five daily fractions).72,76 The Kaplan-Meier estimates of 5-year all-cause mortality and melanoma-related mortality were comparable for both treatment arms.
Authors’ Recommendations Tumor size is the most significant factor that influences the treatment recommendation. In general, COMS guidelines are followed. Small-sized choroidal melanomas that are less than 4 mm in thickness may be treated by TTT.
OPHTHALMIC CANCERS Table 3 Classification system and design of Collaborative Ocular Melanoma Trials.a
Size (mm)
Classification
Diameter
Nevus Small melanoma Medium melanoma
Large melanoma
<5.0 5.0 – 16 ≤16
>16
Height
Design
Treatment groups
<1.0 1.0 – 2.5 Nonrandomized Observationb 2.5 – 10 Randomized Iodine 1 – 25 plaque radiotherapy Enucleation >10 Randomized Preenucleation radiotherapy Enucleation
a
After November, 1990. Treatment offered initially or during follow-up based on the discretion of the patient and treating ophthalmologist.
b
Prognosis The survival in patients with uveal melanoma is compromised due to the tendency of uveal melanoma to undergo hepatic metastasis. Among other factors, tumor size, tumor cell type, and presence of cytogenetic changes are significant prognostic factors.62 The COMS data indicates approximate melanoma-related mortality at 5 years of 1, 10, and 25% for small, medium, and large choroidal melanoma respectively. Currently used screening protocols of annual chest X rays and liver function tests every 6 months are ineffective, unless liver imaging studies are included.77
RETINOBLASTOMA Historical Background James Wardrop in 1809 was the first to recognize retinoblastoma as a specific entity and called it fungus haematodes. Enucleation with removal of a long stump of optic nerve as treatment was recommend by von Graefe in 1884. Hilgartner first used external beam radiotherapy in 1903.78 Kupfer used chemotherapy in 1953 for the treatment of retinoblastoma.79 Reese and Ellsworth proposed a classification for retinoblastoma, which has guided the treatment for the last 50 years.80 More recently, there is a move toward new classification of retinoblastoma which has better correlation with currently used methods of treatment.81
Anatomy The retina is the innermost layer of the eyeball. It is a highly complex neural structure and embryologically represents a direct extension of the brain. Retinoblastoma arises from the as yet unidentified retinoblasts that transiently populate the pediatric retina.
Biology and Epidemiology Retinoblastoma is the most common primary intraocular malignant tumor in children with an incidence of 1 in 15 000 live births. The average annual incidence of retinoblastoma
717
in the United States is 10.9 per million for children younger than 5 years.82 Retinoblastoma is a familial disorder with an autosomaldominant inheritance. Retinoblastoma can be classified in three different ways: familial or sporadic, bilateral or unilateral, and heritable or nonheritable. Approximately 10% of newly diagnosed retinoblastoma cases are familial and 90% are sporadic. All familial cases are due to an inherited germ line mutation. All bilateral cases and about 15% of unilateral cases represent new onset of germ line mutations.83 Human retinoblastoma susceptibility gene (RB1), a tumor suppressor gene, is located on chromosome 13 q14.84 The mutations are distributed throughout the RB1 gene with no mutational hotspots. Retinoblastoma protein (pRB) arrests the cell cycle at G1 restriction point by binding to E2F transcription factors. In the absence of both alleles of RB1, one due to inactivating the mutation and the other due to either a separate somatic mutation or due to chromosomal mechanisms, there is almost complete lack of pRB. In the absence of pRB, there is uncontrolled cell proliferation leading to retinoblastoma.85
Pathology Retinoblastoma growth patterns can be classified as one of four patterns: exophytic, endophytic, mixed, and diffuse infiltrating. Endophytic tumor grows toward the vitreous cavity and mimics endophthalmitis. Exophytic tumors tend to grow outwards leading to retinal detachment. Diffuse infiltrating tumors lack localized tumor prominence and are usually misdiagnosed as uveitis. Histopathologically, retinoblastoma is a neuroblastic tumor with large basophilic nuclei and scanty cytoplasm. Necrosis, mitosis, and calcification are common features. Flexner-Wintersteiner rosettes are highly characteristic of retinoblastoma.86
Clinical Presentation and Diagnostic Considerations Leukocoria (a white pupil) and strabismus are the most common presenting signs (see Figure 4).87 Other less common presentations include intraocular inflammation, discoloration of the iris due to neovascularization, hyphema (blood in the anterior chamber), and glaucoma. As children with a family history of retinoblastoma are screened prospectively by periodic indirect ophthalmoscopy, early retinoblastoma may be detected by screening examinations. Diagnosis is usually based on the ophthalmoscopic appearance of a white tumor with intrinsic calcifications demonstrated by ultrasonography or CT scan. Intraocular biopsy in the setting of retinoblastoma is contraindicated because of concerns regarding seeding of malignant cells.88 Metastatic disease is rare at presentation; therefore, lumbar puncture with cerebrospinal fluid analysis, bone marrow aspiration, and bone scan are not performed on a routine basis.89 However, these tests are indicated in children with advanced intraocular disease or who show evidence of extraocular disease at presentation. Retinocytoma or the socalled spontaneously regressed retinoblastoma is the benign equivalent of retinoblastoma.90
718
NEUROLOGICAL MALIGNANCIES
Authors’ Recommendations The decision-making process for treatment of retinoblastoma is rather complex as several factors such as family history, status of the other eye, number, size, and location of tumors, and the visual potential are important considerations. In general, advanced unilateral disease is best managed by enucleation with continued monitoring of the uninvolved eye. Treatment of bilateral disease is initiated by chemotherapy (chemoreduction) with local therapy comprising of cryotherapy, thermotherapy, or brachytherapy. External beam radiation therapy as an initial treatment for bilateral disease is generally not recommended.
Prognosis
Figure 4 Leukocoria due to retinoblastoma.
Treatment In recent years, there has been a trend away from enucleation, with the increased use of alternative globe conserving methods of treatment including, laser photocoagulation, cryotherapy, TTT, plaque radiotherapy, external beam radiotherapy, and chemotherapy. Laser photocoagulation and cryotherapy are used to treat very small tumors.91 TTT is used for small tumors in conjunction with chemotherapy.92 Plaque radiotherapy is highly effective in treating medium-sized tumors.93 External beam radiation therapy was used for large and multiple tumors associated with vitreous seeding.94 It is used less frequently nowadays due to concerns of inducing second malignant neoplasms (SMNs). Enucleation is still a valid primary therapeutic option for advanced unilateral retinoblastoma. Since the 1990s, chemoreduction has been increasingly used for the management of retinoblastoma to avoid external beam radiotherapy or enucleation.95 Chemotherapy is delivered intravenously to reduce the volume of intraocular retinoblastoma and also to make it amenable to focal therapy such as cryotherapy, thermotherapy, or brachytherapy. Six-cycle chemoreduction using three agents (vincristine, etoposide, and carboplatin) is generally prescribed.96 On the basis of available (noncomparative series) data, it can be concluded that chemoreduction combined with adjuvant focal therapy offers about a 50 to 90% probability of avoiding enucleation or external beam radiotherapy depending upon the severity of disease at initial presentation.97 It must be realized that chemoreduction is not without its problems. Recurrence of the neoplasm while on chemotherapy has been observed.98 Immediate complications related to transient bone marrow suppression requiring hospital admissions and intravenous antibiotics with consequent delay in examinations under anesthesia are frequent.99 Risk of late complications such as drug-induced leukemia has not yet been excluded. It is recommended that chemoreduction therapy for retinoblastoma should only be offered at a specialist center.
The survival in patients with retinoblastoma is compromised by three independent diseases: metastatic retinoblastoma, trilateral retinoblastoma, and SMNs. Recent advances in the treatment of retinoblastoma have led to improved 5-year survival rates of greater than 90% in the developed countries.100,101 However, in the underdeveloped countries, retinoblastoma is associated with high mortality rate because it tends to present at a much more advanced stage. The metastasis in retinoblastoma usually occurs within 1 year of diagnosis of retinoblastoma and almost never after 5 years.102 Involvement of the central nervous system and hematogenous spread are the commonest sites of metastasis. Survival with metastatic retinoblastoma is generally limited to 6 months.102,103 Several histopathologic risk factors for metastasis have been identified, in the presence of which, chemotherapy is generally recommended.104,105 In about 8% of cases with germ line retinoblastoma, a primary intracranial tumor is observed.106 The nature of primary intracranial malignant tumor can be varied in its location and histopathologic features. The majority of tumors are located in the pineal region but the tumors can also occur in the suprasellar and parasellar regions. The histopathologic appearance is of a pinealoblastoma or a primitive neuroectodermal tumor.106,107 In the United States, more patients die of SMN than from their initial retinoblastoma.108 SMNs typically occur in adolescence in children with germline mutations. Treatment with external beam radiotherapy enhances the risk for SMN.109 A variety of tumors are seen within the spectrum of SMN, but osteogenic sarcoma, soft tissue sarcomas, and cutaneous melanoma are the most frequently observed.
REFERENCES 1. Straatsma BR. Meibomian gland tumors. AMA Arch Ophthalmol 1956; 56: 71 – 93. 2. Kass LG, Hornblass A. Sebaceous carcinoma of the ocular adnexa. Surv Ophthalmol 1989; 33: 477 – 90. 3. Shields JA, et al. Sebaceous carcinoma of the ocular region: a review. Surv Ophthalmol 2005; 50: 103 – 22. 4. Margo CE, Mulla ZD. Malignant tumors of the eyelid: a populationbased study of non-basal cell and non-squamous cell malignant neoplasms. Arch Ophthalmol 1998; 116: 195 – 8. 5. Ni C, et al. Sebaceous cell carcinomas of the ocular adnexa. Int Ophthalmol Clin 1982; 22: 23 – 61.
OPHTHALMIC CANCERS 6. Abdi U, et al. Tumours of eyelid: a clinicopathologic study. J Indian Med Assoc 1996; 94: ,16,18 405 – 9. 7. Stockl FA, et al. Sebaceous carcinoma of the eyelid in an immunocompromised patient with Muir-Torre syndrome. Can J Ophthalmol 1995; 30: 324 – 6. 8. Rundle P, et al. Sebaceous gland carcinoma of the eyelid seventeen years after irradiation for bilateral retinoblastoma. Eye 1999; 13: 109 – 10. 9. Kivela T, et al. Sebaceous carcinoma of the eyelid associated with retinoblastoma. Ophthalmology 2001; 108: 1124 – 8. 10. Rumelt S, et al. Four-eyelid sebaceous cell carcinoma following irradiation. Arch Ophthalmol 1998; 116: 1670 – 2. 11. Yen MT, Tse DT. Sebaceous cell carcinoma of the eyelid and the human immunodeficiency virus. Ophthal Plast Reconstr Surg 2000; 16: 206 – 10. 12. Rao NA, et al. Sebaceous carcinomas of the ocular adnexa: a clinicopathologic study of 104 cases, with five-year follow-up data. Hum Pathol 1982; 13: 113 – 22. 13. Sinard JH. Immunohistochemical distinction of ocular sebaceous carcinoma from basal cell and squamous cell carcinoma. Arch Ophthalmol 1999; 117: 776 – 83. 14. Doxanas MT, Green WR. Sebaceous gland carcinoma. Review of 40 cases. Arch Ophthalmol 1984; 102: 245 – 9. 15. Muqit MM, et al. Improved survival rates in sebaceous carcinoma of the eyelid. Eye 2004; 18: 49 – 53. 16. Hornblass A, Lauer SA. Sebaceous carcinoma of the eyelids. Ophthalmology 2004; 111: 2149 – 50. 17. Boniuk M, Zimmerman LE. Sebaceous carcinoma of the eyelid, eyebrow, caruncle, and orbit. Trans Am Acad Ophthalmol Otolaryngol 1968; 72: 619 – 42. 18. Putterman AM. Conjunctival map biopsy to determine pagetoid spread. Am J Ophthalmol 1986; 102: 87 – 90. 19. Cook BE Jr, Bartley GB. Treatment options and future prospects for the management of eyelid malignancies: an evidence-based update. Ophthalmology 2001; 108: 2088 – 98. 20. Lisman RD, Jakobiec FA, Small P. Sebaceous carcinoma of the eyelids. The role of adjunctive cryotherapy in the management of conjunctival pagetoid spread. Ophthalmology 1989; 96: 1021 – 6. 21. Rosner M, Hadar I, Rosen N. Successful treatment with mitomycin C eye drops for conjunctival diffuse intraepithelial neoplasia with sebaceous features. Ophthal Plast Reconstr Surg 2003; 19: 477 – 9. 22. Nunery WR, Welsh MG, McCord CD Jr. Recurrence of sebaceous carcinoma of the eyelid after radiation therapy. Am J Ophthalmol 1983; 96: 10 – 5. 23. Yen MT, et al. Radiation therapy for local control of eyelid sebaceous cell carcinoma: report of two cases and review of the literature. Ophthal Plast Reconstr Surg 2000; 16: 211 – 5. 24. Callahan EF, et al. Sebaceous carcinoma of the eyelid: a review of 14 cases. Dermatol Surg 2004; 30: 1164 – 8. 25. Wilson MW, et al. Sentinel node biopsy for orbital and ocular adnexal tumors. Ophthal Plast Reconstr Surg 2001; 17: 338 – 44. 26. Duke-Elder S, Leigh A. Diseases of the outer eye. In Duke-Elder S (ed) Systems of Ophthalmology. Vol. 7 St. Louis, Missouri: CV Mosby, 1985. 27. Spencer W. Conjunctiva. In Spencer W (ed) Ophthalmic Pathology: An Atlas and Textbook. Philadelphia, Pennsylvania: W. B. Saunders Co, 1996. 28. Lee GA, Hirst LW. Ocular surface squamous neoplasia. Surv Ophthalmol 1995; 39: 429 – 50. 29. Grossniklaus HE, et al. Conjunctival lesions in adults. A clinical and histopathologic review. Cornea 1987; 6: 78 – 116. 30. Sun EC, Fears TR, Goedert JJ. Epidemiology of squamous cell conjunctival cancer. Cancer Epidemiol Biomarkers Prev 1997; 6: 73 – 7. 31. Lee GA, Hirst LW. Incidence of ocular surface epithelial dysplasia in metropolitan Brisbane. A 10-year survey. Arch Ophthalmol 1992; 110: 525 – 7. 32. Pizzarello LD, Jakobiec FA. Bowen’s disease of the conjunctiva: a misnomer. In Jakobiec FA (ed) Ocular and Adnexal Tumors. Birmingham, Alabama: Aesculapius, 1978.
719
33. McDonnell JM, McDonnell PJ, Sun YY. Human papillomavirus DNA in tissues and ocular surface swabs of patients with conjunctival epithelial neoplasia. Invest Ophthalmol Vis Sci 1992; 33: 184 – 9. 34. Brash DE, et al. Sunlight and sunburn in human skin cancer: p53, apoptosis, and tumor promotion. J Investig Dermatol Symp Proc 1996; 1: 136 – 42. 35. Kestelyn P. Ocular problems in AIDS. Int Ophthalmol 1990; 14: 165 – 72. 36. Waring GO, Roth AM 3rd, Ekins MB. Clinical and pathologic description of 17 cases of corneal intraepithelial neoplasia. Am J Ophthalmol 1984; 97: 547 – 59. 37. Shields CL, Shields JA. Tumors of the conjunctiva and cornea. Surv Ophthalmol 2004; 49: 3 – 24. 38. Rao NA, Font RL. Mucoepidermoid carcinoma of the conjunctiva: a clinicopathologic study of five cases. Cancer 1976; 38: 1699 – 709. 39. McKelvie PA, et al. Squamous cell carcinoma of the conjunctiva: a series of 26 cases. Br J Ophthalmol 2002; 86: 168 – 73. 40. Lee GA, Hirst LW. Retrospective study of ocular surface squamous neoplasia. Aust N Z J Ophthalmol 1997; 25: 269 – 76. 41. Erie JC, Campbell RJ, Liesegang TJ. Conjunctival and corneal intraepithelial and invasive neoplasia. Ophthalmology 1986; 93: 176 – 83. 42. Shields JA, Shields CL, De Potter P. Surgical management of conjunctival tumors. The 1994 Lynn B. McMahan Lecture. Arch Ophthalmol 1997; 115: 808 – 15. 43. Fraunfelder FT, Wingfield D. Management of intraepithelial conjunctival tumors and squamous cell carcinomas. Am J Ophthalmol 1983; 95: 359 – 63. 44. Buus DR, et al. Microscopically controlled excision of conjunctival squamous cell carcinoma. Am J Ophthalmol 1994; 117: 97 – 102. 45. Frucht-Pery J, Rozenman Y, Pe’er J. Topical mitomycin-C for partially excised conjunctival squamous cell carcinoma. Ophthalmology 2002; 109: 548 – 52. 46. Frucht-Pery J, et al. Mitomycin C treatment for conjunctival-corneal intraepithelial neoplasia: a multicenter experience. Ophthalmology 1997; 104: 2085 – 93. 47. Yeatts RP, et al. Topical 5-fluorouracil in treating epithelial neoplasia of the conjunctiva and cornea. Ophthalmology 1995; 102: 1338 – 44. 48. Singh AD, et al. Neoadjuvant topical mitomycin C chemotherapy for conjunctival and corneal intraepithelial neoplasia. Eye 2005; In Press. 49. Vann RR, Karp CL. Perilesional and topical interferon alfa-2b for conjunctival and corneal neoplasia. Ophthalmology 1999; 106: 91 – 7. 50. Smith M, et al. Lack of toxicity of a topical recombinant interferon alpha. Cornea 1989; 8: 58 – 61. 51. Singh AD. Excision and cryosurgery of conjunctival malignant epithelial tumours. Eye 2003; 17: 125 – 6. 52. Glasson WJ, et al. Invasive squamous cell carcinoma of the conjunctiva. Arch Ophthalmol 1994; 112: 1342 – 5. 53. Schachat AP. Management of uveal melanoma: a continuing dilemma. Collaborative Ocular Melanoma Study Group. Cancer 1994; 74: 3073 – 5. 54. Fine SL. No one knows the preferred management for choroidal melanoma. Am J Ophthalmol 1996; 122: 106 – 8. 55. Markowitz JA, et al. A review of mortality from choroidal melanoma. I. Quality of published reports, 1966 through 1988. Arch Ophthalmol 1992; 110: 239 – 44. 56. Diener-West M, et al. A review of mortality from choroidal melanoma. II. A meta-analysis of 5-year mortality rates following enucleation, 1966 through 1988. Arch Ophthalmol 1992; 110: 245 – 50. 57. Zimmerman LE, McLean IW, Foster WD. Does enucleation of the eye containing a malignant melanoma prevent or accelerate the dissemination of tumour cells. Br J Ophthalmol 1978; 62: 420 – 5. 58. Singh AD, et al. Zimmerman-McLean-Foster hypothesis: 25 years later. Br J Ophthalmol 2004; 88: 962 – 7. 59. Singh AD, Topham A. Incidence of uveal melanoma in the United States: 1973 – 1997. Ophthalmology 2003; 110: 956 – 61. 60. Chang AE, Karnell LH, Menck HR. The National Cancer Data Base report on cutaneous and noncutaneous melanoma: a summary of 84,836 cases from the past decade. The American College of Surgeons Commission on Cancer and the American Cancer Society. Cancer 1998; 83: 1664 – 78.
720
NEUROLOGICAL MALIGNANCIES
61. Singh AD, Bergman L, Seregard S. Uveal melanoma: epidemiologic aspects. Ophthalmol Clin North Am 2005; 18: 75 – 84. 62. Singh AD, Shields CL, Shields JA. Prognostic factors in uveal melanoma. Melanoma Res 2001; 11: 255 – 63. 63. Singh AD, Borden EC. Metastatic uveal melanoma. Ophthalmol Clin North Am 2005; 18: 143 – 150, ix. 64. Vajdic CM, et al. Incidence of ocular melanoma in Australia from 1990 to 1998. Int J Cancer 2003; 105: 117 – 22. 65. Michalova K, et al. Iris melanomas: are they more frequent in New Zealand? Br J Ophthalmol 2001; 85: 4 – 5. 66. Strickland D, Lee JA. Melanomas of eye: stability of rates. Am J Epidemiol 1981; 113: 700 – 2. 67. Egan KM, et al. Epidemiologic aspects of uveal melanoma. Surv Ophthalmol 1988; 32: 239 – 51. 68. Singh AD, et al. Uveal melanoma: genetic aspects. Ophthalmol Clin North Am 2005; 18: 85 – 97. 69. Group COMS. Histopathologic characteristics of uveal melanomas in eyes enucleated from the Collaborative Ocular Melanoma Study. COMS report no. 6. Am J Ophthalmol 1998; 125: 745 – 66. 70. Group COMS. COMS Manual of Procedures: accession no. PBS 179693. Springfield, Virginia: National Technical Information Service, 1995. 71. Diener-West M, et al. The COMS randomized trial of iodine 125 brachytherapy for choroidal melanoma, III: initial mortality findings. COMS Report No. 18. Arch Ophthalmol 2001; 119: 969 – 82. 72. Group COMS. The Collaborative Ocular Melanoma Study (COMS) randomized trial of pre-enucleation radiation of large choroidal melanoma II: initial mortality findings. COMS report no. 10. Am J Ophthalmol 1998; 125: 779 – 96. 73. Group COMS. Mortality in patients with small choroidal melanoma. COMS report no. 4. The Collaborative Ocular Melanoma Study Group. Arch Ophthalmol 1997; 115: 886 – 93. 74. Melia BM, et al. Collaborative ocular melanoma study (COMS) randomized trial of I-125 brachytherapy for medium choroidal melanoma. I. Visual acuity after 3 years COMS report no. 16. Ophthalmology 2001; 108: 348 – 66. 75. Diener-West M, et al. The COMS randomized trial of iodine 125 brachytherapy for choroidal melanoma, II: characteristics of patients enrolled and not enrolled. COMS report no. 17. Arch Ophthalmol 2001; 119: 951 – 65. 76. Group COMS. The Collaborative Ocular Melanoma Study (COMS) randomized trial of pre-enucleation radiation of large choroidal melanoma III: local complications and observations following enucleation COMS report no. 11. Am J Ophthalmol 1998; 126: 362 – 72. 77. Diener-West M, et al. Screening for metastasis from choroidal melanoma: the Collaborative Ocular Melanoma Study Group Report 23. J Clin Oncol 2004; 22: 2438 – 44. 78. Albert DM. Historic review of retinoblastoma. Ophthalmology 1987; 94: 654 – 62. 79. Kupfer C. Retinoblastoma treated with intravenous nitrogen mustard. Am J Ophthalmol 1953; 36: 1721 – 3. 80. Reese AB. Tumors of the eye. New York: Harper and Row Publishers, 1976: 127. 81. Linn Murphree A. Intraocular retinoblastoma: the case for a new group classification. Ophthalmol Clin North Am 2005; 18: 41 – 53. 82. Young JJL, et al. Retinoblastoma. In Ries LAG, et al. (eds) Cancer Incidence and Survival among Children and Adolescents: United States SEER Program 1975 – 1995. Bethesda, Maryland: National Cancer Institute, SEER Program. NIH Pub. No. 99 – 4649, 1999. 83. Gallie BL, et al. The genetics of retinoblastoma. Relevance to the patient. Pediatr Clin North Am 1991; 38: 299 – 315.
84. Friend SH, et al. A human DNA segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma. Nature 1986; 323: 643 – 6. 85. Gallie BL, et al. How do retinoblastoma tumours form? Eye 1992; 6: 226 – 31. 86. Sang DN, Albert DM. Retinoblastoma: clinical and histopathologic features. Hum Pathol 1982; 13: 133 – 47. 87. Abramson DH, et al. Presenting signs of retinoblastoma. J Pediatr 1998; 132: 505 – 8. 88. Shields CL, et al. Vitrectomy in eyes with unsuspected retinoblastoma. Ophthalmology 2000; 107: 2250 – 5. 89. Pratt CB, Crom DB, Howarth C. The use of chemotherapy in extraocular retinoblastoma. Med and Pediatr Oncol 1985; 13: 330 – 3. 90. Singh AD, et al. Observations on 17 patients with retinocytoma. Arch Ophthalmol 2000; 118: 199 – 205. 91. Shields JA, et al. The role of cryotherapy in the management of retinoblastoma. Am J Ophthalmol 1989; 108: 260 – 4. 92. Journee-de Korver HG, Midena E, Singh AD. Infrared thermotherapy: from laboratory to clinic. Ophthalmol Clin North Am 2005; 18: 99 – 110. 93. Shields JA, et al. Plaque radiotherapy for residual or recurrent retinoblastoma in 91 cases. J Pediatr Ophthalmol Strabismus 1994; 31: 242 – 5. 94. Hernandez JC, et al. External beam radiation for retinoblastoma: results, patterns of failure, and a proposal for treatment guidelines. Intl J Radiat Oncol Biol Phys 1996; 35: 125 – 32. 95. Ferris FL, Chew EY. A new era for the treatment of retinoblastoma. Arch Ophthalmol 1996; 114: 1412. 96. Chan HS, et al. Chemotherapy for retinoblastoma. Ophthalmol Clin North Am 2005; 18: 55 – 63. 97. Shields CL, Honavar SG, Meadows AT. Chemoreduction for retinoblastoma: factors predictive of failure and need for treatment with external beam radiotherapy or enucleation. Am J Ophthalmol 2002; 133: 657 – 64. 98. Scott IU, et al. New retinoblastoma tumors in children undergoing systemic chemotherapy. Arch Ophthalmol 1998; 12: 1685 – 6. 99. Benz MS, et al. Complications of systemic chemotherapy as treatment of retinoblastoma. Arch Ophthalmol 2000; 118: 577 – 8. 100. Tamboli A, Podgor MJ, Horm JW. The incidence of retinoblastoma in the United States: 1974 through 1985. Arch Ophthalmol 1990; 108: 128 – 32. 101. Ajaiyeoba IA, et al. Retinoblastomas in Ibadan: treatment and prognosis. West Afr J Med 1993; 12: 223 – 7. 102. Kopelman JE, McLean IW, Rosenberg SH. Multivariate analysis of risk factors for metastasis in retinoblastoma treated by enucleation. Ophthalmology 1987; 94: 371 – 7. 103. McCay CJ, Abramson DH, Ellsworth RM. Metastatic patterns of retinoblastoma. Arch Ophthalmol 1984; 102: 391 – 6. 104. Honavar SG, et al. Post-enucleation adjuvant chemotherapy in highrisk retinoblastoma. Arch Ophthalmol 2002; 120(7): 923 – 31. 105. Singh AD, Shields CL, Shields JA. Prognostic factors in retinoblastoma. J Pediatr Ophthalmol Strabismus 2000; 37: 1 – 8. 106. Singh AD, Shields CL, Shields JA. New insights into trilateral retinoblastoma. Cancer 1999; 86: 3 – 5. 107. Kivela T. Trilateral retinoblastoma: a meta-analysis of hereditary retinoblastoma associated with primary ectopic intracranial retinoblastoma. J Clin Oncol 1999; 17: 1829 – 37. 108. Eng C, et al. Mortality from second tumors among long-term survivors of retinoblastoma. J Natl Cancer Inst 1993; 85: 1121 – 8. 109. Abramson DH, Frank CM. Second nonocular tumors in survivors of bilateral retinoblastoma: a possible age effect on radiation-related risk. Ophthalmology 1998; 105: 573 – 9.
Section 11 : Pediatric Malignancies
66
Rare Pediatric Malignancies of the Head and Neck
Ted A. James, Larry L. Myers, Nestor Rigual, Janet S. Winston, Thom R. Loree and Wesley L. Hicks, Jr
INTRODUCTION Only trauma exceeds malignancy as the leading cause of pediatric mortality. According to data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results’ tumor database, 12% of all pediatric tumors develop in the head and neck. The average annual incidence of pediatric head and neck cancer between 1973 and 1996 increased by 35% compared to a 25% rise for all other tumors in children less than 15 years.1 Most commonly, pediatric head and neck cancers include lymphoma, rhabdomyosarcoma, and thyroid cancers. Less frequent causes of pediatric head and neck malignancy include nasopharyngeal cancer, neuroblastoma, sarcoma, salivary gland tumors, and malignant teratomas. The approach to pediatric head and neck tumors often parallels the treatment approach to the adult variants of these neoplastic processes. However, with a greater understanding of the specific pathophysiology and tumor biology in pediatric malignancies, tailored multidisciplinary approaches have evolved and have heralded improved outcomes. Tumors occurring in the pediatric population are generally best treated by a multidisciplinary approach. Meticulous posttreatment surveillance is necessary as adverse effects may arise several years following cancer treatment. The following discussion will focus on the histopathology, clinical features, and management of the uncommon pediatric head and neck malignancies of extraosseous Ewing’s sarcoma (EOE), ameloblastoma odontogenic tumors, sinonasal undifferentiated carcinoma (SNUC), and esthesioneuroblastoma.
EXTRAOSSEOUS EWING’S SARCOMA Introduction The first description of this sarcoma was by James Ewing in 1921. The initial description of EOE, which is identical
in histologic appearance to osseous Ewing’s sarcoma (OES), was by Tefft and colleagues in 1969.2 They reported four pediatric patients with paravertebral “round cell” soft tissue sarcomas. In 1975, Angervall and Enzinger reported on 39 patients with EOE.3 In 1997, the Intergroup Rhabdomyosarcoma Study (IRS) reported on 130 pediatric patients with EOE, and is the largest series to date.4
Epidemiology The mean age at diagnosis is 11 years.5 Most patients will present before 30 years of age.3 The male-to-female ratio for children with EOE is 1.17 : 1. African-Americans in the United States are rarely affected compared to their white counterparts.
Pathology The gross appearance of EOE is varied. It can be multilobulated, soft, and friable, but rarely exceeds 10 cm in greatest dimension. On cut surface it has a gray-yellow to tan appearance, often with large areas of necrosis and cyst formation with hemorrhage. Microscopically, it has a solidly packed cellular lobular pattern comprised of small, uniform round cells with ovoid vesicular nuclei and scant ill-defined cytoplasm. In the majority of cases, the cytoplasm contains irregular vacuoles, secondary to the presence of glycogen (see Figure 1). The tumor is richly vascular, but the vessel walls are compressed due to the cellular nature of the tumor. Degenerating cells characterize the “filigree pattern” of EOE with distinct vascular structures having thickened, fibrotic walls and dilated lumina. Trabecular, globoid, and pseudorosette patterns may also be seen. The cells are usually glycogen positive. A dimorphic distribution of large variable and small necrotic cells is characteristic, but not specific, for Ewing’s sarcoma. Immunohistochemical and ultrastructural studies5 allowed pathologists to classify EOE into three categories: (i) tumors
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
722
PEDIATRIC MALIGNANCIES
soft tissue sarcomas, the duration of symptoms is usually less than 1 year prior to presentation. The most frequent primary tumor site in the head and neck region is the paravertebral region.
Evaluation
Figure 1 Extraskeletal Ewing’s sarcoma. There is a proliferation of uniform small round cells with scanty cytoplasm. Concentration around vascular structures gives a filigree appearance (×100).
with bidirectional neuroblastic and Schwannian differentiation, (ii) tumors with monodirectional neuroblastic differentiation, and (iii) tumors with monodirectional Schwannian differentiation.
Biology It is now known that Ewing’s sarcoma and peripheral tumors called primitive neuroectodermal tumor (PNET), share a common and unique chromosomal translocation.6 Ewing’s sarcoma is of neural crest origin and can synthesize acetylcholine transferase. The chromosomal translocation is t(11; 22)(q24; q12), and is present in up to 95% of these tumors.7 As a consequence of this translocation, the 5 portion of the EWS gene from band 22q12 is fused to the 3 portion of the FLI1 gene from band 11q24. The fusion results in the formation of a hybrid EWS-FLI1 protein. This EWSFLI1 contributes to the transformed phenotype of Ewing’s sarcoma cells by inducing a deregulated activation of genes, or an unscheduled activation of genes, or both.8
Patterns of Spread EOE is an aggressive, poorly differentiated tumor with local enlargement or destruction, with a potential to metastasize via hematogenous or lymphatic routes. The most common sites of metastasis are the lung and bone. Rarely, EOE may metastasize to regional lymph nodes.
Clinical Presentation EOE arises in the head and neck region in 49,10 to 18%4 of all patients with this disease. Most patients with EOE present with a rapidly enlarging mass usually arising in the soft tissues adjacent to the cervical spine11 and may have associated deep-seated pain or tenderness.4 According to a study of 42 patients at the Mayo Clinic,12 75% of patients presented with a palpable neck mass and 66% with pain. Superficially located tumors are rare, but do occur. If the peripheral nerves or the spinal cord is involved, it may cause progressive sensory or motor disturbances. As with similar
The radiologic evaluation of EOE can be effectively performed by computed tomography (CT) scan or magnetic resonance imaging (MRI). These investigations will delineate the tumor and show the relationship to vital structures. If bony involvement of this tumor is suspected, we prefer CT scan for initial evaluation. CT scans usually demonstrate an enhancing soft tissue mass in the involved region. MRI is the preferred imaging study if extensive soft tissue tumor invasion is a prominent feature. Magnetic resonance angiogram (MRA) is useful in delineating critical vascular structures and avoids the attendant morbidity of an angiogram. On plain roentgenograms, though not a preferred radiographic diagnostic technique, the vertebrae adjacent to the soft tissue mass may show periosteal thickening and new bone formation.2
Differential Diagnosis The diagnosis of EOE can be confirmed on the basis of immunohistochemical techniques. Immunohistochemistry with the monoclonal antibody HBA-71 to the p30/32MIC2 antigen have been expressed almost exclusively on the cell membranes of Ewing’s sarcoma and PNET, but not on those of other small cell tumors of childhood.13 Histologically, EOE can be confused with alveolar rhabdomyosarcoma, primitive peripheral neuroectodermal tumor, malignant lymphoma, small cell osteosarcoma, angiosarcoma, malignant hemangiopericytoma, melanotic neuroectodermal tumor of infancy, polyphenotypic small cell tumors, and intraabdominal desmoplastic small cell tumor. EOE tumors show evidence of neural differentiation as manifested by positive neuron-specific enolase (NSE) and/or S-100 stains, indicating neuroblastic or Schwannian cell differentiation. Often, neurosecretory cells can be seen on electron microscopy. Ultrastructural studies can sometimes be useful to distinguish primitive rhabdomyosarcoma from EOE.4 EOE is derived from peripheral neuroectodermal tumors.6 The evidence of a chromosomal translocation denoted as t(11;22)(q24;q12), which creates a fusion between parts of the EWS and FLI1 genes, is present in an overwhelming majority of patients with EOE and PNET. This cytogenetic evaluation may be used to differentiate EOE and PNET from other small, round cell tumors.
Treatment In the past, treatment for this malignancy consisted of surgical excision alone or in combination with radiation therapy. Mortality, local recurrence, and metastasis3 were high. The multimodality regimens utilized in the treatment of OES are generally used to treat both adults and children with EOE. Until recently, patients with
RARE PEDIATRIC MALIGNANCIES OF THE HEAD AND NECK
EOE were treated on IRS protocols. The IRS group recently reported on 130 patients with EOE and 23 (18%) had tumors of the head and neck region.4 All patients were given multiagent chemotherapy (vincristine, dactinomycin, and cyclophosphamide) and most received radiation therapy. None of the patients received a bone marrow transplant. We presently use the following algorithm in the treatment of pediatric EOE of the head and neck. Primary local treatment of Ewing’s sarcoma of the head and neck is radiation and chemotherapy. Surgical resection may be preferred for smaller lesions. In the past, chemotherapy and radiation therapy have been started concomitantly. Presently, we choose to give two to five cycles of chemotherapy. Agents active in the treatment of EOE are vincristine, cyclophosphamide, ifosfamide, dactinomycin, doxorubicin, and etoposide. Presently, we would treat these patients with chemotherapy/radiation before surgical resection of residual or unresponsive lesion is undertaken. Chemotherapy should be continued during radiation therapy for 6 to 12 months afterward, depending on the regimen used. Radiation is given for a total dose of 5000 to 5500 cGy at 180 to 200 cGy per fraction. A hyperfractionation regimen has been used with 120 cGy twice a day for a total dose of 5040 to 6000 cGy.14 The 2-year local control rate was over 90% when aggressive systemic treatment was given. Future treatment strategies may be centered on new biology-based approaches. A recent experiment has demonstrated a direct correlation between vascular endothelial growth factor and the regulation of angiogenesis in Ewing’s sarcoma. This may eventually provide a novel means of targeted therapy for this tumor.15
Prognosis Favorable prognostic indicators include small tumor volume, the absence of distant metastases, young age, and a good initial response to chemotherapy.16 Outcomes have seen an improvement with the use of multimodal therapy. The IRS study4 showed that 82% of 130 patients treated with vincristine, dactinomycin, and cyclophosphamide, with radiation therapy achieved a complete response. Over 70% of EOE patients in this report survived for more than 5 years. Survival at 10 years was most likely for patients with tumors that arose in the head and neck region and in those who underwent gross tumor removal before initiation of chemotherapy. A recent study demonstrated that event-free survival reached 50% in patients with EOE treated with a combination of systemic and local therapy.17 Newer chemotherapy combination protocols utilizing ifosfamide and etoposide have also demonstrated improved outcome in patients with localized Ewing’s sarcoma.18 However, some authors have suggested the use of less-intensive chemotherapy regimens and a reduction of radiation therapy in certain cases, especially when complete surgical resection has been achieved.19
723
ODONTOGENIC TUMORS AMELOBLASTOMA AND MALIGNANT AMELOBLASTOMA Introduction Originally described by Fischer in 1913, ameloblastoma, also referred to as adamantinoma, is a rare, benign, yet locally aggressive, odontogenic tumor. We have chosen to discuss these two disease processes together because of their common histologic, clinical, and radiographic presentations. In fact, the diagnosis of malignant ameloblastoma is usually made only when there is the clinical presence of regional (neck metastasis) or distant (lung, bone) disease. The tumor arises from the elements of odontogenic tissue responsible for enamel formation in normal dentition. Rarely, extraosseous peripheral ameloblastoma is found in the gingiva; a few have also been found in the buccal mucosa. Peripheral lesions usually do not invade underlying bone, and they recur infrequently. Ameloblastoma with the usual benign histologic appearance, that metastasizes and has the same histologic appearance in both the primary and secondary deposits, is termed malignant ameloblastoma.20
Epidemiology Ameloblastoma represents 10% of all tumors of the jaw, and is localized in the mandible in approximately 80% of cases, while localization in the maxilla occurs in the remaining 20% of cases.21 There does not appear to be any predilection for gender.22 In the Western world, the average age of presentation of pediatric ameloblastoma is 14.3 years.23 Although rare in the United States, odontogenic tumors are not as uncommon in other parts of the world such as Nigeria, where ameloblastoma of the mandible and malignant odontogenic tumors are more frequently encountered in 9.6% of all oral cavity and mandible biopsies.24
Embryology Initial signs of tooth development occur at approximately the sixth week of gestation with downward proliferation of oral epithelium into the underlying mesenchyme. Proliferation progresses to a phase in development termed as the cap stage. During this period, the inner enamel epithelium, outer enamel epithelium, stratum intermedium, and stellate reticulum can be identified. Organization of the underlying connective tissue beneath the enamel organ also occurs, with development of the dental papilla and the future dental sac. At approximately the 14th week of gestation, differentiation into ameloblasts and odontoblasts occurs, accompanied by disintegration of the dental lamina that leaves epithelial remnants (epithelial rests of Malassez). Although the ultimate origin of ameloblastoma is the dental lamina, it can form from derivations of this structure such as epithelial rests, the epithelial lining of odontogenic cysts, basal cells of surface epithelium, epithelium of enamel organ, and heterotopic epithelium such as the primary gland.
724
PEDIATRIC MALIGNANCIES
Pathology Histologically, ameloblastoma is characterized by cords of epithelium with a peripheral layer of vacuolated columnar epithelial cells in palisades and a central region of loosely arranged stellate epithelial cells (sometimes called the stellate reticulum) that may develop features of squamous metaplasia.20 Cohesion of the epithelial cells may further decrease, with cystlike spaces appearing in the epithelial cords. The tumor cells usually show mild nuclear hyperchromatism. Ameloblastoma has been classified into five histologic patterns: follicular, acanthomatous, plexiform, granular cell, and desmoplastic and basal cell. This classification is of questionable usefulness because there is frequently more than one cell type present and the histological type does not correlate with biological behavior.25 Common to all subtypes is the polarization of cells around the proliferating nests in a pattern similar to that of ameloblasts of the enamel organ. We classify ameloblastoma into two histologic classifications: solid and cystic. This division has clinical relevance because treatment and prognosis differ substantially between solid and cystic lesions. The solid and multicystic ameloblastoma requires more aggressive treatment than its unicystic counterpart. It also has a higher (50–90%) recurrence rate if treated with curettage. The unicystic lesion has a single cystic space in which there is intraluminal or mural growth. It may represent a small ameloblastoma that is unilocular, or it may represent an odontogenic cyst in which there has been ameloblastic transformation of the epithelial lining. Unicystic ameloblastoma is seen in a younger age-group (second and third decades of life) and typically in the mandibular molar area. It has a recurrence rate of less than 10% when curettage is the primary form of treatment.
Biology (Ameloblastoma) The biological behavior of ameloblastoma is unpredictable. Despite the histological appearance of an indolent tumor, ameloblastoma can display an aggressive growth pattern with bony destruction. Local recurrence and malignant degeneration are also observed, particularly when the tumor is insufficiently excised initially.21 The identification of increased cellular proliferative activity in recurrent ameloblastoma led to the premise that this could be an explanation for the aggressive biological variation observed.26 Other authors have not established this correlation between biologic activity and proliferative rates. The interplay of other cellular mechanisms involving stromal tissue factors and proteolytic enzymes is speculated as contributing to the local invasiveness characteristic of this lesion.22,27,28
Biology/Patterns of Spread (Malignant Ameloblastoma) The stimulus for the neoplastic transformation of ameloblastoma to malignant ameloblastoma is unknown. There have been approximately 60 cases of malignant ameloblastoma reported in the medical literature between 1923 and 1994.20 The most common route of metastatic dissemination is thought to be hematogenous because of the possible scattered
spread and the findings of tumor cells in blood vessels.29,30 In 1989, Laughlin reviewed 42 previously published cases of malignant ameloblastoma and reported an additional case. He noted that up to 75% of metastases were in the chest, including the lungs, pleura, and hilar nodes, with cervical lymph nodes and spine each representing approximately 15% of metastatic sites.31 Metastases from ameloblastoma are frequently preceded by local recurrences.32,33 It has been postulated that dissemination may result from increased malignant behavior stimulated by multiple recurrences or that repeated surgical procedures required for the treatment of these recurrences cause implantation of tumor cells into the blood vessels or lymphatic channels.34
Clinical Presentation Ameloblastoma and malignant ameloblastoma are usually asymptomatic and are discovered either during radiographic evaluation or because of asymptomatic jaw expansion. Occasionally, tooth movement or malocclusion may be the initial presenting sign. Ameloblastoma may occur anywhere in the mandible or maxilla, although the mandibular molar –ramus area is the most common site. In the maxilla, the molar area is more commonly affected than the premolar or anterior regions. Rarely, extraosseous peripheral ameloblastomas are found in the gingiva or buccal area. Lesions in the buccal region are seen in older adults between 40 and 60 years of age.
Evaluation Both Panorex roentgenogram and computerized axial tomography are useful in establishing the diagnosis and extent of disease in ameloblastoma. Radiographically, ameloblastoma appears as osteolytic processes. This tumor, typically found in the tooth-bearing areas of the mandible, may exhibit a unilocular or multilocular appearance. Because ameloblastoma is slow growing, the radiographic margins are usually well defined and sclerotic. In cases where connective tissue desmoplasia occurs in conjunction with tumor proliferation, ill-defined radiographic margins may be seen. This type also has a predilection for the anterior jaws. The generally slow tumor growth rate is also responsible for the movement of tooth roots. Radiographic evidence of root resorption may also appear in association with ameloblastoma growth. Muller et al. described five cases (three primary mandible and two metastases) in which ameloblastoma was diagnosed by fine needle aspiration cytology.35 The smears were hypercellular and occasionally showed tissue fragments of basaloid cells with peripheral palisading. A distinct, two-cell population was seen, consisting of small, hyperchromatic, basaloid-type cells and scattered larger cells with more open chromatin. They concluded that, in the right clinical setting and with proper radiologic evidence, the cytologic features of primary and metastatic ameloblastoma are unique and are sufficient for an accurate diagnosis and follow-up of patients with a history of ameloblastoma. Screening for metastatic spread is advisable for all patients diagnosed with ameloblastoma20 and malignant ameloblastoma. The index of suspicion should be higher for recurrent
RARE PEDIATRIC MALIGNANCIES OF THE HEAD AND NECK
cases. Because malignant ameloblastoma commonly metastasize to the chest, cervical lymph nodes and bone, the minimum metastatic workup should include chest roentgenogram. If positive, a chest CT and bone scanning should be performed.
Differential Diagnosis When considering the location and radiographic appearance together, the clinical differential diagnosis can generally be limited to several entities in the three categories of mandibular disease: odontogenic tumors, cysts, and benign nonodontogenic tumors. Among odontogenic tumors, the calcifying epithelial odontogenic tumors (CEOTs) (radiolucent variety) and odontogenic myxomas bear close resemblance. The dentigerous cysts and odontogenic keratocysts can also be included. In young patients, lesions that are radiographically similar to ameloblastoma include nonodontogenic lesions such as central giant cell granuloma, ossifying fibroma, central hemangioma, and possibly idiopathic histiocytosis. Microscopically, an ameloblastoma, especially the plexiform uni- and multicystic type, may be confused with an odontogenic cyst in which there is hyperplasia of the lining. In ameloblastoma basal cell palisading is evident and inflammatory cells are usually scant. Maxillary ameloblastoma occasionally appears less differentiated, requiring distinction from adenocarcinoma and squamous cell carcinoma of maxillary sinus origin.
Treatment Surgery
Studies evaluating the surgical approach to ameloblastic lesions have revealed recurrence rates of 29–33% in patients treated with conservative management compared to 0–7% recurrence rates in patients treated with wide local resection.36,37 Although en bloc resection of the mandible or maxilla with 1–2 cm-margins of normal bone is an acceptable standard, there is agreement that conservative treatment with marsupialization or enucleation and bone curettage may be considered in select cases of unicystic lesions without extraosseous extension. Additionally, a modified radical conservative resection has been applied to mandibular ameloblastomas with good results, and may represent another acceptable alternative.38 The authors agree with the validity of a more conservative approach for select unicystic lesions. If the radiographic appearance and preoperative cytology suggest an encapsulated unicystic mandibular lesion, enucleation with curettage appears to be adequate therapy. In multicystic lesions of the mandible and in both unicystic and multicystic lesions of the maxilla, complete surgical excision with negative margins (1 to 2 cm) is the treatment of choice. The surgical approach outlined above has resulted in local recurrence rates of less than 15%. It appears that surgical resection is currently the most appropriate modality for the treatment of malignant ameloblastoma, with radiotherapy being used for metastatic disease not amenable to surgical resection.39
725
Radiation Therapy
The role of radiation therapy in the treatment of ameloblastoma and malignant ameloblastoma has never been clearly defined in the pediatric or adult population. The rarity of pediatric ameloblastoma, and the fact that many of the cases of radiation-treated ameloblastomas were handled before modern (megavoltage) irradiation, leave open the question of whether primary radiation therapy should be indicated in the treatment of ameloblastoma or malignant ameloblastoma. Radiotherapy should be considered in patients who are poor surgical candidates or in those who refuse extensive surgical procedures. Though generally considered radioresistant,40 there is no well-documented evidence in the literature concerning the true radioresponsiveness or radioresistance of ameloblastomas. Atkinson et al. retrospectively reviewed 10 patients with ameloblastoma treated with megavoltage irradiation at the Princess Margaret Hospital.41 Seven cases were treated with radiation alone and six (86%) responded initially. Only one patient subsequently had a recurrence and was successfully salvaged surgically. Three patients were treated with combined radiation therapy and surgery. They concluded that ameloblastoma is not an inherently radioresistant tumor, and that properly applied megavoltage irradiation has a useful role in the management of ameloblastoma. In particular, they concluded, where a full surgical excision would be technically difficult because of local invasion or where other medical factors would make surgery inappropriate, primary radiation should be considered. Recommended treatment dosages for use are between 3000 cGy and 5000 cGy. Complications of radiation therapy consist of osteonecrosis and the theoretical, yet rare, danger of postirradiation sarcoma or carcinoma, especially in pediatric patients.40 Chemotherapy as primary treatment of ameloblastoma does not appear indicated; results of such therapy for nonmetastatic disease have been poor.42 However, in the setting of metastatic disease,43 the use of cisplatin, adriamycin, and cyclophosphamide was found to be beneficial. Cure of ameloblastoma is possible with surgical intervention. Unfortunately, in the pediatric patient with multicystic ameloblastoma, the attendant morbidity from such a resection may be grave. Four factors should be considered in the treatment of ameloblastoma in children: (i) cure, (ii) preservation of facial skeleton growth centers, (iii) cosmesis, and (iv) dental rehabilitation. Growth in the maxilla in a forward and downward direction occurs by expansion of the orbits and growth of the nasal septum and cranial sutures. By 24 to 36 months of life, suture growth in the maxilla has subsided and further growth in the upper facial skeleton occurs by accretion onto the maxilla, alveolar process, and palate. In the mandible, the upper end of the condylar cartilages serve as growth centers bilaterally, serving the same function as the epiphysial plate in the long bone. This growth center in the mandible controls the ultimate vertical height of the ramus and the forward projection of the mandible. As stated previously, preservation of these growth centers in children during the resection of multicystic ameloblastoma is an important consideration for the ultimate cosmetic and functional rehabilitation of these patients. Of the many
726
PEDIATRIC MALIGNANCIES
reconstructive options available to the modern head and neck surgeon, two seem to offer the potential best morbidityto-benefit ratio in the pediatric population. First is free tissue transfer (fibula free flap) and second is distraction osteogenesis. The use of either one of these methods versus free iliac bone grafts should be individualized. The use of plate reconstruction has fallen out of favor.
AMELOBLASTIC CARCINOMA Introduction In 1974, Shafer et al. coined the term ameloblastic carcinoma.44 Until 1994, only 40 cases of ameloblastic carcinoma have been described in the English medical literature.45 This very rare tumor is defined as primary ameloblastoma where there has been histologic malignant transformation in association with less-differentiated metastatic growths, or where the features of ameloblastoma intermingles with those of carcinoma.46 Elzay proposed a definition for malignant ameloblastoma, which accounts for the more aggressive clinical tendencies of ameloblastoma, which have undergone a histologically and, ultimately, clinically aggressive transformation of their epithelial elements.47 This classification differs from the 1972 World Heath Organization classification system because it takes into account those primary lesions of the mandible or maxilla whose epithelium displays malignant transformation. Ameloblastic carcinoma is distinct from malignant ameloblastoma, which refers to lesions that present with distant or regional metastasis of histologically welldifferentiated elements.
Epidemiology The mean age of patients with ameloblastic carcinoma is 33 years, with a range of 4 to 75 years. Men are slightly more affected than women. Eighty percent of these malignant lesions are located in the mandible.46
Pathology Macroscopically, the cut surface of ameloblastic carcinoma is gray/white and smooth with a central cystic area. Histologically, ameloblastic carcinoma shows architectural features of ameloblastoma, but with malignant cytologic features, including nuclear pleomorphism, high nuclear-tocytoplasmic ratio, and increased mitotic activity. Peripheral palisading arrangement and acanthomatous differentiation may also be seen (see Figure 2). The central cystic area is partially lined with epithelium displaying squamous differentiation.
Biology Ameloblastic carcinoma is thought to arise from epithelial remnants within the mandible. This lesion, like all odontogenic carcinomas, is locally invasive and may metastasize to regional lymph nodes or distant sites. Genetic variances have long been suggested to contribute to the malignant behavior of this tumor.48 Muller et al. compared the DNA content of
Figure 2 Ameloblastic carcinoma. The tumor is composed of islands of basaloid cells with high nuclear-to-cytoplasmic ratio. Peripheral palisading is present (×200).
17 primary and 5 recurrent odontogenic tumors. The authors found that aneuploidy was significantly more common in ameloblastic carcinoma than in benign ameloblastoma, and was a strong predictor of malignant potential.35 Expression profiles employing DNA microarray have demonstrated upregulation and down-regulation of genetic sequences responsible for transcription, signaling transduction, cell cycle regulation, apoptosis, and differentiation.49 The identification of altered gene expression may help elucidate the biology of ameloblastic carcinoma, as well as identify potential genetic targets for therapy. Muller et al.35 compared the DNA content of 17 primary and 5 recurrent “benign” ameloblastomas and 5 ameloblastic carcinomas. Of the primary “benign” ameloblastomas, 14 (82%) were diploid; 3 (60%) of the 5 recurrent “benign” were diploid. No significant difference between the primary and the recurrent “benign” ameloblastomas was demonstrated. Of the five ameloblastic carcinomas, four (80%) were aneuploid. Flow cytometry was performed on three carcinomas; 100% concordance between image and flow cytometric data was seen. They found that aneuploidy was significantly more common in ameloblastic carcinoma than in “benign” ameloblastoma and is a strong predictor for malignant potential.
Patterns of Spread These tumors infiltrate locally and to distant sites of metastasis, which include the lungs, lymph nodes, chest wall, liver, peritoneum, brain, intestine, skull, and vertebrae. Both local/regional, as well as distant metastases, are prominent features of ameloblastic carcinoma.
Clinical Presentation/Evaluation Ameloblastic carcinoma is often clinically occult, except for the swelling. When a critical size is reached, pain and bleeding are the prominent clinical features.50 Determining the three-dimensional extent of ameloblastic carcinoma is essential prior to surgical resection to delineate
RARE PEDIATRIC MALIGNANCIES OF THE HEAD AND NECK
the tumor’s relationship with vascular and neural structures. CT scans offer both bone and soft tissue windows to allow determination of macroscopic tumor boundaries. MRI, although not as useful as CT for evaluation of bone, provides information regarding edge definition and tumor consistency, and is free from metal-induced artifacts.51
Differential Diagnosis The differential diagnosis of ameloblastic carcinoma should include primary intraalveolar epidermoid carcinoma and metastatic tumors from other sites including breast, lung, gastrointestinal tract, and salivary glands.
Treatment In ameloblastic carcinoma, surgery with negative margins and postoperative radiation therapy is the treatment of choice. In general, the three main modalities for surgical excision of odontogenic tumors are as follows: (i) enucleation, or curettage, (ii) marginal or partial resection, and (iii) composite resection. If cervical metastasis is present, surgical therapy should include a neck dissection. Our own experience and review of the literature suggests the following treatment paradigm in the management of ameloblastic carcinoma in the pediatric population. Surgical resection with negative margins remains a central dogma in the initial management of this disease followed by postoperative radiation therapy to the primary site and regional lymphatics. The clinically positive neck should clearly have a lymphadenectomy, and because of the aggressive nature of ameloblastic carcinoma, even the clinically negative neck should be treated with either primary surgical intervention or radiation therapy. The morbidity associated with surgical resection of this tumor remains of some concern to us, as previously stated – loss or disruption of craniofacial growth centers must be considered when planning treatment for pediatric patients. In the pediatric patient, we would prefer a modified neck dissection (sparing the sternocleidomastoid muscle, internal jugular vein, and spinal accessory nerve) and give only postoperative radiation therapy for pathologically positive disease. The advantage of this is to limit the radiation fields to only the primary site, sparing, if possible, treatment to critical growth centers of the cranial–facial skeleton. Adjuvant chemotherapy is a more controversial question. In our experience and review of the literature, distant failure has been the primary cause of demise of these patients. It makes intuitive sense that chemotherapy, which has been efficacious in diminishing the incidence of distant failure in other squamous cell cancers of the head and neck, might be of benefit in ameloblastic carcinoma. However, our own clinical experience and that reported in the literature leaves this as a provocative and unanswered question.
727
died 1 month to 2 years after metastases were first noted. Six of seven patients with anaplastic tumors (without verification of metastasis) died 1 year to 45 years after initial diagnosis of the tumor.46
SINONASAL UNDIFFERENTIATED CARCINOMA/ANAPLASTIC CARCINOMA OF THE PARANASAL SINUSES Introduction SNUC was first described in 198652 as a rare and highly aggressive malignancy arising in the nasal cavity and paranasal sinuses. Clinicopathologic information concerning this tumor has been sparse, mainly consisting of limited case series.53 – 55 This tumor remains an extremely difficult clinical challenge.
Epidemiology The mean age at diagnosis of SNUC is reported to be in the sixth decade of life. Although there have been only a small number of series reported, there appears to be no sex predilection.
Pathology Characteristics of SNUC include a hypercellular proliferation with trabecular, sheetlike, ribbon, lobular, and organoid growth patterns. The cells vary in size from medium-to-large and are round-to-oval in shape. They contain pleomorphic and hyperchromatic nuclei, and have various amounts of eosinophilic cytoplasm with a high nuclear-to-cytoplasmic ratio. Mitotic activity is infrequent and atypical mitoses, tumor necrosis, and apoptosis are observed.56 There is necrosis of both the individual tumor cells and the central areas of tumor islands. There may also be tumor involvement of vascular channels or perineural invasion (see Figure 3). Immunocytochemical staining for cytokeratin and epithelial membrane antigen is positive. NSE may be positive. The staining for S-100 protein and leukocyte common antigen are negative.
Prognosis The prognosis for ameloblastic carcinoma is poor.35 In a study of 25 cases of ameloblastic carcinoma, seven of nine patients with less-differentiated or dedifferentiated tumors
Figure 3 Undifferentiated carcinoma. The tumor cells are large and pleomorphic with vesicular chromatin and prominent nucleoli (×400).
728
PEDIATRIC MALIGNANCIES
Biology SNUC of the paranasal sinuses is locally highly aggressive, causing bone destruction and invasion of adjacent structures. Undifferentiated carcinoma of the nasopharynx has a wellknown association with the Epstein-Barr virus (EBV), but only an inconsistent relationship has been identified in morphologically similar undifferentiated carcinomas occurring at other sites. The possible association of EBV with SNUC has recently been investigated.57 Using in situ hybridization techniques, 22 cases of pathologically proven SNUC were investigated for the presence of EBV ribonucleic acid (RNA). EBV RNA was identified in 7 of 11 SNUCs from Asian patients, but in none of those from Western patients. These authors concluded that genetic predisposition or environmental and geographical cofactors play an important role in determining the strength of the association of SNUC with EBV. The presence of the virus may play a role in the tumorigenesis in this neoplasm. The absence of EBV in SNUCs from several Asian patients and in those from all cases of Western patients indicates that other, yet unidentified, pathogenetic factors must also be operative.
Patterns of Spread Presentation with extensive involvement of the nasal cavity and paranasal sinuses is common. Direct invasion can occur into the orbit, cribriform plate, infratemporal fossa, pterygoid region, and anterior cranial fossa. Cervical metastasis has been reported, but is generally a rare occurrence.55
Clinical Presentation SNUC typically produces few symptoms despite extensive disease mostly because of the “silent anatomic spaces” of the paranasal sinuses, which permit development of large neoplasms without focused symptomatology. It is characterized by extensive sinonasal tissue destruction, with frequent involvement of the orbit and anterior cranial fossa. The most common initial symptoms are facial pain, nasal obstruction, proptosis, and epistaxis. Cranial nerve palsies and diplopia may occur. Rarely, altered mental status or enlargement of the jugulodigastric node may be an initial presenting symptom. Virtually all neoplasms involve the nasal, maxillary, and ethmoid complexes at the time of presentation.
Evaluation CT and MRI best perform radiographic evaluation of tumors of the paranasal sinuses. A recent study by Phillips et al. retrospectively studied 11 patients with histopathologically proven SNUC.58 All 11 patients underwent CT and 6 underwent MR imaging. On contrast-enhanced CT scans, all tumors enhance to varying degrees. These tumors tend to be noncalcified and often cause sinus obstruction. MRI signal intensity of the lesions is isointense to muscle on T1-weighted images and iso- to hyperintense on T2 images. A heterogeneous enhancement of tumors is seen in gadolinium-enhanced images. This study concluded that
SNUC could not be distinguished from other tumors of this region on the basis of imaging features.
Differential Diagnosis The differential diagnoses of SNUC are melanoma, malignant lymphoma, olfactory neuroblastoma, rhabdomyosarcoma, neuroendocrine carcinoma, and lymphepithelioma. It is important not to confuse olfactory neuroblastoma with SNUC because the former has a better prognosis. Olfactory neuroblastoma, unlike SNUC, is typically composed of small uniform cells with sparse cytoplasm and a small round nucleus without a large nucleolus. Homer Wright rosettes with background neurofibrillary processes are typically seen with olfactory neuroblastoma. SNUC has larger pleomorphic cells with prominent nucleoli, a greater number of mitotic figures, extensive necrosis, and vascular invasion. Despite morphologic similarities with other nasopharyngeal carcinoma and SNUC, one major clinical difference is in the presentation. Nasopharyngeal carcinoma often presents with regional lymph node metastasis and an occult primary lesion, while SNUC usually presents as bulky, locally destructive disease.
Treatment and Prognosis There are only a few centers where a series of managed patients have been reported. A recent review of the data on the management of SNUC has revealed treatment trends that may improve local control and survival. The report confirms the use of aggressive multimodality treatment in the management of these lesions. The study points to the improved disease-free survival with the combined use of surgical excision, chemotherapy with a platinumbased agent, and the administration of radiation therapy.59 Patients were found to benefit from systemic therapy given in either the adjuvant or neoadjuvant setting; however, both were dependent on achieving appropriate local control. An evaluation of a standardized treatment approach at the University of Virginia revealed 4 of 10 patients treated with the intent of cure, by means of neoadjuvant chemotherapy and craniofacial resection, to be free of disease at 4, 36, 49, and 164 months following surgery. The authors also found that in those cases of advanced disease thought to be unresectable, palliative chemotherapy may still be of benefit to the patient.60 The use of intraarterial chemotherapy in conjunction with radiation and surgery has been explored in patients with advanced sinonasal malignancies with favorable results.61 Direct application of this technique to SNUC has been limited, and further prospective studies will be required to determine its effectiveness. We advocate a combined modality approach, particularly preoperative chemotherapy and radiation therapy, followed by radical surgical resection, which in select cases has the best chance at local control in this lethal malignancy. We believe that this regimen offers the best chance at local control in this highly lethal malignancy.
RARE PEDIATRIC MALIGNANCIES OF THE HEAD AND NECK
ESTHESIONEUROBLASTOMA Introduction Esthesioneuroblastoma, also known as olfactory neuroblastoma, is an extremely rare cancer infrequently occurring in the pediatric population. French physicians Berger and Luc first described the tumor in 1924.62 Esthesioneuroblastoma is thought to be of neural crest origin and arises from the olfactory mucosa. It accounts for approximately 3% of all nasal epithelial neoplasms.63
Epidemiology Esthesioneuroblastoma is very rare in the United States, but may have higher incidence elsewhere. Although still rare, the rate of disease occurrence has seen an increase in the last 20 years. This most likely represents improved pathological recognition, rather than a rising incidence of the disease. Males and females are affected with similar frequency. There are two peaks of age occurring at approximately 15 and 55 years.63
Pathology Esthesioneuroblastoma falls into the category of “small round blue cell tumors”. The tumor typically is composed of irregular nests of small hyperchromatic cells separated by fibrous septa. Also commonly seen are Flexner-type rosettes and Homer Wright–type pseudorosettes. Electron microscopy reveals fibrils within the cytoplasm.64,65
Biology
729
Further tumor extension can result in proptosis, extraocular muscle paralysis, blindness, or intracranial complications.
Evaluation CT is used as the initial radiographic test, and is essential in the staging of the disease. The tumor appears as a homogeneous enhancing mass. Evidence of tumor extending into surrounding bony structures should be carefully sought. MRI can be helpful to further evaluate sinonasal, intraorbital, or intracerebral extension. Most patients will require both CT and MRI.69 Following radiographic assessment, tissue is obtained by biopsy and submitted for regular staining and immunohistochemistry. No one staging system has been universally accepted for esthesioneuroblastoma. Kadish developed the first system for staging esthesioneuroblastoma. In this system, tumors are classified as A: limited to the nasal fossa, B: extending to the paranasal sinuses, or C: extending beyond the paranasal sinuses. A more recent Tumor-Node-Metastasis (TNM) staging system has also been developed. T1 tumors involve the nasal cavity and/or paranasal sinuses, excluding the sphenoid and superior ethmoidal cells. T2 tumors involve the nasal cavity and/or paranasal sinuses, including the sphenoid, with extension to or erosion of the cribriform plate. T3 lesions extend into the orbit or protrude into the anterior cranial fossa without dural invasion. T4 tumors directly involve the brain. Nodal classification under the TNM is as follows: N0 for no cervical lymph node metastasis and N1 for the presence of cervical node metastasis. The presence or absence of metastases is delineated M1 or M0 respectively.
Differential Diagnosis
Contrary to the name, esthesioneuroblastoma does not share the genetic abnormalities seen in neuroblastoma. Previous studies had included esthesioneuroblastoma in the Ewing sarcoma family of tumors.66 The reciprocal translocation t(11;22)(q24;q12) is found in esthesioneuroblastoma, and is indistinguishable from that seen in Ewing’s sarcoma. Newer findings, however, separate esthesioneuroblastoma from the Ewing sarcoma family of tumors, noting a lack of the fusion gene product EWS/FLI1 seen in Ewing’s sarcoma and PNET.67,68
Esthesioneuroblastoma is distinguished from melanoma by its negative UMB 45 and sparse pattern of S-100 staining. In comparison with SNUC, esthesioneuroblastoma lacks mitotic activity, necrosis, vascular invasion, and glandular differentiation. Immunohistochemistry also differs by the lack of cytokeratin antibody staining. Recent findings suggest that hASH1 mRNA may represent another means of distinguishing esthesioneuroblastoma from other poorly differentiated tumors of the sinonasal region.70
Patterns of Spread
Treatment
The lesion typically starts at the superior portion of the nasal cavity. Tumor progression is in the submucosal plane, and can involve the paranasal sinuses and nasal cavity, or can cross the cribriform plate and directly invade the brain. Tumor growth and progression varies from slow to rapid. Metastasis is via lymphatic and hematogenous routes. Unfortunately, by the time of diagnosis most tumors have already spread extensively.
The mainstay of treatment has been complete surgical excision. Additionally, aggressive multimodality approaches utilizing surgery and adjuvant radiation therapy have yielded favorable outcomes.71,72 The use of combination chemotherapy has also demonstrated efficacy.73 Recent protocols have evaluated combinations using vincristine, doxorubicin, cyclophosphamide, ifosfamide, cisplatin, and etoposide. A recent study evaluated the role of multimodal therapy specifically in pediatric and adolescent patients undergoing treatment for esthesioneuroblastoma. The authors reported a 5-year overall survival of 73%, and 5-year event-free survival of 55%. Patients with advanced lesions, in Kadish stage C, had more favorable outcomes when managed with a combination of surgery, chemotherapy, and radiation compared to local therapy with surgery and radiation alone
Clinical Presentation The tumor presents clinically as a polypoid mass, which when biopsied can bleed profusely. Symptoms are not specific and typically include nasal obstruction, epistaxis, facial pain, headache, swelling, and sinusitis. Many diagnoses are unexpected pathological findings at the time of sinus surgery.
730
PEDIATRIC MALIGNANCIES
(65 vs 20% respectively).74 Another study demonstrated the efficacy of high-dose chemotherapy and autologous bone marrow transplantation in achieving survival in the pediatric population.75
7.
8.
Prognosis If treated early, a favorable outcome can be achieved. However, advanced disease can often be fatal. Recent literature reports 5-year survival rates between 50 and 80%. In one retrospective review, multimodality therapy yielded overall 5-year and 10-year survivals of 89 and 81%, respectively.76 The pattern of survival follows the extent of disease involvement, with decreased survival observed in disease extending beyond the nasal cavity and with distant spread. Metastases to cervical lymph nodes are most common, followed by metastases to the lung, abdominal viscera, and bone. The presence of cervical lymph node involvement has been proven to be a highly negative prognostic indicator.77 The rate of recurrence ranges from 20 to 40%, and recurrent disease has been observed greater than 10 years after treatment. The distant recurrence rate is less than 10%; however, it carries an unfavorable prognosis. Multiple studies support histopathological grade as another important prognostic indicator.78 – 80
SUMMARY
9. 10.
11.
12. 13. 14.
15.
16. 17. 18.
Malignant neoplasms in pediatric population are relatively rare diseases, and approximately 10% of these lesions occur in the head and neck region. EOE, malignant ameloblastoma, ameloblastic carcinoma, SNUC, and esthesioneuroblastoma are rare head and neck malignancies in the pediatric age-group. In general, the treatment strategies applicable to pediatric head and neck malignancies are similar to those of adults with the same neoplastic process. These include a combination of surgery and radiation therapy, and, in some instances, chemotherapy. A multidisciplinary team is necessary to ensure a well-planned therapeutic approach. Improved understanding of the disease processes and more effective treatments have led to increased overall prognosis in these patients.
19.
REFERENCES
27.
1. Albright JT, Topham AK, Reilly JS. Pediatric head and neck malignancies: US incidence and trends over 2 decades. Arch Otolaryngol Head Neck Surg 2002; 128: 655 – 9. 2. Tefft M, Vawter GF, Mitus A. Paravertebral “round cell” tumors in children. Radiology 1969; 92: 1501 – 9. 3. Angervall L, Enzinger FM. Extraskeletal neoplasm resembling Ewing’s sarcoma. Cancer 1975; 36: 240 – 51. 4. Raney RB, et al. Ewing’s sarcoma of soft tissues in childhood. A report from the Intergroup Rhabdomyosarcoma Study, 1972 – 1991. J Clin Oncol 1997; 15: 574 – 82. 5. Shimida H, et al. Pathologic features of extraosseous Ewing’s sarcoma: a report from the Intergroup Rhabdomyosarcoma Study. Hum Pathol 1988; 19: 442 – 53. 6. Horowitz ME, et al. Ewing’s sarcoma family of tumors. Ewing’s sarcoma of bone and soft tissue and the peripheral primitive neuroectodermal tumors. In Pizzo PA, Poplack DG (eds) Principles and
28.
20. 21.
22. 23. 24.
25.
26.
29. 30. 31. 32.
33.
34.
Practice of Pediatric Oncology. Philadelphia, Pennsylvania: LippincottRaven, 1997: 831 – 863. Folpe AL, et al. Morphologic and immunophenotypic diversity in Ewing family tumors: a study of 66 genetically confirmed cases. Am J Surg Pathol 2005; 29: 1025 – 33. Batsakis JG, Mackay B, El Naggar AK. Ewing’s sarcoma and peripheral primitive neuroectodermal tumor: an interim report. Ann Otol Rhinol Laryngol 1996; 105: 838 – 43. Siegal GP, et al. Primary Ewing’s sarcoma involving the bones of the head and neck. Cancer 1987; 60: 2829 – 40. Marcus RB, Post JC, Mancuso AA. Pediatric tumors of the head and neck. In Million RR, Cassisi NJ (eds) Management of Head and Neck Cancer: A Multidisciplinary Approach. Philadelphia, Pennsylvania: JB Lippincott Company, 1994. Gustafson R, Maragos N, Reiman H. Extraskeletal Ewing’s sarcoma occurring as a mass in the neck. Otolaryngol Head Neck Surg 1982; 90: 491. Wilkins RM, et al. Ewing’s sarcoma of bone. Experience with 140 patients. Cancer 1986; 58: 2551. Dehner LP. Primitive neuroectodermal tumor and Ewing’s sarcoma. Am J Surg Pathol 1993; 17: 1 – 13. Marcus RB, et al. Local control and function after twice-a-day radiotherapy for Ewing’s sarcoma of bone. Int J Radiat Oncol Biol Phys 1991; 21: 1509 – 15. Dalal S, et al. Vascular endothelial growth factor: a therapeutic target for tumors of the Ewing’s sarcoma family. Clin Cancer Res 2005; 11: 2364 – 78. Kutluk MT, et al. Treatment results and prognostic factors in Ewing sarcoma. Pediatr Hematol Oncol 2004; 21: 597 – 610. Koscielniak E, Morgan M, Treuner J. Soft tissue sarcoma in children: prognosis and management. Paediatr Drugs 2002; 4: 21 – 8. Grier HE, et al. Addition of ifosfamide and etoposide to standard chemotherapy for Ewing’s sarcoma and primitive neuroectodermal tumor of bone. N Engl J Med 2003; 348: 694 – 701. Chow E, et al. Cutaneous and subcutaneous Ewing’s sarcoma: an indolent disease. Int J Radiat Oncol Biol Phys 2000; 46: 433 – 8. Witterick IJ, et al. Malignant ameloblastoma. Am J Otolaryngol 1996; 17: 122 – 6. Becelli R. et al. Mandibular ameloblastoma: analysis of surgical treatment carried out in 60 patients between 1977 and 1998. J Craniofac Surg 2002; 13: 395 – 400. Rosenstein T, et al. Cystic ameloblastoma – behavior and treatment of 21 cases. J Oral Maxillofac Surg 2001; 59: 1311 – 6. Ord RA, et al. Ameloblastoma in children. J Oral Maxillofac Surg 2002; 60: 762 – 70. Ladeinde AL, et al. Odontogenic tumors: a review of 319 cases in a Nigerian teaching hospital. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2005; 99: 191 – 5. Kahn MA. Ameloblastoma in young persons. A clinicopathologic analysis and etiologic investigation. Oral Surg Oral Med Oral Pathol 1989; 67: 706 – 15. Piattelli A, et al. Expression of proliferating cell nuclear antigen in ameloblastomas and odontogenic cysts. Oral Oncol 1998; 34: 408 – 12. Vered M, et al. Myofibroblasts in stroma of odontogenic cysts and tumors can contribute to variations in the biological behavior of lesions. Oral Oncol 2005; 41(10): 1028 – 33 [Epub 2005 Aug 31]. Pinheiro JJ, et al. Local invasiveness of ameloblastoma: role played by matrix metalloproteinases and proliferative activity. Histopathology 2004; 45: 65 – 72. Schweitzer FC, Barnfield WF. Ameloblastoma of the mandible with metastasis to the lungs: Report of a case. J Oral Surg 1943; 1: 287 – 95. Azumi T, et al. Malignant ameloblastoma with metastasis to the skull: Report of a case. J Oral Surg 1981; 39: 690 – 6. Laughlin EH. Metastasizing ameloblastoma. Cancer 1989; 64: 776. Buff SJ, et al. Pulmonary metastasis from ameloblastoma of the mandible: Report of a case and review of the literature. J Oral Surg 1980; 38: 374 – 6. Madiedo G, Choi H, Kleinman JG. Ameloblastoma of the maxilla with distant metastases and hypercalcemia. Am J Clin Pathol 1981; 4: 585 – 91. Kunze E, et al. Biology of metastasizing ameloblastoma. Pathol Res Pract 1985; 180: 526 – 35.
RARE PEDIATRIC MALIGNANCIES OF THE HEAD AND NECK 35. Muller S, DeRose PB, Cohen C. DNA ploidy of ameloblastoma and ameloblastic carcinoma of the jaws: analysis of image and flow cytometry. Arch Pathol Lab Med 1993; 117: 1126 – 31. 36. Nakamura N, et al. Comparison of long-term results between different approaches to ameloblastoma. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2002; 93: 13 – 20. 37. D’Agostino A, et al. Retrospective evaluation on the surgical treatment of jaw bones ameloblastic lesions. Experience with 20 clinical cases. Minerva Stomatol 2001; 50: 1 – 7. 38. Bataineh AB. Effect of preservation of the inferior and posterior borders on recurrence of ameloblastomas of the mandible. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2000; 90: 155 – 63. 39. Houston G, et al. Malignant (metastatic) ameloblastoma: report of a case. J Oral Maxillofac Surg 1993; 51: 1152 – 5. 40. Gardner DG. Radiotherapy in the treatment of ameloblastoma. Int J Oral Maxillofac Surg 1988; 17: 201 – 5. 41. Atkinson CH, Harwood AR, Cummings BJ. Ameloblastoma of the jaw: a reappraisal of the role of megavoltage irradiation. Cancer 1984; 53: 869 – 73. 42. Pandya NJ, Stuteville OH. Treatment of ameloblastoma. Plast Reconstr Surg 1972; 50: 242 – 8. 43. Ramadas K, et al. Pulmonary metastasis form ameloblastoma of the mandible treated with cisplatin, adriamycin, and cyclophosphamide. Cancer 1990; 66: 1475. 44. Shafer WG, Hine MK, Levy BM. A Textbook of Oral Pathology, 3rd ed. Philadelphia, Pennsylvania: WB Saunders Co, 1974: 236 – 284. 45. Lee L, Maxymiw WG, Wood RE. Ameloblastic carcinoma of the maxilla metastatic to the mandible: case report. J Craniomaxillofac Surg 1990; 18: 247 – 50. 46. Slootweg PJ, Muller H. Malignant ameloblastoma or ameloblastic carcinoma. Oral Surg Oral Med Oral Pathol 1984; 57: 168 – 76. 47. Elzay RP. Primary intraossseous carcinoma of the jaw: review and update of odontogenic carcinomas. Oral Surg Oral Med Oral Pathol 1982; 54: 299 – 303. 48. Nodit L, et al. Allelic loss of tumor suppressor genes in ameloblastic tumors. Mod Pathol 2004; 17: 1062 – 7. 49. Carinci F, et al. Expression profiling of ameloblastic carcinoma. J Craniofac Surg 2004; 15: 264 – 9. 50. Lolachi CM, Madan SK, Jacobs JR. Ameloblastic carcinoma of the maxilla. J Laryngol Otol 1995; 109: 1019 – 22. 51. Williams TP. Management of ameloblastoma. J Oral Maxillofac Surg 1993; 51: 1064 – 70. 52. Frierson HF, et al. Sinonasal undifferentiated carcinoma. An aggressive neoplasm derived from schneiderian epithelium and distinct from olfactory neuroblastoma. Am J Surg Pathol 1986; 10: 771 – 9. 53. Kim BS, Vongtama R, Juillard G. Sinonasal undifferentiated carcinoma: case series and literature review. Am J Otolaryngol 2004; 25: 162 – 6. 54. Gorelick J, et al. Sinonasal undifferentiated carcinoma: case series and review of the literature. Neurosurgery 2000; 47: 750 – 4. 55. Righi PD, et al. Sinonasal undifferentiated carcinoma: a 10 year experience. Am J Otolaryngol 1996; 17: 167 – 71. 56. Ejaz A, Wenig BM. Sinonasal undifferentiated carcinoma: clinical and pathologic features and a discussion on classification, cellular differentiation, and differential diagnosis. Adv Anat Pathol 2005; 12: 134 – 43. 57. Lopategui JR, et al. Detection of Epstein-barr viral RNA in sinonasal undifferentiated carcinoma from Western and Asian patients. Am J Surg Pathol 1994; 18: 391 – 8. 58. Phillips CD, et al. Sinonasal undifferentiated carcinoma. CT and MR imaging of an uncommon neoplasm of the nasal cavity. Radiology 1997; 202: 477 – 80.
731
59. Enepekides DJ. Sinonasal undifferentiated carcinoma: an update. Curr Opin Otolaryngol Head Neck Surg 2005; 13: 222 – 5. 60. Musy PY, Reibel JF, Levine PA. Sinonasal undifferentiated carcinoma: the search for a better outcome. Laryngoscope 2002; 112: 1450 – 5. 61. Madison Michael L 2nd et al. The treatment of advanced sinonasal malignancies with pre-operative intra-arterial cisplatin and concurrent radiation. J Neurooncol 2005; 72: 67 – 75. 62. Berger L, Luc H, Richard R. L’esthesioneuro epitheliome olfac-tif. Bull Assoc Fr Etud Cancer 1924; 13: 410 – 20. 63. Broich G, Pagliari A, Ottaviani F. Esthesioneuroblastoma: a general review of the cases published since the discovery of the tumor in 1924. Anticancer Res 1997; 17: 2683 – 706. 64. Min KW. Usefulness of electron microscopy in the diagnosis of “small” round cell tumors of the sinonasal region. Ultrastruct Pathol 1995; 19: 347 – 63. 65. Du ZM, Li YS, Wang BF. Electron microscopic and immunohistochemical findings in a case of olfactory neuroblastoma. J Clin Pathol 1993; 46: 83 – 5. 66. Sorensen PH, et al. Olfactory neuroblastoma is a peripheral primitive neuroectodermal tumor related to Ewing sarcoma. Proc Natl Acad Sci U S A 1996; 93: 1038 – 43. 67. Kumar S, et al. Absence of EWS/FLI1 fusion in olfactory neuroblastomas indicates these tumors do not belong to the Ewing’s sarcoma family. Hum Pathol 1999; 30: 1356 – 60. 68. Argani P, Perez-Ordonez B, Xiao H. Olfactory neuroblastoma is not related to the Ewing family of tumors: absence of EWS/FLI1 gene fusion and MIC2 expression. Am J Surg Pathol 1998; 22: 391 – 8. 69. Oskouian RJ Jr, et al. Esthesioneuroblastoma: clinical presentation, radiological, and pathological features, treatment, review of the literature, and the University of Virginia experience. Neurosurg Focus 2002; 12: e4. 70. Paulette Mhawech MD, et al. Human achaete-scute Homologue (hASH1) mRNA level as a diagnostic marker to distinguish Esthesioneuroblastoma from poorly differentiated tumors arising in the sinonasal tract. Am J Clin Pathol 2004; 122: 100 – 5. 71. Irish J, et al. Outcome and analysis of the surgical management of esthesioneuroblastoma. J Otolaryngol 1997; 26: 1 – 7. 72. Eden BV, Debo RF, Larner JM. Esthesioneuroblastoma. Long-term outcome and patterns of failure – the University of Virginia experience. Cancer 1994; 73: 2556 – 62. 73. Bhattacharyya N, Thornton AF, Joseph MP. Successful treatment of esthesioneuroblastoma and neuroendocrine carcinoma with combined chemotherapy and proton radiation. Results in 9 cases. Arch Otolaryngol Head Neck Surg 1997; 123: 34 – 40. 74. Eich HT, et al. Esthesioneuroblastoma in childhood and Adolescence Better Prognosis with Multimodal Treatment? Strahlenther Onkol 2005; 181: 378 – 84. 75. Nguyen QA, et al. Esthesioneuroblastoma in the pediatric age-group: the role of chemotherapy and autologous bone marrow transplantation. Int J Pediatr Otorhinolaryngol 1996; 37: 45 – 53. 76. Diaz EM Jr, et al. Olfactory neuroblastoma: the 22-year experience at one comprehensive cancer center. Head Neck 2005; 27: 138 – 49. 77. Koka VN, et al. Aesthesioneuroblastoma. J Laryngol Otol 1998; 112: 628 – 33. 78. McElroy EA, Buckner JC, Lewis JE Jr. Chemotherapy for advanced esthesioneuroblastoma: the Mayo Clinic experience. Neurosurgery 1998; 42: 1023 – 7. 79. Foote RL, et al. Esthesioneuroblastoma: the role of adjuvant radiation therapy. Int J Radiat Oncol Biol Phys 1993; 15: 835 – 42. 80. Dias FL, et al. Patterns of failure and outcome in Esthesioneuroblastoma. Arch Otolaryngol Head Neck Surg 2003; 129: 1186 – 92.
Section 11 : Pediatric Malignancies
67
Uncommon Pediatric Tumors of the Thorax Joanne M. Hilden, Sharon O. Meerbaum and Louis P. Dehner
INTRODUCTION Although common in the adult population, primary pulmonary malignancies in the age-group less than 20 years old at diagnosis are quite rare.1,2 The most common lung neoplasms in children are, in fact, metastases from other primary tumors3 (see Figure 1). The purpose of this chapter is to discuss a group of neoplasms of diverse histogenetic types, which have in common their origin from intrathoracic organs or structures or the chest wall, and have been described in children. Most of these neoplasms are malignant, although the clinical behavior is dependent upon the specific tumor type and the pathologic stage. Several of the tumor types are diagnosed predominately in children. One notable example is the pleuropulmonary blastoma (PPB) to be described later in this chapter. The epidemiology of primary pulmonary neoplasms in children is not well documented when compared to the established associations in adults of carcinomas and mesotheliomas to environmental factors such as tobacco smoke and asbestos exposure. Children with primary or secondary immunodeficiency syndromes and neurofibromatosis occasionally develop pulmonary or mediastinal neoplasms of a lymphoproliferative and/or sarcomatous nature.4 A predisposition to PPB in children from PPB-associated families has been described.5 Other than these few associations and relationships, epidemiologic factors have yet to be elucidated. Presenting symptomatology of an intrathoracic tumor in a child is generally the consequence of the direct mass effect or airway obstruction. Cough, dyspnea, chest pain, fever, and superior vena cava syndrome are some of the more common clinical manifestations. A minority of neoplasms are associated with digital clubbing as seen in intrathoracic lymphomas,6 spinal cord compression with associated limb weakness or bowel/bladder dysfunction in neurogenic or vertebral malignancies,7 or autoimmune phenomena in mediastinal lymphoreticular malignancies or thymic neoplasms.8
Because most pulmonary neoplasms produce a mass, the tumor is often initially recognized in a standard chest radiograph in a child with suspected respiratory tract infection or pneumothorax.9 If the tumor is asymptomatic, the identification of a mass may be entirely incidental, which is the experience in some cases of inflammatory myofibroblastic tumor (IMT), also known as inflammatory pseudotumor and plasma cell granuloma.10 Some metastatic lesions to the lung and PPB may have a combination of cystic lesions in the lung and a pneumothorax. Diffuse reticulonodular infiltrates in the parenchyma with or without mediastinal adenopathy, or a mass(es) in an immunosuppressed child should alert one to the possibility of a lymphoproliferative disorder, smooth muscle neoplasm, or an infectious process.11 A similar presentation is seen in children with Langerhans cell histiocytosis or juvenile xanthogranuloma with lung involvement. Computed tomography (CT) allows for more exact localization of the lesion(s) and improves the sensitivity for metastatic disease.12 Magnetic resonance imaging provides additional refinements about tissue densities, an opportunity for multiple views, and clear images of the vertebral column and neural foramina. Ultrasonography, fluoroscopy, gallium radionuclide scanning, and angiography can also be of diagnostic value. A histologic diagnosis is necessary for nearly all pulmonary neoplasms, with the exception of lymphoreticular malignancies diagnosed in the blood or bone marrow. Flexible, fiberoptic bronchoscopy with biopsy is sometimes carried out before definitive surgery. Communication between specialists such as the surgeon, pulmonologist, hematologist/oncologist, and pathologist is essential for comprehensive preoperative planning. A formal thoracotomy is the standard diagnostic approach,13 and with pediatric cancer protocols, risk-assigning according to tumor characteristics, it is often necessary to obtain sufficient tumor tissue for complete evaluation. Consultation with a pediatric surgeon will help determine whether a minimally invasive approach is feasible.14
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
UNCOMMON PEDIATRIC TUMORS OF THE THORAX
Figure 1 Wilm’s tumor metastatic to the lung showing nodules of neoplastic blastoma in the interstitium. H&E, ×200.
MEDIASTINUM Most mediastinal tumors in children are malignant (60–70% of all cases).15 Approximately 40% of all mediastinal tumors in the pediatric age-group are Hodgkin’s or non-Hodgkin’s lymphomas (see Figure 2), and 20–25% of neoplasms are neuroblastic tumors.16 – 18
733
Lymphoreticular malignancies in children often present in the anterior mediastinum, particularly the thymus. Thymicbased tumors other than the lymphoproliferative malignancies are notably uncommon in children.19 Only 1–2% of mediastinal tumors in children are derived from the epithelial component of the thymus that defines a thymoma.16,17 Although some reports suggest that thymomas in children are more aggressive neoplasms than the adult counterparts, others have reported a generally favorable outcome for thymomas in children.20 – 22 Thymomas are rare in children, but nonetheless well documented in the first two decades of life.19,23 Thymic hyperplasia is in the differential diagnosis when thymic enlargement is discovered.24,25 The most consistently reliable feature that differentiates a “benign” from a “malignant” thymoma is the presence or absence of capsular invasion (see Figures 3 and 4). The invasiveness of a thymoma rather than its subtype is the more significant factor in prognosis in most cases. Complete surgical resections, when possible, are the treatment of choice.19,26,27 Those thymic neoplasms with the features of a carcinoid or one with overtly carcinomatous features behave in a malignant fashion, without particular regard to the local manifestations.28 The World Health Organization (WHO) classification system for thymomas, developed in 1999, was evaluated in two studies (273 and 132 patients); the authors concluded that this classification system correlated with clinical behavior.29,30 When complete resection is not feasible in the presence of invasive thymoma in contiguous structures, radiotherapy has been utilized, with more extensive experience in adults.31,32 It occasionally metastasizes widely to the lung, liver, lymph nodes, and bones. Complete resection is the treatment of choice. Radiotherapy has been used more extensively in adults than in children, although Kaplinsky et al.33 reported a child who received combination radiochemotherapy and is alive 7 years after treatment. Given the high fatality rates in children, combination chemotherapy
(a)
(b) Figure 2 Lymphoblastic lymphoma presenting as an enlarging anterior mediastinal mass in a 16-year-old male. A mediastinal biopsy showing an infiltrate of malignant thymic lymphoblasts with convoluted and nonconvoluted nuclei (a). An immunoperoxidase stain for CD99 or 013, the antibody to the MIC2 gene product showing a uniform membrane reaction (b). The tumor cells were also immunopositive for CD45RO. H&E, ×400; Immunoperoxidase, ×400.
Figure 3 CT image of thymoma presenting as an anterior mediastinal mass in a 12-year-old boy.
734
PEDIATRIC MALIGNANCIES
(a)
(b)
(c)
Figure 4 Encapsulated thymoma in a 12-year-old boy (see Figure 3), showing a prominent fibrous capsule (a), the predominant lymphocytic component (b) and dilated perivascular spaces (c). H&E, ×100, ×400, ×100.
with platinum-containing regimens should be considered since remissions have been reported in adults and children with this therapy.33 – 36 There are known associations between thymoma and myasthenia gravis,37 and pure red cell aplasia.38 Lymphoepithelioma of the thymus, resembling nasopharyngeal carcinoma, has an association with Epstein-Barr virus.39 – 41 Carcinoid tumor of the thymus is uncommon in any agegroup, but is particularly rare in children, with only a few cases described.42 – 46 It is highly aggressive, with a propensity to metastasize, and can be associated with Cushing’s syndrome and other endocrine symptoms.45 – 47 Reported pediatric cases reveal a high mortality rate, although therapy has not always been well described.42 – 45 Complete resection, if possible, is the treatment of choice.46,48 Chemotherapy and radiotherapy have not been shown to be effective in any agegroup.48 Germ cell neoplasms, although rare, account for nearly 10% of mediastinal masses in children. The mediastinum is the second most common extragonadal site for such neoplasms, exceeded only by the sacrococcygeal region.49 These
tumors occur almost exclusively in the anterior mediastinum, where they comprise almost 25% of all masses in that specific compartment.50 Most germ cell neoplasms of the anterior mediastinum (75–80% of cases) are mature teratomas, whose predominantly cystic or mixed cystic-and-solid areas are composed of mature tissues of ectodermal, endodermal, and mesodermal types.51,52 Unlike germ cell neoplasms in adults, which are often characterized by chromosomal gain of 12q, imbalances in chromosome 1, deletions of 4q and 6q and gains of 20q are commonly found in childhood.53 The remaining germ cell neoplasms of the mediastinum are malignant as defined by one or more of the following histopathologic subtypes: germinoma (seminoma), embryonal carcinoma, yolk sac carcinoma (endodermal sinus tumor), choriocarcinoma, immature teratoma, and nongerminal sarcomas.52 These neoplasms often have residual teratomatous elements in addition to the malignant components. In females, yolk sac carcinoma is often the only malignant element, whereas in males, mixed malignant patterns are commonly seen.54 Mediastinal malignant germ cell neoplasms occur overwhelmingly in males between the ages of 12 and 35 years.49,55 There is an apparent increased risk for mediastinal germ cell neoplasms in Klinefelter syndrome.56 Chest pain, hemoptysis, respiratory distress, and/or superior vena cava syndrome are the presenting clinical manifestations. If the tumor has a component of choriocarcinoma, gynecomastia may be present as well. Serum α-fetoprotein and/or human chorionic gonadotropin are invariably elevated at diagnosis. Since the mediastinum is a common site for metastases from a testicular germ cell neoplasm, the possibility of an occult primary tumor in the testis should not be overlooked in the clinical evaluation.57 Germinoma (seminoma) is the most common subtype of malignant germ cell tumor of the mediastinum; 10% of cases are diagnosed between the ages of 10 and 19 years.58 These tumors may be cystic and have a prominent lymphocytic component that may resemble Hodgkin’s lymphoma, thymoma, or large B cell lymphoma. Complete resection is the treatment of choice, and the prognosis correlates with extent of disease. Other malignant patterns include yolk sac carcinoma (endodermal sinus tumor) and choriocarcinoma.59,60 Mixed histologic patterns of malignant germ cell neoplasms occur in the mediastinum with combinations of embryonal carcinoma, teratoma (often immature) and yolk sac carcinoma. Immature teratoma behaves in a malignant fashion when diagnosed in an adolescent or young adult, in contrast to the benign behavior of a similar-appearing tumor in an infant or young child (see Figure 5). Sarcomatous transformation in the form of angiosarcoma, rhabdomyosarcoma, or a poorly differentiated hematopoietic malignancy with features of myelomonocytic leukemia is rare, but a well-documented phenomenon in mediastinal germ cell neoplasms (see Figure 6).61 Complete surgical resection and cisplatin-based chemotherapy have yielded the best results in the management of these tumors. When the tumor is unresectable, the diagnosis is confirmed by a biopsy that may reveal only necrosis or a less than representative sample of the total tumor. The treatment decision will then require
UNCOMMON PEDIATRIC TUMORS OF THE THORAX
735
Figure 5 Immature teratoma of the anterior mediastinum, showing an embryonic or primitive-appearing neural canal with palisading neuroblast-like cells. Fibrous stroma and cartilage are also present. H&E, ×200.
Figure 7 Composite ganglioneuroblastoma (nodular stroma-rich ganglioneuroblastoma) of the posterior mediastinum in a 1-year-old female. The circumscribed grayish-tan mass represents the stroma-rich ganglioneuromatous component surrounded by hemorrhagic tissue containing the neuroblasts.
Figure 6 Epithelioid angiosarcoma arising in a malignant mixed germ cell neoplasm of the anterior mediastinum in a 15-year-old male. The signet ringlike cells are the malignant endothelial cells. H&E, ×400.
the incorporation of clinical findings. Those children with advanced disease may experience a variable disease-free interval after platinum-based regimens.62 In those children with initially unresectable tumors, chemotherapy may shrink the tumor so that an attempted complete resection becomes feasible.62 One study has compared the use of combination chemotherapy including cisplatin, bleomycin, and etoposide to the use of carboplatin, bleomycin, and etoposide. It was concluded that carboplatin-based regimens are not only less toxic, but more effective than cisplatin-based regimens.63 The intergroup (Children’s Cancer Group and Pediatric Oncology Group) study 8882 evaluated response rate and survival using etoposide and bleomycin with platinum randomized to high or low dose. Mediastinal cases were analyzed for clinical and operative findings, and the results emphasized the need for aggressive surgical resection either before or after chemotherapy.62 Kesler, et al. looked at postchemotherapy histology and found that postchemotherapy tumor necrosis and teratoma were predictors of survival, and that those with persistent germ cell tumor or sarcomatous degeneration warranted further surgery.64
Neurogenic neoplasms, including neuroblastoma, ganglioneuroblastoma, ganglioneuroma, neurofibroma, and schwannoma account for 25–30% of mediastinal tumors in children; in excess of 90% of these neoplasms present in the posterior mediastinum (see Figures 7 and 8).65,66 Most neuroblastic tumors arising in the posterior mediastinum have favorable pathologic features, often with diffuse ganglioneuroblastomatous or immature ganglioneuromatous features, low mitotic-karyorrhectic index and unamplified Nmyc.67 Other mediastinal neurogenic malignancies in children are rare. Paraganglioma (also termed extra-adrenal pheochromocytoma) can present in the anterior or posterior mediastinum.68 – 70 The patient can experience paroxysmal hypertension as with intra-adrenal pheochromocytoma. This must be anticipated and controlled preoperatively. Surgical excision with follow-up scanning is the treatment of choice.70 Metastases are rare.69 Among the sarcomas of the mediastinum, rhabdomyosarcoma is the most common in children.61 These tumors may arise anteriorly or in a paraspinal location. Most rhabdomyosarcomas have embryonal features. Complete resection of all types is recommended when feasible. In a review of Intergroup Rhabdomyosarcoma Study (IRS)-I and IRSII pilot patients, 10 mediastinal soft tissue sarcomas were described.71 Nine children received radiation and chemotherapy, 1 died shortly after diagnosis, 5 died in the second year after diagnosis, 2 were lost to follow-up, and 2 were long-term survivors.71 The IRS-II study showed that with multiagent chemotherapy and radiation, tumor size greater than 10 cm predicts poor outcome. In an IRS review of soft tissue tumors of the trunk, ten patients had paraspinal soft tissue tumors and received radiation therapy. Seven were
736
PEDIATRIC MALIGNANCIES
(a)
(b)
(c)
Figure 8 Composite ganglioneuroblastoma (nodular stroma-rich ganglioneuroblastoma), showing a sharp plane of fibrous tissue separating the ganglioneuromatous from the neuroblastomatous component (a). Poorly differentiated neuroblastoma (b) and ganglioneuroma (c) are seen at higher magnification. H&E, ×100, ×400, ×400.
without evidence of disease 2.9–6.9 years after diagnosis, and 3 developed recurrences within 2 years of diagnosis and died of disease.72 The malignant peripheral nerve sheath tumor, another mediastinal sarcoma, typically arises in the setting of a child with neurofibromatosis type 1 whose plexiform neurofibroma of the middle or posterior mediastinum has undergone malignant transformation. Ewing’s sarcoma-primitive neuroectodermal tumor (EWS-PNET), melanotic neuroectodermal tumor of infancy (melanotic progonoma), congenitalinfantile fibrosarcoma II, synovial sarcoma, hemangiopericytoma, myxoid liposarcoma, malignant rhabdoid tumor, chondrosarcoma, alveolar soft part sarcoma, epithelioid sarcoma and lipoblastoma are other tumor types that have been reported in the mediastinum of children.73 – 82 Primary neoplasms of the heart, pericardium, and great vessels constitute a rare group of tumors in children and adults. These tumors are almost exclusively mesenchymal in derivation, with a few exceptions such as the pericardial teratoma. An excellent review of fetal and neonatal cardiac tumors was published by Isaacs,83 and a review of anesthesia implications of pediatric cardiac tumors has been published by Kussman.84 Whereas the myxoma is the most common
primary cardiac tumor in adults accounting for 50–60% of all cases, this tumor type is infrequent in children.85 Rhabdomyoma and fibroma together represent 50% or more of primary heart tumors in children 15 years of age or younger at diagnosis. Primary cardiac neoplasms in children are pathologically benign in 85–90% of cases, but that does not diminish the fact that these tumors may cause death and even sudden death secondary to a lethal dysrhythmia.86 Cardiac rhabdomyoma is arguably a hamartoma rather than a true neoplasm. It is estimated that at least 50% or more of infants and young children with rhabdomyoma have tuberous sclerosis.87 The clinical presentation varies from an asymptomatic incidental finding to fetal hydrops and/or tachypnea or a murmur. Multiple lobulated masses arising in the interventricular septum is the typical gross presentation. These masses are composed of large rounded cells with abundant clear cytoplasm.88 Surgical resection may be necessary in some cases when there is obstruction to blood flow or a refractory arrhythmia.89 These tumors are well known for their ability to undergo spontaneous regression.90 Fibroma, like the cardiac rhabdomyoma, often presents in the neonatal period and early infancy, but is also diagnosed into adolescence.91,92 A murmur or congestive heart failure are the two most common modes of clinical presentation. This tumor may be seen in children with the Gorlin-Golz or multiple basal cell carcinoma syndrome. Unlike the rhabdomyoma, the fibroma is usually a solitary mass, which seems to originate from the interventricular septum or as a large intracavitary circumferential mass involving the septum and free wall of the left ventricle (see Figure 9). Multiple fibromas of the heart may be found in neonates with congenital generalized myofibromatosis. A uniform proliferation of spindle cells, arranged in broad fascicles, replace the myocardium. Necrosis and dystrophic calcifications are other microscopic features. These tumors are thought to be equivalent pathologically to infantile myofibromatosis of the soft tissues. If a surgical resection cannot be accomplished, a cardiac transplantation may be required in some cases.93 The other myofibroblastic tumor, which has been reported in the heart in children of all ages, is the IMT.94 This tumor may involve any portion of the heart unlike the more restricted localization of the cardiac fibroma. Direct extension of an IMT from the lung to the heart has been reported.95 Lipoblastoma has been reported in the heart of a child.96 Histiocytoid or oncocytic cardiomyopathy (cardiac hamartoma) is an enigmatic tumefactive lesion(s) presenting with potentially life-threatening arrhythmias or sudden death in infants and young children under 2 years of age.97 Multifocal yellowish nodules are present in the subendocardium, myocardium, and even the heart valves. These lesions are composed of palestaining polygonal cells arranged in cohesive aggregates, often situated beneath the endocardium (see Figure 10). The cytoplasm has an eosinophilic to granular appearance. This nonneoplastic disorder affecting cardiac myocytes is now thought to represent a type of mitochondriopathy with point mutations in the mitochondrial cytochrome B gene.98 Only 40% of all cardiac neoplasms in children are primary sarcomas of various types, including rhabdomyosarcoma,
UNCOMMON PEDIATRIC TUMORS OF THE THORAX
(a)
737
Figure 10 Histiocytoid (oncocytic) cardiomyopathy in an infant who died suddenly without an apparent cause of death. Multiple yellowish nodules were present throughout the heart and were composed of uniform polygonal cells resembling granular cells. H&E, ×200.
Immature somatic elements in these tumors do not alter the excellent prognosis when successful surgical resection has been accomplished. Malignant mesothelioma rarely occurs in the pericardium of children. The single most common category of cardiac neoplasms in the pediatric age-group is represented by those malignancies with secondary or metastatic involvement of the heart. Two groups of investigators, Chan and associates101 and Burke and coworkers,91 have reported that approximately 75% of all cardiac neoplasms in children were leukemic or lymphomatous infiltrates or metastases from various types of bone and soft tissue sarcomas and Wilm’s tumor. Lymphoblastic leukemia-lymphoma seems to have a particular predilection for cardiac involvement. (b) Figure 9 Cardiac fibroma in a neonate who presented with tachypnea and cyanosis while feeding. A chest X ray showed marked cardiomegaly (a). The infant died at 16 days of age. At autopsy, the heart weighed 54 gm (normal = 21 ± 5 gm) and the left ventricle and cavity had been replaced by a grayish-white myocardial mass (b).
EWS-PNET, leiomyosarcoma, and synovial sarcoma.99,100 We have recently encountered a primary alveolar soft part sarcoma of the heart with the t(x;17)(p11;q25) translocation in a child. Sarcomas of the aorta and pulmonary artery in children are even less common than those arising in the heart.101,102 Teratoma of the pericardium is the most common primary neoplasm in this site in childhood.91,101 These tumors usually present in infancy, with cardiorespiratory distress and an enlarged cardiac silhouette or are detected prenatally.103,104 An attachment of the glistening cystic mass to the adventitia of the aorta with a positioning of the tumor on the right side of the heart is the appearance of the tumor at surgery. The predominantly multicystic mass is composed of mature, and occasionally some immature teratomatous elements.
AIRWAYS Tracheal and bronchial tumors presenting as an endophytic mass are restricted principally to a group of epithelial neoplasms arising from the submucosal glands and several types of mesenchymal–stromal tumors.105,106 IMT may present as an obstructing endotracheal or endobronchial mass.107 This tumor, like its pulmonary counterpart, is a spindle cell proliferation with a background of lymphocytes and plasma cells. An IMT may be mistaken pathologically for embryonal rhabdomyosarcoma. Excision is the treatment of choice. Local recurrences are uncommon. Another possibly related lesion is the fibrous histiocytoma, which may have some microscopic features in common with the IMT, but is composed of mononuclear cells with occasional giant cells (see Figure 11). Although the fibrous histiocytoma is regarded in the pathologic sense as a benign tumor, it may recur locally on multiple occasions, just as the IMT does, in a minority of cases.108,109 The differentiation of an endobronchial fibrous histiocytoma from juvenile xanthogranuloma has proven to be problematic in some cases. Mucoepidermoid carcinoma and adenoid cystic carcinoma may present in the trachea, as well as more commonly in the bronchus.110,111 These two tumor types, together with
738
PEDIATRIC MALIGNANCIES
Figure 12 Mucoepidermoid carcinoma of the bronchus in 10-year-old male who presented with wheezing and shortness of breath. Solid nests of low-grade squamous epithelium and glands and cysts lined by well-differentiated mucin epithelium are the features of this low-grade neoplasm. H&E, ×200.
Figure 11 Fibrous histiocytoma of the bronchus in a 16-year-old male who presented with shortness of breath. The tumor is composed predominantly of mononuclear cells with some spindle cells in the background. These tumors may be locally aggressive, as in this case, with multiple local recurrences.
the carcinoid, have been designated collectively in the past as the so-called bronchial adenomas, yet each of these three neoplasms is a malignancy with its own distinctive natural history. Presenting symptoms are those of airway obstruction. The diagnosis can be made by bronchoscopy, but great care must be taken because of the possibility of significant bleeding.112,113 An apparent primary acinic cell carcinoma of the bronchus has been reported in a 4-year-old girl.114 Most mucoepidermoid carcinomas of the bronchus are low-grade neoplasms with an excellent outcome, and even the histologically high-grade tumors do not always behave in an aggressive fashion (see Figure 12).112,113,115,116 Though most low-grade mucoepidermoid carcinomas are readily identified pathologically, the high-grade tumor may be difficult to distinguish from standard squamous cell carcinomas. Treatment consists of complete resection, with preservation of lung tissue whenever possible. Sleeve resection of the affected segment of bronchus has been performed with success in children and adults alike.117 A few cases of adenoid cystic carcinoma (cylindroma) have been reported in children.118 In our experience, this tumor presents as often in the trachea as in the bronchus of children. Cough and postobstructive pneumonia are the most common clinical manifestations. Complete resection is accomplished more often when this tumor presents in the trachea where infiltration is more limited than in the bronchus. Local recurrence and the delayed onset of metastatic disease
Figure 13 Bronchial carcinoid in a 15-year-old female, showing the infiltrating cords and nests of small, uniform tumor cells between the submucous glands of the bronchus. H&E, ×400.
are two consistent aspects of adenoid cystic carcinoma that account for treatment failure. Bronchial carcinoid is the most common primary airway malignancy in childhood.2,10 Approximately 10% of bronchial carcinoid cases occur in the first two decades of life, and the majority of these are diagnosed between 3 and 19 years of age.119,120 The obstructing nature of these tumors is manifested by wheezing, treatment-resistant pneumonitis, and atelectasis.121 Partial or complete occlusion of the bronchus by insular and trabecular profiles of uniform tumor cells is the principal pathologic finding (see Figure 13). Extensive infiltration of the peribronchial tissues by nests of tumor cells is a characteristic feature of these neoplasms, whose gross appearance conveys the impression of a wellcircumscribed and minimally infiltrating tumor. A segmental resection of the bronchus or lobectomy is curative in 90% or more of cases.117,122,123 A local recurrence or metastasis to a regional lymph node(s) (often the immediate peribronchial lymph nodes) is usually the extent of aggressive behavior. Lack, however, reports that two of six children
UNCOMMON PEDIATRIC TUMORS OF THE THORAX
developed metastatic disease.112 Choroidal metastases have been described, which in one case led to blindness.112 Although radiotherapy has not been shown to be effective, it has been administered in conjunction with chemotherapy to adults with symptomatic lesions.124 Combination chemotherapy consisting of platinum and etoposide has been used in adults with some success.124 The carcinoid syndrome is rarely encountered with bronchial carcinoids.124,125 Sarcomas of the respiratory tree include rhabdomyosarcoma of the larynx126,127 or bronchus,128 leiomyosarcoma, and fibrosarcoma.129,130 The latter two neoplasms with a congenital or neonatal presentation have been collectively designated as congenital peribronchial myofibroblastic tumor in the WHO Classification.131 These tumors may be associated with nonimmune fetal hydrops. Rhabdomyosarcomas are treated with multimodal therapy according to IRS staging. Bronchial fibrosarcomas in children are generally low grade with a good outcome after surgery.129,132
739
(a)
LUNG PARENCHYMA Primary malignant neoplasms in the lung parenchyma in children arise from the airway or the nonairway parenchyma. In most cases this anatomic distinction is unnecessary for treatment purposes, and cannot be made with certainty even upon completion of a thorough pathologic examination of the resected specimen. As noted in the previous section, the socalled bronchial adenomas, which are invariably associated with a major airway, account for 35–40% of primary malignant pulmonary neoplasms in children.133 Rhabdomyosarcoma has been reported originating from a cystic adenomatoid malformation (CAM) and the literature reviewed.134 This report reviewed 29 primary pulmonary rhabdomyosarcomas; 15 of these arose in CAM and two in bronchogenic cysts, emphasizing the need for careful histologic analysis of these benign cysts when they are excised. It is thought that most of the rhabdomyosarcomas arising in CAMs are examples of cystic PPBs (see subsequent section on pleuropulmonary blastoma). Bronchogenic carcinomas of the common adult types are extremely rare in children, as noted in the few available epidemiologic studies.135,136 Carcinomas in all anatomic sites comprise 2% or less of all malignant neoplasms in children. Hancock and associates, in their review of primary pediatric malignant neoplasms of the lung, reported that approximately 17% of tumors were bronchogenic carcinomas.1 The three major subtypes of bronchogenic carcinoma (squamous cell carcinoma, adenocarcinoma, and small cell undifferentiated carcinoma) have been described in older children and adolescents.137 – 140 The pathologic features and natural history of these tumor types are seemingly comparable to the same tumors in adults (see Figure 14). Some selected problems that may arise in the pathologic diagnosis of these tumors include the differentiation of mucoepidermoid carcinoma and tracheobronchial papillomatosis from squamous cell carcinoma,141 and avoiding the misinterpretation of a bronchial carcinoid or a lymphoproliferative process as a small cell undifferentiated carcinoma.
(b)
Figure 14 Bronchioloalveolar carcinoma in a 15-year-old male who presented with acute respiratory distress and lung infiltrates, who expired before a diagnosis was established. An adenocarcinoma with the characteristic growth of neoplastic columnar cells along airspaces was seen on microscopic examination (a). The tumor cells were strongly immunopositive for carcinoembryonic antigen (b). H&E, ×200; immunoperoxidase, ×200.
Adenocarcinoma of all subtypes are proportionately more common in adolescents and young adults than the other major subtypes of bronchogenic carcinoma. The acinar and bronchioloalveolar carcinomas are documented in later childhood and adolescence. Several examples of bronchioloalveolar carcinoma or atypical goblet cell hyperplasia arising in type I congenital CAMs would suggest more than a coincidental relationship.142 – 144 A thorough staging with mediastinoscopy should be done in the young patient with a bronchogenic carcinoma, as in an adult with the same tumor. The decision about surgery is predicated on the extent of disease. A consultation with a medical oncologist is appropriate since the therapeutic options are basically the same without regard to age. Fetal pulmonary adenocarcinoma (pulmonary endodermal tumor resembling fetal lung) is a type of well-differentiated adenocarcinoma whose complex glandular pattern resembles the pseudoglandular stage of lung development.145 The majority of these tumors present in adults, but are encountered as well in late childhood and adolescence.146,147 Elongated tubular rounded glands with or without papillary enfoldings with minimal interstitium are principal microscopic features (see Figure 15). Solid formations of tumor cells have a morular appearance. The columnar cells have
740
PEDIATRIC MALIGNANCIES
Figure 15 Fetal pulmonary adenocarcinoma in a 14-year-old female, showing the characteristic tubular glands lined by stratified columnar cells with clear atypical cytoplasm. H&E, ×400.
abundant glycogen-rich clear cytoplasm and nuclei with low to intermediate-grade cytologic abnormalities. Lobectomy is the treatment of choice, and the prognosis is favorable in the absence of high-grade cytologic abnormalities in the epithelial component or a sarcomatous stroma. Koss and associates have proposed a histogenetic relationship among fetal pulmonary adenocarcinoma, PPB, and classic pulmonary blastoma.148 PPB is a malignancy of the lung or less commonly of the parietal pleura with predominantly cystic or cystic-andsolid or exclusively solid features.149 This tumor constituted almost 40% of the pediatric lung neoplasms seen in consultation over a 9-year period (see Table 1). Earlier authors had interpreted this tumor as rhabdomyosarcoma or mesenchymoma arising in a congenital lung cyst or an adenomatoid malformation.150,151 If it is assumed that most or all of the “pulmonary blastomas” in children have been PPB, then approximately 15% of all primary malignant pulmonary tumors in children represent this tumor type. The delineation
of the PPB as a unique pathologic entity of early childhood aligns this neoplasm with other dysembryonic neoplasms of childhood.2 The PPB presents in early childhood, typically in the first 4 years of life,152 although we have seen a case in a 36-year-old male. An interesting aspect of PPB is the observation that 25% of cases have occurred in a constitutional or familial setting, in which the affected children themselves or other young family members have dysembryonic or neoplastic conditions.5 Symptoms of an upper respiratory tract infection, with coughing and some shortness of breath, or chest pain with or without pneumothorax, are the more common presenting manifestations. Imaging studies may reveal the presence of a pulmonary cyst, a cyst with a solid component, or a mass partially filling the hemithorax (see Figure 16). Multifocal cystic PPBs have been reported as individual exceptional cases.153,154 Some PPBs are positioned to suggest a mediastinal tumor rather than one that is based in the lung. Baez-Giangreco and associates reported a PPB in the posterior mediastinum.155 These
Table 1 Tumors of the lung in children. A 9-year review.a
Type
Number
%
Pleuropulmonary blastoma Inflammatory myofibroblastic tumor Other tumor typesb Metastasisc Congenital myofibroblastic tumor Ewing’s sarcoma-primitive neuroectodermal tumor Adenocarcinoma Sarcoma, not otherwise specified Malignant lymphoma Carcinoid Squamous cell carcinoma Neuroendocrine carcinoma
30 9 8 5 4 4 4 4 3 3 2 2
39 11 10 6 5 5 5 5 4 4 2.5 2.5
Total
78
99
a
From the Lauren V. Ackerman Laboratory of Surgical Pathology, Barnes-Jewish and St. Louis Children’s Hospitals, Washington University Medical Center, St. Louis, MO. Fibrous histiocytoma (2), epithelioid hemangioendothelioma (1), hemangiopericytoma (1), malignant peripheral nerve sheath tumor (1), mucoepidermoid carcinoma (1), Langerhans cell histiocytosis (1), and fibrous tumor of the pleura (1). c Metastatic choriocarcinoma, melanoma, epithelioid sarcoma, alveolar soft part sarcoma, giant cell tumor of bone.
(a)
(b)
b
Figure 16 Pleuropulmonary blastoma type I may present as a lung cyst of presumed congenital type (a), or alternatively, as a solid intrathoracic mass representing type III PPB (b).
UNCOMMON PEDIATRIC TUMORS OF THE THORAX
tumors may simulate an empyema or a congenital adenomatoid malformation.152,156,157 There are some potential hazards in the nonoperative management when the assumption of presumed congenital lung cyst in a young child is made.158,159 Because of the possibility of a PPB in these cases, we have recommended surgical resection so that a complete pathologic examination can be performed.160 Another presentation is a soft tissue mass density in the inferior-lateral chest or as a mass in or on the diaphragm in a minority of cases. An extralobar sequestration may be suggested from the imaging. Each of these imaging characteristics reflect the three basic pathologic types of PPB: the cystic type I, the cystic-andsolid type II, and the solid type III.161 Type I PPB is a delicate multicystic lesion typically located at or beneath the visceral pleura. Importantly, a specimen with these gross features must be thoroughly examined microscopically since the delicate septa are either diffusely or focally populated by primitive malignant small cells, with rhabdomyoblastic differentiation in some cases with or without diminutive nodules of cartilage162 (see Figure 17). Thickened areas within the septa or nodules within an otherwise cystic lesion are the macroscopic features of a type II PPB. The delicate septal structures have the microscopic appearance of the type I PPB, whereas the nodules often
741
(a)
(b) Figure 18 Pleuropulmonary blastomas showing the blastomatous (a) and chondrosarcomatous (b) areas of a type II or type III PPB. H&E, ×200, ×200.
(a)
(b) Figure 17 Pleuropulmonary blastoma showing a dense stromal infiltrate of primitive malignant small cells in the wall of a cyst lined by ciliated respiratory epithelium (a). Larger tumor cells with rhabdomyoblastic differentiation may be seen among the smaller, more primitive malignant cells (b). H&E, ×100, ×400.
have the mixed blastomatous and sarcomatous pattern, which are present in the type II and type III PPB (see Figure 18). The type III or solid PPB is a friable, partially necrotic and hemorrhagic mass whose viable areas have a grayish-white to white mucoid appearance. (see Figure 19). These tumors are characterized by their complex multipatterned sarcomatous, blastomatous, and anaplastic features. There is a clear relationship between the age at diagnosis and the pathologic type of PPB with following mean ages at presentation: type I PPB (median age 9 months), type II (median age 31 months) and type III (median age 42 months).163 Approximately 5% of PPBs are type I tumor at diagnosis and it is thought that pathologic progression begins at the stage of a type I PPB. Unfortunately, examples of type I PPB have been regarded clinically and radiographically as a congenital lung cyst on initial detection.164 Surgical resection is the treatment of choice. Resection of type I PPBs appears to be relatively uncomplicated and the survival rate for these children is excellent: six of seven patients have survived 5 months to 17 years after the diagnosis without recurrences. Three of these patients
742
PEDIATRIC MALIGNANCIES
Figure 19 Pleuropulmonary blastoma consisting of several fragments of resected grayish-white tissue with focal hemorrhage. These soft friable tumors are difficult to resect in an en bloc fashion.
underwent surgery and did not receive any further therapy. One child with a type I PPB developed a recurrence whose pathologic features were those of a type II PPB. There was a second recurrence in this same child; the tumor was solid and pathologically resembled a type III PPB. Soon thereafter, the patient died of tumor.152 Type II and III PPBs are clearly aggressive neoplasms, with a 5-year survival rate of less than 50%. If the type II PPB is predominantly cystic, the resection is usually uncomplicated. Tumor growth into the pleura, diaphragm, or mediastinum by the predominantly solid type II PPB or exclusively solid type III PPB serves as a substantial impediment to complete surgical resection. The diffuse involvement of the thorax and the friability of the tumor often lead to a piecemeal excision as a debulking procedure. Despite multimodality therapy, over 50% of those children with type II or type III PPBs developed local recurrences or metastases, often to the brain, and usually within 18 months of diagnosis.152 A recent analysis of those PPBs that had metastasized to the brain showed that the two most important histologic predictors of aggressive behavior in PPB, type III disease and anaplasia, also correlated with the development of central nervous system (CNS) metastasis. Tumor growth pattern, mitotic rate, and amount of necrosis were not useful in predicting metastasis.165 Recommendations for treatment for children with PPB can be summarized as follows: complete surgical resection, when possible, followed by chemotherapy even in cases of type I PPB.152,166 Children with type II and III PPBs may benefit from more aggressive chemotherapy augmented by intracavitary cisplatin or intracavitary 32 P.152 Pulmonary blastoma of the classic adult type with its biphasic histologic pattern is an uncommon neoplasm in general, and even more so in the pediatric age-group. Pulmonary blastomas have a poor prognosis, as do most other high-grade carcinomas of
the lung with regional lymph node and distant metastases. There may be a relationship between the pulmonary blastoma and the well-differentiated fetal type adenocarcinoma, but apparently none with the PPB. Various types of sarcomas, as presumably primary neoplasms in the lung, have been reported in children as single case studies in most instances. Rhabdomyosarcoma of the lung parenchyma without a known primary site elsewhere has been seen rarely in children.71 Some cases are type I PPBs whose sarcomatous component often has the microscopic and immunophenotypic features of embryonal rhabdomyosarcoma. Several cases have been reported as embryonal rhabdomyosarcoma arising in a congenital lung cyst.167 A primary neoplasm of the lung in a young child with rhabdomyosarcomatous features accompanied by other sarcomatous patterns and immature blastomatous foci is very likely a PPB. Primary alveolar rhabdomyosarcoma has been seen in the lung.71 Fibrosarcoma, leiomyosarcoma, malignant fibrous histiocytoma, epithelioid hemangioendothelioma, and synovial sarcoma of the lung are other sarcomas seen in the first two decades of life.132,168 – 172 Many pulmonary fibrosarcomas present in early childhood, and not infrequently in the neonatal period; these tumors together with the so-called congenital leiomyosarcoma are collectively designated as congenital peribronchial myofibroblastic tumors (see Figure 20).173,131 Moran and associates reported their experience with 18 cases of pulmonary leiomyosarcoma; two patients in this series were 5 and 17 years at diagnosis.174,175 Both benign and malignant smooth muscle neoplasms have been documented in several visceral sites, including the lung, in children with human immunodeficiency virus (HIV) acquired immunodeficiency syndrome (AIDS) or who have been organ transplant recipients.175 Epstein-Barr virus has been implicated in the pathogenesis of those smooth muscle neoplasms in the immunosuppressed setting.11,176,177 High pathologic grade is correlated with a poor prognosis. Kaposi sarcoma with its human herpesvirus – 8 associates has been observed
Figure 20 Congenital fibrosarcoma of the lung, showing a cellular neoplasm with focal cleft-like vascular spaces resembling a hemangiopericytoma. Despite the large size, cellularity, and mitotic activity of these neoplasm presenting in the neonatal period, the prognosis is excellent upon successful resection. H&E, ×200.
UNCOMMON PEDIATRIC TUMORS OF THE THORAX
in the lung and other locations also in immunosuppressed children.178,179 Synovial sarcoma with the characteristic translocation, t(X;18) has been identified in the lung as a pulmonary neoplasm in young patients.180 Metastatic pulmonary lesions of synovial sarcoma are seen with greater frequency than the same tumor as a primary sarcoma of the lung. Poorly differentiated synovial sarcoma can resemble a primitive round cell neoplasm of childhood rather than a biphasic or monophasic spindle cell sarcoma. Hemangiopericytoma has been observed rarely; the outcome was fatal in those cases despite surgery and chemotherapy.181,182 Malignant fibrous histiocytoma has been reported in the lung of children.183,184 Epithelioid hemangioendothelioma (intravascular bronchoalveolar tumor) is a low-grade angiosarcoma, which is reported in older children and adolescents as multiple, often bilateral nodules.185,186 The tumor is composed of individual histiocytoid endothelial cells or strands of these cells in a myxohyaline matrix. A similar-appearing neoplasm occurs in the liver, so that the lung involvement is metastatic in some cases.186 The clinical course is a slowly progressive one. Both Ewing’s sarcoma-primitive neuroectodermal tumor and desmoplastic small round cell tumor (DSRCT) have been reported in the lung.187,188 IMT is the most common primary neoplasm of lung in children and adolescents, which typically presents as a solitary mass. A minority of children may have symptoms and signs of an inflammatory process, which is mediated by cytokines produced by the tumor. Multifocal tumors in separate sites are known to occur in less than 5% of cases.189 The neoplastic nature of the IMT has been supported by the observation that a substantial proportion of cases have a translocation in the anaplastic lymphoma kinase (ALK) receptor tyrosine-kinase locus region at 2p23.190 – 192 Dystrophic calcification may or may not be identified in the gross examination of these firm, well-circumscribed tumors with a tan nodular surface (see Figure 21). Spindle cells, lymphocytes, plasma cells, and foci of dense fibrosis are the basic microscopic features (see Figure 22). The spindle cells have the ultrastructure and immunophenotype of myofibroblasts (see Figure 23). A
743
(a)
(b)
(c)
Figure 22 Inflammatory myofibroblastic tumor, showing three characteristic microscopic patterns in a tumor resected from the lung of a 4-year-old female (see Figure 19). An entrapped airspace surrounded by dense fibro-inflammatory tissue (a), spindle cells intermixed with plasma cells and lymphocytes (b), and foci of calcification in a dense hypocellular collagen (c) are the histologic features. H&E, ×200, ×400, ×200.
Figure 21 Inflammatory myofibroblastic tumor resected from the right lower lobe of a 4-year-old female. A well-circumscribed mass measuring 1.5 cm has a faintly nodular, pale tannish-white surface with small yellowish puncta of calcification. (Contributed by Stanley B. Smith, MD, Miami, FL.)
lobectomy is curative in most cases. A minority of IMTs may behave aggressively, with extension into the chest wall that may be accompanied by an alteration in the pathology from a spindle cell to a round cell neoplasm with histiocytelike features (see Figure 24). The presence of necrosis and anaplastic cells should be viewed as evidence of sarcomatous transformation or that the tumor is a sarcoma from its onset, as in the case of a malignant fibrous histiocytoma.183,193
744
PEDIATRIC MALIGNANCIES
Figure 23 Inflammatory myofibroblastic tumor, showing immunoreactivity for smooth muscle actin in the spindle cell component. Immunoperoxidase, ×200.
children with metastatic disease at diagnosis have a poor prognosis.198 In one small study, all children presenting with metastatic disease died within 10 to 17 months.198 Chest wall rhabdomyosarcoma is treated according to disease stage (IRS protocols); complete resection, whenever possible while preserving form and function, is the recommendation.195 Although an early IRS review detailed a rather poor response rate for children with chest wall rhabdomyosarcoma (9/14 died),194 a more recent study details a higher survival rate with surgery, combination chemotherapy (actinomycin-D, doxorubicin, cyclophosphamide, vincristine, with or without bleomycin), and radiotherapy.195 The patients with thoracic sarcomas treated on IRS-II and III were reviewed.201 Of the 84 patients (76 chest wall, 3 parenchymal, 4 pleural, 1 cardiac), 13 presented as group I, 18 as group II, 31 as group III, and 22 as group IV. Thirtyfive were surviving (42%, mean follow-up 1.8 year). Those with complete resections fared best, but a trend toward higher recurrence in group I patients led the authors to caution that care is needed in definition of margins, and to emphasize local control. Other rare malignant chest wall or pleural tumors in children include extraskeletal myxoid chondrosarcoma,202 hemangiopericytoma,203,204 pleuropulmonary blastoma,152,157 and malignant mesothelioma. 205
REFERENCES
Figure 24 Inflammatory myofibroblastic tumor of the lung presenting in a 14-year-old female. In addition to the typical spindle cell pattern in the lung mass, a portion of the tumor had grown into the pleural space and chest wall whose histologic appearance was a mixed cellular infiltrate with atypical round cells with the immunophenotype of histiocytes rather than myofibroblasts (see Figure 21). H&E, ×400.
CHEST WALL, DIAPHRAGM, AND PLEURA Primary malignant lesions of the chest wall, diaphragm, and pleura, although uncommon in children, are well documented.194 – 196 Pinto and associates have reviewed the topic of pleural tumors in childhood10 and Wong et al. have reviewed thoracic wall lesions in children.197 EWS-PNET is the most common primary malignant tumor of the chest wall in a review by Shamberger and Grier.198 Another similar neoplasm in terms of its pathologic features and immunophenotype, the DSRCT, has been reported as a pleural-based tumor. Both the EWS-PNET and DSRCT have translocations involving the EWS gene on 22q12 and FL1 on 11q 24 and WT1 on 11p13, respectively.199 These aggressive neoplasms have been treated with resection, radiotherapy, and adjuvant chemotherapy including vincristine, doxorubicin, cyclophosphamide, and actinomycin-D with mixed results.198,200 Those
1. Hancock BJ, et al. Childhood primary pulmonary neoplasms. J Pediatr Surg 1993; 28: 1133 – 6. 2. Cohen MC, Kaschula RO. Primary pulmonary tumors in childhood: a review of 31 years’ experience and the literature. Pediatr Pulmonol 1992; 14: 222 – 32. 3. Abel RM, et al. Pulmonary metastasectomy for pediatric solid tumors. Pediatr Surg Int 2004; 20: 630 – 2. 4. Balarezo FS, Joshi VV. Proliferative and neoplastic disorders in children with acquired immunodeficiency syndrome. Adv Anat Pathol 2002; 9: 360 – 70. 5. Priest JR, et al. Pleuropulmonary blastoma: a marker for familial disease. J Pediatr 1996; 128: 220 – 4. 6. Staalman CR, Umans U. Hypertrophic osteoarthopathy in childhood malignancy. Med Pediatr Oncol 1993; 21: 676 – 9. 7. Baten M, Vannucci RC. Intraspinal metastatic disease in childhood cancer. J Pediatr 1977; 90: 207 – 12. 8. Carr I. The Ophelia syndrome: Memory loss in Hodgkin’s disease. Lancet 1982; 10: 844 – 5. 9. Cohen MD. Imaging of Children with Cancer Mosby Year Book: St. Louis, 1982: 8 – 19. 10. Pinto A, Machin GA, Trevenen CL: Respiratory tract and serosal tumors. In Parham DR (ed): Pediatric Neoplasia: Morphology and Biology. Philadelphia: Lippincott-Raven, 1996: pp 423 – 447. 11. Jenson HB, et al. Benign and malignant smooth muscle tumors containing Epstein-Barr virus in children with AIDS. Leuk Lymphoma 1997; 27: 303 – 14. 12. Rosenfield NS, et al. CT differentiation of benign and malignant lung nodules in children. J Pediatr Surg 1992; 27: 459 – 61. 13. Massie RJ, Van Asperen PP, Mellis CM. A review of open biopsy for mediastinal masses. J Paediatr Child Health 1997; 33: 230 – 3. 14. Rodgers BM. The role of thoracoscopy in pediatric surgical practice. Semin Pediatr Surg 2003; 12: 62 – 70. 15. Takeda S, et al. Clinical spectrum of primary mediastinal tumors: a comparison of adult and pediatric population at a single Japanese institution. J Surg Oncol 2003; 83: 24 – 30. 16. Grosfeld JL, et al. Mediastinal tumors in children: experience with 196 cases. Ann Surg Oncol 1994; 1: 121 – 7.
UNCOMMON PEDIATRIC TUMORS OF THE THORAX 17. Shimosato Y, Mukai K. Tumors of the Mediastinum. Atlas of Tumor Pathology. Third Series. Fascicle 21 Washington DC: Armed Forces Institute of Pathology, 1997. 18. Simpson I, Campbell PE. Mediastinal masses in childhood: a review from a paediatric pathologist’s point of view. Prog Pediatr Surg 1991; 27: 92 – 126. 19. Spigland N, et al. Malignant thymoma in children: a 20-year review. J Pediatr Surg 1990; 25: 1143 – 6. 20. Hasegawa T, et al. Malignant thymoma in a patient with growth hormone deficiency during growth hormone therapy. Eur J Pediatr 1993; 152: 802 – 4. 21. Lam WW, et al. Paediatric thymoma: unusual occurrence in two siblings. Pediatr Radiol 1993; 23: 124 – 6. 22. Pescarmona E, et al. Thymoma in childhood: a clinicopathological study of five cases. Histopathology 1992; 21: 65 – 8. 23. Dhall G, et al. Thymoma in children: report of two cases and review of literature. J Pediatr Hematol Oncol 2004; 26: 681 – 5. 24. Rice H, et al. Massive thymic hyperplasia: characterization of a rare mediastinal mass. J Pediatr Surg 1994; 29: 1561 – 4. 25. Ocal T, et al. Thymic enlargement in childhood. Turk J Pediatr 2000; 42: 298 – 303. 26. Close PM, et al. Reproducibility of a histogenetic classification of thymic epithelial tumors. Histopathology 1995; 26: 339 – 43. 27. Quintanilla-Martinez L, et al. Histologic subclassification is an independent prognostic factor. Cancer 1994; 74: 606 – 17. 28. Stephan JL, et al. Epstien-Barr virus-positive undifferentiated thymic carcinoma in a 12-year-old white girl. J Pediatr Hematol Oncol 2000; 22: 162 – 6. 29. Okumuru M, et al. The World Health organization histologic classification system reflects the oncologic behavior of thymoma: A clinical study of 273 patients. Cancer 2002; 94: 624 – 32. 30. Rea F, et al. Long-term survival and prognostic factors in thymic epithelial tumors. Eur J Cardiothorac Surg 2004; 26: 412 – 8. 31. Ramon Y, Cajal S, Suster S. Primary thymic epithelial neoplasms in children. Am J Surg Pathol 1991; 15: 466 – 74. 32. Regnard JF, et al. Prognostic factors and long-term results after thymoma resection: a series of 307 patients. J Thorac Cardiovasc Surg 1996; 112: 376 – 84. 33. Kaplinsky C, et al. Childhood malignant thymoma: A phase II study of the European Organization for Research and Treatment of Cancer Lung Cancer Cooperative Group. J Clin Oncol 1992; 14: 814 – 20. 34. Loehrer PJ Sr, et al. Cisplatin, doxorubicin, and cyclophosphamide plus thoracic radiation therapy for limited-stage unresectable thymoma: an intergroup trial. J Clin Oncol 1997; 15: 3093 – 9. 35. Oshita F, et al. Intensive chemotherapy with cisplatin, doxorubicin, cyclophosphamide, etoposide and granulocyte colony-stimulating factor for advanced thymoma or thymic cancer: preliminary results. Jpn J Clin Oncol 1995; 25: 208 – 12. 36. Giaccone G, et al. Cisplatin and etoposide combination chemotherapy for locally advanced or metastatic thymoma: A phase II study of the European Organization for Research and Treatment of Cancer Lung Cancer Cooperative Group. J Clin Oncol 1996; 14: 814 – 20. 37. Furman WL, et al. Thymoma and myasthenia gravis in a 4-year-old child. Case report and review of the literature. Cancer 1985; 56: 2703 – 6. 38. Masaoka A, et al. Thymomas associated with pure red cell aplasia. Histologic and follow-up studies. Cancer 1989; 26: 419 – 24. 39. Niehuis T, et al. Treatment of pediatric malignant thymoma: long-term remission in a 14-year-old boy with EBV-associated thymic carcinoma by aggressive, combined modality treatment. Med Pediatr Oncol 1996; 26: 419 – 24. 40. Matsuno Y, et al. Detection of Epstein-Barr virus DNA in a Japanese case of lymphoepithelioma-like thymic carcinoma. Jpn J Clin Oncol 1992; 83: 127 – 30. 41. Leyvraz S, et al. Association of Epstein-Barr virus with thymic carcinoma. N Engl J Med 1985; 312: 1296 – 9. 42. McCaughey ES, et al. Ectopic ACTH production by a thymic carcinoid tumor. Eur J Pediatr 1987; 146: 590 – 1. 43. Gartner LA, Voorhess ML. Adrenocorticotropic hormone-producing thymic carcinoid in a teenager. Cancer 1993; 71: 106 – 11.
745
44. Brown LR, et al. Roentgenologic diagnosis of primary corticotropinproducing carcinoid tumors of the mediastinum. Radiology 1982; 142: 143 – 8. 45. Salyer WR, Dalyer DC, Eggleston JC. Carcinoid tumors of the thymus. Cancer 1976; 37: 958 – 73. 46. Wick MR, et al. Primary mediastinal carcinoid tumors. Am J Surg Pathol 1982; 6: 195 – 205. 47. Rosai J, Levine GD: Tumors of the thymus. In: Atlas of tumor pathology. Second series, fascicle 13. Armed Forces Institute of Pathology. 1978: 170. 48. Dusmet ME, McKneally MF. Pulmonary and thymic carcinoid tumors. World J Surg 1996; 20: 189 – 95. 49. Moran CA, Suster S. Primary germ cell tumors of the mediastinum: I. Analysis of 322 cases with special emphasis on teratomatous lesions and a proposal for histopathologic classification and clinical staging. Cancer 1997; 80: 681 – 90. 50. Mullen B, Richardson JD. Primary anterior mediastinal tumors in children and adults. Ann Thorac Surg 1986; 42: 338 – 45. 51. Dulmet EM, et al. Germ cell tumors of the mediastinum. A 30-year experience. Cancer 1993; 72: 1894 – 900. 52. Rosado-de-Christenson ML, Templeton PA, Morn CA. From the archives of the AFIP. Mediastinal germ cell tumors: radiologic and pathologic correlation. Radiographics 1992; 12: 1013 – 30. 53. Schneider DT, et al. Genetic analysis of childhood germ cell tumors with comparative genomic hybridization. Klin Padiatr 2001; 213: 204 – 11. 54. Billmire D, et al. Malignant mediastinal germ cell tumors: an intergroup study. J Pediatr Surg 2001; 36: 18 – 24. 55. Gooneratne S, et al. Anterior mediastinal endodermal sinus (yolk sac) tumor in a female infant. Cancer 1985; 56: 1430 – 3. 56. Zon R, et al. Benign hematologic neoplasm associated with mediastinal mature teratoma in a patient with Klinefelter’s syndrome: a case report. Med Pediatr Oncol 1994; 23: 376 – 9. 57. Dehner LP. Germ cell tumors of the mediastinum. Semin Diagn Pathol 1990; 7: 266 – 84. 58. Moran CA, et al. Primary germ cell tumors of the mediastinum. II. Mediastinal seminomas-a clinicopathologic and immunohistochemical study of 120 cases. Cancer 1997; 80: 691 – 8. 59. Moran CA, Suster S, Koss MN. Primary germ cell tumors of the mediastinum. III. Yolk sac tumor, embryonal carcinoma, choriocarcinoma, and combined nonteratomatous germ cell tumors of the mediastinum – A clinicopathologic and immunohistochemical study of 64 cases. Cancer 1997; 80: 699 – 707. 60. Moran CA, Suster S. Primary mediastinal choriocarcinomas: a clinicopathologic and immunohistochemical study of eight cases. Am J Surg Pathol 1997; 21: 1007 – 12. 61. Suster S, Moran CA, Koss MN. Rhabdomyosarcomas of the anterior mediastinum: report of four cases unassociated with germ cell, teratomatous, or thymic carcinomatous components. Hum Pathol 1994; 25: 349 – 56. 62. Billmire D, et al. Malignant mediastinal germ cell tumors: An intergroup study. J Pediatr Surg 2001; 36: 18 – 24. 63. Mann JR, et al. UKCCG’s germ cell tumor (GCT) studies: Improving outcome for children with malignant extracranial non-gonadal tumours-carboplatin, etoposide, and bleomycin are effective and less toxic than previous regimens. Med Pediatr Oncol 1998; 30: 217 – 27. 64. Kesler K, et al. Primary mediastinal nonseminomatous germ cell tumors: the influence of postchemotherapy pathology on long-term survival after surgery. J Thorac Cardiovasc Surg 1999; 118: 692 – 700. 65. Saenz NC, et al. Posterior mediastinal masses. J Pediatr Surg 1993; 28: 172 – 6. 66. Strollo DC, Rosado-de-Christenson ML, Jett JR. Primary mediastinal tumors: part II. Tumors of the middle and posterior mediastinum. Chest 1997; 112: 1344 – 57. 67. Lonergan GJ, et al. Neuroblastoma, ganglioneuroblastoma and ganglioneuroma: Radiologic-pathologic correlation. Radiographics 2002; 22: 911 – 34. 68. Moran CA, et al. Mediastinal paragangliomas. A clinicopathologic and immunohistochemical study of 16 cases. Cancer 1993; 72: 2358 – 64.
746
PEDIATRIC MALIGNANCIES
69. Hodgkinson DJ, et al. Extra-adrenal intrathoracic functioning paraganglioma (pheochromocytoma) in childhood. Mayo Clin Proc 1980; 55: 271 – 6. 70. Spector J, Willis D, Ginsburg H. Paraganglioma (pheochromocytoma) of the posterior mediastinum: a case report and review of the literature. J Pediatr Surg 2003; 38: 1114 – 6. 71. Crist WM, et al. Intrathoracic soft tissue sarcomas in children. Cancer 1982; 50: 598 – 604. 72. Wharam MD, et al. Radiation Therapy for Rhabdomyosarcoma: Local failure risk for Clinical Group III patients on intergroup Rhabdomyosarcoma study II. Int J Radiat Oncol Biol Phys 1997; 38: 797 – 804. 73. Schmidt D, Harms D, Burdach S. Malignant peripheral neuroectodermal tumours of childhood and adolescence. Virchows Arch [Pathol Anat] 1985; 406: 351 – 65. 74. D’Abrera VS, Burfitt-Williams W. A melanotic neuroectodermal neoplasm of the posterior mediastinum. J Pathol 1973; 111: 165 – 72. 75. Khoddami M, et al. Melanotic neuroectodermal tumor of infancy: a molecular genetic study. Pediatr Dev Pathol 1998; 1: 295 – 9. 76. Flieder DB, Moran CA, Suster S. Primary alveolar soft -part sarcoma of the mediastinum: a clinicopathological and immunohistochemical study of two cases. Histopathology 1997; 31: 469 – 73. 77. Gross E, et al. Epithelioid sarcoma in children. J Pediatr Surg 1996; 31: 1663 – 5. 78. Klimstra DS, et al. Liposarcoma of the anterior mediastinum and thymus. A clinicopathologic study of 28 cases. Am J Surg Pathol 1995; 19: 782 – 91. 79. Mikkilineni RS, et al. Liposarcoma of the posterior mediastinum in a child. Chest 1994; 106: 1288 – 9. 80. Suster S, Moran CA. Malignant cartilaginous tumors of the mediastinum: clinicopathological study of six cases presenting as extraskeletal soft tissue masses. Hum Pathol 1997; 28: 588 – 94. 81. Suster S, Moran CA, Koss MN. Epithelioid hemangioendothelioma of the anterior mediastinum. Clinicopathologic, immunohistochemical, and ultrastructural analysis of 12 cases. Am J Surg Pathol 1994; 18: 871 – 81. 82. Mack TM. Sarcomas and other malignancies of soft tissue, retroperitoneum, peritoneum, pleura, heart, mediastinum, and spleen. Cancer 1995; 75: 211 – 44. 83. Isaacs H. Fetal and neonatal cardiac tumors. Pediatr Cardiol 2004; 25: 252 – 73. 84. Kussman B, et al. Anesthetic implications of primary cardiac tumors in infants and children. J Cardiothorac Vasc Anesth 2002; 16: 582 – 6. 85. Burke AP, Virmani R. Tumors of the Heart and Great Vessels. Third Series, Fascicle 16 Washington, D.C.: Armed Forces Institute of Pathology, 1996: 1 – 11. 86. Beghetti M, et al. Pediatric primary benign cardiac tumors: A 15-year review. Am Heart J 1997; 134: 1107 – 14. 87. Smith WH, et al. Cardiac rhabdomyomata in tuberous sclerosis: their course and diagnostic value. Arch Dis Child 1989; 64: 196 – 200. 88. Burke AP, Virmani R. Cardiac rhabdomyoma: a clinicopathologic study. Mod Pathol 1991; 4: 70 – 4. 89. Takach TJ, et al. Primary cardiac tumors in infants and children: immediate and long-term operative results. Ann Thorac Surg 1996; 62: 559 – 64. 90. Smyth JF, et al. Natural history of cardiac rhabdomyoma in infancy and childhood. Am J Cardiol 1990; 66: 1247 – 9. 91. Burke BA, Edwareds JE, Titus JL. Tumor and tumor-like lesions of the heart and great vessels in the young. Arch Pathol Lab Med 1992; 5: 357 – 402. 92. de Montpreville VT, et al. Fibroma and inflammatory myofibroblastic tumor of the heart. Ann Diagn Pathol 2001; 5: 335 – 42. 93. Michler RE, Goldstein DJ. Treatment of cardiac tumors by orthotopic cardiac transplantation. Semin Oncol 1997; 24: 534 – 9. 94. LI L, Cerilli LA, Wick MR. Inflammatory pseudotumor (myofibroblastic tumor) of the heart. Ann Diagn Pathol 2002; 6: 116 – 21. 95. Corneli G, et al. Invasive inflammatory pseudo-tumor involving the lung and the mediastinum. Thorac Cardiovasc Surg 2001; 49: 124 – 6. 96. Dishop MK, et al. Primary cardiac lipoblastoma. Pediatr Dev Pathol 2001; 4: 276 – 80. 97. Shehata BM, et al. Histiocytoid cardiomyopathy: three new cases and review of the literature. Pediatr Dev Pathol 1998; 1: 56 – 89.
98. Vallance HD, et al. A case of sporadic infantile histiocytoid cardiomyopathy caused by the A8344G (MERRF) mitochondrial DNA mutation. Pediatr Cardiol 2004; 25: 538 – 40. 99. Ai SZ, et al. Pleomorphic rhabdomyosarcoma of the heart metastatic to bone. Report of a case with fine needle aspiration biopsy findings. Acta Cytol 1995; 39: 555 – 8. 100. Burke AP, Cowan D, Virmani R. Primary sarcomas of the heart. Cancer 1992; 69: 387 – 95. 101. Chan HSL, et al. Primary and secondary tumors of childhood involving the heart, pericardium, and great vessels. A report of 75 cases and review of the literature. Cancer 1985; 56: 825 – 36. 102. Walles T, Macchiarini P. Clinical-pathologic conference in general thoracic surgery: pulmonary artery fibrohistiocytic tumor in a child. J Thorac Cardiovasc Surg 2004; 128: 319 – 22. 103. Perez-Aytes A, et al. Non-immunological hydrops fetalis and intrapericardia teratoma: case report and review. Chest Surg Clin N Am 1995; 6: 875 – 98. 104. Tollens M, et al. Pericardial teratoma: prenatal diagnosis and course. Fetal Diagn Ther 2003; 18: 432 – 6. 105. Mathisen DJ. Tracheal tumors. Chest Surg Clin N Am 1996; 6: 875 – 98. 106. Al-Qahtani AR, Di Lorenzo M, Yazbeck S. Endobronchial tumors in children: Institutional experience and literature review. J Pediatr Surg 2003; 38: 733 – 6. 107. Dewar AL, Connett GJ. Inflammatory pseudotumor of the trachea in a ten-month-old infant. Pediatr Pulmonol 1997; 23: 307 – 9. 108. Sculerati N, et al. Fibrous histiocytoma of the trachea: management of a rare cause of upper airway obstruction. Int J Pediatr Otorhinolaryngol 1990; 19: 295 – 301. 109. Tagge E, et al. Obstructing endobronchial fibrous histiocytoma: potential for lung salvage. J Pediatr Surg 1991; 26: 1067 – 9. 110. Bellah RD, Mahboubi S, Berdon WE. Malignant endobronchial lesions of adolescence. Pediatr Radiol 1992; 22: 563 – 7. 111. Hause DW, Harvey JC. Endobronchial carcinoid and mucoepidermoid carcinoma in children. J Surg Oncol 1991; 46: 270 – 2. 112. Lack EE, et al. Primary bronchial tumors in childhood. A clinicopathologic study of six cases. Cancer 1983; 51: 492 – 7. 113. Giusti RJ, Flores RM. Mucoepidermoid carcinoma of the bronchus presenting with a negative chest X-ray and normal pulmonary function in two teenagers: two case reports and review of the literature. Pediatr Pulmonol 2004; 37: 81 – 4. 114. Sabaratnam RM, Anunathan R, Govender D. Acinic cell carcinoma: an unusual case of bronchial obstruction in an child. Pediatr Dev Pathol 2004; 7: 521 – 6. 115. Granata C, et al. Mucoepidermoid carcinoma of the bronchus: a case report and review of the literature. Pediatr Pulmonol 1997; 22: 226 – 32. 116. Seo IS, et al. Mucoepidermoid carcinoma of the bronchus in a 4-yearold child. A high-grade variant with lymph node metastasis. Cancer 1984; 53: 1600 – 4. 117. Gaissert HA, et al. Tracheobronchial sleeve resection in children and adolescents. J Pediatr Surg 1994; 29: 192 – 7. 118. Ahel V, Zubovic I, Rozmanic V. Bronchial adenoid cystic carcinoma with saccular bronchiectasis as a cause of recurrent pneumonia in children. Pediatr Pulmonol 1992; 12: 260 – 2. 119. Wang LT, Wilkins EW Jr, Bode HH. Bronchial carcinoid tumors in pediatric patients. Chest 1993; 103: 1426 – 8. 120. Moraes T, et al. Pediatric pulmonary carcinoid: a case report and review of the literature. Pediatr Pulmonol 2003; 35: 318 – 22. 121. Hulka GF, et al. Carcinoid tumor of the trachea in a pediatric patient. Otolaryngol Head Neck Surg 1996; 114: 822 – 5. 122. Schreurs AJ, et al. A twenty-five-year follow-up of ninety-three resected typical carcinoid tumors of the lung. J Thorac Cardiovasc Surg 1992; 104: 1470 – 5. 123. Vadasz P, et al. Diagnosis and treatment of bronchial carcinoid tumors: clinical and pathological review of 120 operated patients. Eur J Cardiothorac Surg 1993; 7: 8 – 11. 124. Davila DG, et al. Bronchial carcinoid tumors. Mayo Clin Proc 1993; 68: 795 – 803. 125. Doppman JL, et al. Ectopic adrenocorticotropic hormone syndrome: Localization studies in 28 patients. Radiology 1989; 172: 115 – 24.
UNCOMMON PEDIATRIC TUMORS OF THE THORAX 126. Diehn KW, Hyams VJ, Harris AE. Rhabdomyosarcoma of the larynx: A case report and review of the literature. Laryngoscope 1984; 94: 201 – 5. 127. Dodd-O JM, Wieneke KF, Rosman PM. Laryngeal rhabdomyosarcoma. Case report and literature review. Cancer 1987; 59: 1012 – 8. 128. Fallon G, Schiller M, Kilman JW. Primary rhabdomyosarcoma of the bronchus. Ann Thorac Surg 1971; 12: 650 – 4. 129. Garnett JD, Cook CB. Primary bronchopulmonary fibrosarcoma of the trachea in a child. South Med J 1993; 86: 1283 – 5. 130. Skarin A. Unusual pulmonary lesions. J Clin Oncol 2002; 20: 2745 – 51. 131. Travis WD, et al. Congenital peribronchial myofibroblastic tumour. In Travis WD, et al. (eds): Pathology and Genetics of the Lung, Pleura, Thymus and Heart World Health Classification of Tumors. IARD Press, 2004, pp 102 – 103. 132. Pettinato G, et al. Primary bronchopulmonary fibrosarcoma of childhood and adolescence: Reassessment of a low-grade malignancy. Clinicopathologic study of five cases and review of the literature. Hum Pathol 1989; 20: 463 – 71. 133. Lal DR, et al. Primary epithelial lung malignancies in the pediatric population. Pediatr Blood Cancer 2005; 44: 1 – 4. 134. Ozcan C, et al. Primary pulmonary rhabdomyosarcoma arising within cystic adenomatoid malformation: a case report and review of the literature. J Pediatr Surg 2001; 36: 1062 – 5. 135. Al-Sheyyab M, et al. Malignant epithelial tumors in children: incidence and aetiology. Med Pediatr Oncol 1993; 21: 421 – 8. 136. Stiller CA. International variations in the incidence of childhood carcinomas. Cancer Epidemiol Biomarkers Prev 1994; 3: 305 – 10. 137. Asamura H, et al. AFP-producing squamous cell carcinoma of the lung in an adolescent. Jpn J Clin Oncol 1996; 26: 103 – 6. 138. Curcio LD, et al. Primary lymphoepithelioma-like carcinoma of the lung in a child. Report of an Epstein-Barr virus-related neoplasm. Chest 1997; 111: 250 – 1. 139. Icard P, et al. Primary lung cancer in young patients: a study of 82 surgically treated patients. Ann Thorac Surg 1992; 54: 99 – 103. 140. Kojima R, et al. Pulmonary carcinoma associated with hamartoma in an 11-year-old boy. Am J Pediatr Hematol Oncol 1993; 15: 439 – 42. 141. Zawadzka-Glos L, et al. Lower airway papillomatosis in children. Int J Pediatr Otorhinolaryngol 2003; 67: 1117 – 21. 142. Granata C, et al. Bronchioloalveolar carcinoma arising in congenital cystic adenomatoid malformation in a child: a case report and review of malignancies originating in congenital cystic adenomatoid malformation. Pediatr Pulmonol 1998; 25: 62 – 6. 143. MacSweeney F, et al. An assessment of the expanded classification of congenital cystic adenomatoid malformations and their relationship to malignant transformation. Am J Surg Pathol 2003; 27: 1139 – 46. 144. Stacher E, et al. Atypical goblet cell hyperplasia in congenital cystic adenomatoid malformation as a possible preneoplasia for pulmonary adenocarcinoma in childhood: A genetic analysis. Hum Pathol 2004; 35: 565 – 70. 145. Nakatani Y, et al. Pulmonary adenocarcinomas of the fetal lung type. A clinicopathologic study indicating differences in histology, epidemiology, and natural history of low-grade and high-grade forms. Am J Surg Pathol 1998; 22: 399 – 411. 146. Singh SP, Besner GE, Schauer GM. Pulmonary endodermal tumor resembling fetal lung: report of a case in a 14 year-old girl. Pediatr Pathol Lab Med 1997; 17: 951 – 8. 147. DiFurio MJ, Auerbach A, Kaplan KJ. Well-differentiated fetal adenocarcinoma: rare tumor in the pediatric population. Pediatr Dev Pathol 2003; 6: 564 – 7. 148. Koss MN, Hochholzer L, O’Leary T. Pulmonary blastoma. Cancer 1991; 67: 2368 – 81. 149. Dehner LP, et al. Pleuropulmonary blastoma. In Travis WD, et al. (eds): Pathology and Genetics of the Lung, Pleura, Thymus and Heart. World Health Classification of Tumors. IARD Press, 2004, pp 99 – 100. 150. D’Agostino S, et al. Embryonal rhabdomyosarcoma of the lung arising in cystic adenomatoid malformation: case report and review of the literature. J Pediatr Surg 1997; 32: 1381 – 3. 151. Domizio P, et al. Malignant mesenchymoma associated with a congenital lung cyst in a child. Case report and review of the literature. Pediatr Pathol 1990; 10: 785 – 97.
747
152. Priest JR, et al. Pleuropulmonary blastoma. A clinicopathologic study of 50 cases. Cancer 1997; 80: 147 – 61. 153. Lallier M, et al. Pleuropulmonary blastoma: a rare pathology with an even rarer presentation. J Pediatr Surg 1999; 34: 1057 – 9. 154. Picaud JC, et al. Bilateral cystic pleuropulmonary blastoma in early infancy. J Pediatr 2000; 136: 834 – 6. 155. Baez-Giangreco A, et al. Pleuropulmonary blastoma of the lung presenting as posterior mediastinal mass: a case report. Pediatr Hematol Oncol 1997; 14: 475 – 81. 156. Merriman TE, et al. A rare tumor masquerading as an empyema: pleuropulmonary blastoma. Pediatr Pulmonol 1996; 22: 408 – 11. 157. Katz DS, et al. Pleuropulmonary blastoma simulating an empyema in a young child. J Thorac Imaging 1995; 10: 112 – 6. 158. Tagge EP, et al. Childhood pleuropulmonary blastoma: caution against nonoperative management of congenital lung cysts. J Pediatr Surg 1996; 31: 187 – 9. 159. Papagiannopoulos K, et al. Cystic lung lesions in the pediatric and adult populations: surgical experience at the Brompton hospital. Ann Thorac Surg 2002; 73: 1594 – 8. 160. Dehner LP. Beware of “degenerating” congenital pulmonary cysts. Pediatr Surg Int 2005; 21: 123 – 4. 161. Dehner LP, Watterson J, Priest J. Pleuropulmonary blastoma. A unique intrathoracic pulmonary neoplasm of childhood. Perspect Pediatr Pathol 1995; 18: 214 – 26. 162. Hill DA, Dehner LP. A cautionary note about congenital cystic adenomatoid malformation (CCAM) type 4. Am J Surg Pathol 2004; 28: 554 – 5. 163. Hill DA. USCAP Specialty Conference: case 1-type I pleuropulmonary blastoma. Pediatr Dev Pathol 2005; 8: 77 – 84. 164. Hasiotou M, et al. Pleuropulmonary blastoma in the area of a previously diagnosed congenital lung cyst: report of two cases. Acta Radiol 2004; 45: 289 – 92. 165. Serrano MF, et al. Histologic predictors of central nervous system metastasis in pleuropulmonary blastoma: a pilot study. (abstract 10). Mod Pathol 2005; 18: 303. 166. Calabria R, et al. Management of pulmonary blastoma in children. Am Surg 1993; 59: 192 – 6. 167. Allen BT, Day DL, Dehner LP. Primary pulmonary rhabdomyosarcoma of the lung in children. Report of two cases presenting with spontaneous pneumothorax. Cancer 1987; 59: 1005 – 11. 168. Guccion JG, Rosen SH. Bronchopulmonary leiomyosarcoma and fibrosarcoma. A study of 32 cases and review of the literature. Cancer 1972; 30: 836 – 47. 169. Beluffi G, et al. Primary leiomyosarcoma of the lung in a girl. Pediatr Radiol 1986; 16: 240 – 4. 170. Spillane A, et al. Synovial Sarcoma: a clinicopathologic, staging, and prognostic assessment. J Clin Oncol 2000; 18: 3794 – 803. 171. Etienne-Mastroianni B, et al. Primary sarcomas of the lung: a clinicopathologic study of 12 cases. Lung Cancer 2002; 38: 283 – 9. 172. Kunst PW, et al. Unusual pulmonary lesions: case I. A juvenile bronchopulmonary fibrosarcoma. J Clin Oncol 2002; 20: 2745 – 51. 173. McGinnis M, et al. Congenital peribronchial myofibroblastic tumor (so-called “congenital leiomyosarcoma”). A distinct neonatal lung lesion associated with nonimmune hydrops fetalis. Mod Pathol 1993; 6: 487 – 92. 174. Moran CA, et al. Primary leiomyosarcomas of the lungs: A clinicopathologic and immunohistochemical study of 18 cases. Mod Pathol 1997; 10: 121 – 8. 175. Van Hoerven KH, et al. Visceral myogenic tumours. A manifestation of HIV infection in children. Am J Surg Pathol 1993; 17: 1176 – 81. 176. Cheuk W, Li PC, Chan JK. Epstein-Barr virus-associated smooth muscle tumour: a distinctive mesenchymal tumour of immunocompromised individuals. Pathol 2002; 34: 245 – 9. 177. Monforte-Munoz H, Kapoor N, Saavedra JA. Epstein-Barr virusassociated leiomyomatosis and post-transplant lymphoproliferative disorder in a child with severe combined immunodeficiency: case report and review of the literature. Pediatr Dev Pathol 2003; 6: 449 – 57. 178. Marais BJ, Pienaar J, Gie RP. Kaposi sarcoma with upper airway obstruction and bilateral chylothoraces. Pediatr Infect Dis J 2003; 22: 926 – 8.
748
PEDIATRIC MALIGNANCIES
179. Stebbing J, Portsmouth S, Bower M. Insights into the molecular biology and sero-epidemiology of Kaposi’s sarcoma. Curr Opin Infect Dis 2003; 16: 25 – 31. 180. Begueret H, et al. Primary intrathoracic synovial sarcoma: a clinicopathologic study of 40 t(x;18)-positive cases from the French Sarcoma Group and Mesopath Group. Am J Surg Pathol 2005; 29: 339 – 46. 181. Van Damme H, et al. Primary pulmonary hemangiopericytoma: Early local recurrence after perioperative rupture of the giant tumor mass (two cases). Surgery 1990; 108: 105 – 9. 182. Dagfous J, et al. Hemangiopericytome pulmonaire primitif de l’enfant. Rev Mal Respir 1992; 9: 46 – 8. 183. Nistal M, et al. Malignant fibrous histiocytoma of the lung in a child. An unusual neoplasm that can mimic inflammatory pseudotumor. Eur J Pediatr 1997; 156: 107 – 9. 184. Shah SJ, Craver RD, Yu LC. Primary malignant fibrous histiocytoma of the lung in a child: a case report and review of literature. Pediatr Hematol Oncol 1996; 13: 531 – 8. 185. Buggage RR, et al. Epithelioid hemangioendothelioma of the lung: pleural effusion cytology, ultrastructure, and brief literature review. Diagn Cytopathol 1995; 13: 54 – 60. 186. Roepke JE, Heifetz SA. Pathological case of the month. Epithelioid hemangioendothelioma (intravascular bronchioloalveolar tumor) of the lung. Arch Pediatr Adolesc Med 1997; 151: 317 – 9. 187. Morton RL, et al. Clinicopathologic conference: a large pulmonary cavitary lesion in a 2-year-old boy. J Pediatr 2004; 144: 107 – 11. 188. Syed S, et al. Desmoplastic small round cell tumor of the lung. Arch Pathol Lab Med 2002; 126: 1226 – 8. 189. Crespo C, et al. Intracranial and mediastinal inflammatory myofibroblastic tumour. Pediatr Radiol 2001; 31: 600 – 2. 190. Chun YS, et al. Pediatric inflammatory myofibroblastic tumor: Anaplastic lymphoma kinase (ALK) expression and prognosis. Pediatr Blood Cancer 2004; 15: 796 – 801, [Epub ahead of print]. 191. Debelenko LV, et al. Identification of CARS-ALK fusion in primary and metastatic lesions of an inflammatory myofibroblastic tumor. Lab Invest 2003; 83: 1255 – 65.
192. Dehner LP. Inflammatory myofibroblastic tumor. The continued definition of one type of so-called inflammatory pseudotumor. Am J Surg Pathol 2004; 28: 1652 – 4. 193. Morotti RA, et al. Pediatric inflammatory myofibroblastic tumor with late metastasis to the lung: case report and review of the literature. Pediatr Dev Pathol 2005; 8: 224 – 9, [Epub ahead of print]. 194. Raney RB, et al. Soft-tissue sarcoma of the trunk in childhood. Results of the Intergroup Rhabdomyosarcoma Study. Cancer 1982; 49: 2612 – 6. 195. Saenz NC, et al. Chest wall rhabdomyosarcoma. Cancer 1997; 80: 1513 – 7. 196. Gonzalez-Crussi F, et al. Peripheral neuroectodermal tumors of the chest wall in childhood. Cancer 1984; 54: 2519 – 27. 197. Wong K, et al. Thoracic wall lesions in children. Pediatr Pulmonol 2004; 37: 257 – 63. 198. Shamberger RC, et al. Malignant small round cell tumor (Ewing’s – PNET) of the chest wall in children. J Pediatr Surg 1994; 29: 179 – 85. 199. Qualman SJ, Morotti RA. Risk assignment in pediatric soft-tissue sarcomas: An evolving molecular classification. Curr Oncol Rep 2002; 4: 123 – 30. 200. Gururangan S, et al. Treatment of children with peripheral primitive neuroectodermal tumor or extraosseous Ewing’s tumor with Ewing’sdirected therapy. J Pediatr Hematol Oncol 1998; 20: 55 – 61. 201. Andrassy R, et al. Thoracic sarcomas in children. Ann Surg 1998; 227: 170 – 3. 202. Hachitanda Y, et al. Extraskeletal Myxoid chondrosarcoma in young children. Cancer 1988; 61: 2521 – 6. 203. Raafat F, et al. Recurrent hemangiopericytoma of the chest wall: Report of a case in a 5-year-old boy. Pediatr Pathol 1994; 14: 19 – 25. 204. Kuaffman SL, Stout AP. Hemangiopericytoma in children. Cancer 1960; 13: 695 – 710. 205. Vanneuville G, et al. Malignant pleural tumor in child mimicking a mesothelioma. Eur J Pediatr Surg 1992; 3: 362 – 5.
Section 11 : Pediatric Malignancies
68
Uncommon Tumors of the Gastrointestinal Tract in Children Christopher L. Moertel, Jan Watterson and Louis P. Dehner
INTRODUCTION Approximately 230 000 cases of malignant neoplasms of the digestive system are newly diagnosed each year in the United States.1 The primary sites are the colorectum and anus in almost 60% of cases, followed by the pancreas (12% of cases), stomach (8%), liver and biliary tract (7%), and esophagus (5%). Adenocarcinoma is the most common tumor type, accounting for 90% of all digestive tract neoplasms. The remaining tumor types include carcinoid–neuroendocrine carcinoma, hepatocellular carcinoma, squamous cell carcinoma of the esophagus, malignant lymphoma, and leiomyosarcoma/malignant gastrointestinal stromal tumors (GISTs). With the exception of malignant lymphoma, the overwhelming majority of these neoplasms are diagnosed in individuals over the age of 40 years. Al-Sheyyab and associates estimated the incidence of malignant epithelial tumors from all sites in children 15 years of age or younger at 3.3 cases per million per year (in the United Kingdom).2 Only 1–2% of all malignancies occurring in the first 15 years of life are adult-type epithelial neoplasms. The majority of children (60–70% of cases) with adulttype malignant epithelial neoplasms, inclusive of all primary sites, are between 10 and 15 years of age at diagnosis. The two most common malignant epithelial tumors of the gastrointestinal tract in children are the appendiceal carcinoid and colorectal carcinoma.3 Another perspective on malignant epithelial neoplasms of the digestive system is provided by the experience of the Kiel Pediatric Tumor Registry.4 Among almost 12 000 cases during a 15-year period, 220 (2%) tumors were carcinomas, including 24 hepatocellular carcinomas and carcinomas of the gastrointestinal tract (colon, 4; stomach, 3). McWhirter and associates identified 211 carcinomas in British children and 6% of their cases arose in the gastrointestinal tract, with 8 of 12 cases presenting in the large intestine.5
Recent studies from the United States and United Kingdom report gastrointestinal malignancies in 139 children.6 – 8 Two studies described malignant neoplasms only;6,8 and the other included both benign and malignant tumors.7 The age range of the 111 children with malignant tumors was from 1–17 years, with more boys affected than girls. Non-Hodgkin’s lymphoma was the pathologic diagnosis in approximately 80% of the malignant neoplasms. Carcinoid tumor was the second most common malignancy in one series,6 and accounted for approximately 15% of cases; all these tumors presented in the appendix. There were 14 cases of adenocarcinoma of the colon, one of the stomach and five cases of gastric sarcomas, including four leiomyosarcomas and one malignant GIST.6,8 The benign tumors of the gastrointestinal tract in the other study included five neurogenic tumors, three inflammatory myofibroblastic tumors (IMT), two hemangiomas, and one teratoma.7 Although hepatocellular carcinoma has an increased frequency in Africa and East Asia, where the hepatitis B and C virus are endemic, its incidence is rising in the United States as well.9 Otherwise, the malignant epithelial neoplasms of the digestive tract in children have not shown any specific geographic preference. Stiller noted, for instance, that “colorectal carcinoma was extremely rare virtually everywhere in the world” in children.10
CARCINOID TUMORS Carcinoid tumors are derived from the endodermally derived enterochromaffin cells that comprise the diffuse neuroendocrine system of the gastrointestinal tract and its foregut derivative, the respiratory tract.11 The widespread distribution of neuroendocrine cells in the gastrointestinal tract explains the occurrence of carcinoid tumors throughout the length of the intestinal tract and two of its derivative organs, the liver and pancreas. In the gastrointestinal tract of children, the appendiceal carcinoid has the distinction, as does
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
750
PEDIATRIC MALIGNANCIES
its bronchial counterpart, of being the most common primary epithelial malignancy. Bronchial carcinoids are discussed in Chapter 67, Uncommon Pediatric Tumors of the Thorax. Other rare sites for carcinoids in children include the prostate in a child with multiple endocrine neoplasia IIb and the sacrococcygeal region in a child with recurrent sacrococcygeal teratoma.12,13 Carcinoid tumors are recognized for their ability to elaborate 5-hydroxyindoleacetic acid (5-HIAA).14 A large tumor draining into the systemic circulation may sufficiently perturb the body’s physiology to cause the carcinoid syndrome, described so poetically by Dr William Bean in 1955: “This witch’s brew of unlikely signs and symptoms, intriguing to the most fastidious connoisseur of clinical esoterica. . . The skin underwent rapid and extreme changes. . . resembling in clinical miniature the fickle phantasmagoria of the Aurora Borealis.”12,15 The carcinoid syndrome heralds the later stages of malignant carcinoid disease. In the majority of patients with the syndrome, the primary tumor, usually present in the small intestine, has metastasized to the liver. Two-thirds of such patients have obvious clinical signs of cancer and the remaining one-third have mass disease that is easy to locate using standard radiographic techniques. The major elements of carcinoid syndrome are diarrhea (usually watery and profuse) and flushing. Asthma and pellagra are less common. Carcinoids may produce other hormones such as adrenocorticotropin hormone (ACTH) or antidiuretic hormone, which have their own attendant manifestations.12 Carcinoid syndromes are rarely observed in children, because carcinoids rarely metastasize in young individuals. A child with copious watery diarrhea of the type encountered in the carcinoid syndrome is more likely to have a vasoactive intestinal peptide-producing neuroblastic tumor than a carcinoid.16,17 There is one well-documented case of a “functioning” carcinoid in a 15-year-old patient, which was reported by Field and associates. The patient initially presented with “episodes of abdominal cramping with loud growling that actually could be heard throughout the house and often awakened his parents in another room.” Periodic flushing and dyspnea followed these episodes. The patient eventually developed dryness and flaking of the skin with pellagra-like features and a “constant malar flush was noted to have a peculiar cyanotic hue.” Twenty-four hour urine collections revealed increased levels of 5-HIAA. There was no evidence of heart disease. At surgery, the patient had a 2-cm carcinoid of the ileum and multiple hepatic metastases.18
Carcinoid Tumor of the Appendix Carcinoid tumor of the appendix was first described in 1882 by Beger,19 and the designation “Karzinoide” was assigned to the tumor by Oberndorfer in 1907.20 Gastrointestinal carcinoids are divided into foregut (stomach and duodenum, 10% of cases), midgut (jejunum, ileum, and appendix, 65–80% of cases), and hindgut tumors (colon and rectum, 10–25% of cases).11 In children, over 90% of all gastrointestinal carcinoids occur in the appendix.21 Appendiceal carcinoids appear to be histologically distinct from those found elsewhere in the
Figure 1 Carcinoid of the appendix in a 14-year-old male, showing solid nests of tumor in the submucosa beneath the normal glandular mucosa. H&E.
gut. The tumors appear to grow in the lamina propria beneath the epithelial crypt, with no connection to the endocrine cells within the crypts of Lieberk¨uhn. Strong immunoreactivity with S-100 protein is noted in lesions occurring in the appendix, while lesions elsewhere lack this finding.22 The incidence of appendiceal carcinoids in children varies from 0.08 to 0.3% in appendectomy specimens. Several studies have shown a female predilection of 1.6 to 2.1, but the male to female ratio is equal when all age-groups with gastrointestinal carcinoids are included in the analysis.9,23,24 The tumor is an incidental and unexpected finding on examination of the appendix, in most pediatric cases. A firm, yellowish tan mass measuring 0.5–2.0 cm in diameter at the tip of the appendix is the typical gross appearance. Tumors less than 0.5 cm may not be recognized until the microscopic examination. If the appendix is also involved by acute appendicitis, tumors less than 0.5 cm may be obscured by the accompanying purulent and hemorrhagic changes. Obstruction of the appendiceal lumen by the tumor is an uncommon finding. Histologically, small cords, strands, and insular profiles of cohesive, small, uniform-appearing tumor cells are present in the lamina propria beneath an intact, unremarkable-appearing mucosa, except in those cases with acute inflammation and mucosal ulceration in the vicinity of the tumor (see Figure 1). Individual groups of tumor cells infiltrate into the muscularis propria to the serosal surface and into the periappendiceal soft tissues in 70% or more of cases. Lymphatic invasion is found in most cases and microscopic metastasis to a periappendiceal lymph node may
UNCOMMON TUMORS OF THE GASTROINTESTINAL TRACT IN CHILDREN
be identified as well. Despite the presence of pathologic findings that are usually regarded as having prognostically unfavorable implications, the clinical outcome in children is excellent in virtually 100% of cases. One cautionary note is the need to differentiate pathologically between the typical appendiceal carcinoid and the adenocarcinoid or goblet cell carcinoid with mucin-containing signet ring cells. The latter tumor is uncommon in general and rare in children. Unlike the classic carcinoid of the appendix with its excellent prognosis, the clinical outcome may be poor in the case of the adenocarcinoid.25 The majority of pediatric patients present with clinical symptoms and signs consistent with appendicitis. Approximately 80% of children in the author’s Mayo Clinic series presented with an acute abdomen. Three patients had abdominal pain for a duration of approximately 6 months. Symptoms of carcinoid syndrome or Cushing’s syndrome were not described in any of these patients. Indeed, the carcinoid tumor was an unexpected finding in each case. In the majority of cases, the tumor was 1 cm or less in greatest dimension; only about 14% of patients had a tumor greater than 2 cm in size. Nearly 70% of appendiceal tumors invaded up to the serosa or beyond, and lymphatic invasion was a universal finding. With a median follow-up time of 26 years, no child experienced a recurrence or metastasis.20 Two other recent series have corroborated these findings.26,27 Simple appendectomy is curative in the overwhelming majority of cases. Data derived from an adult series28 suggest that tumors greater than 2 cm in greatest dimension have a propensity to metastasize. In such cases, right hemicolectomy is recommended as an anxiety-relieving procedure. Corpron and associates have questioned the necessity for a hemicolectomy in children, since the favorable outcome seemed unaltered by the size of the tumor in the appendix.21 Further postoperative workup, such as “second-look” surgery for lymph node sampling, computed tomography (CT) scans of the abdomen searching for hepatic metastases, or following 5-HIAA as a tumor marker are all burdensome and anxietyproducing for the patient and the parent, and not indicated.12
Carcinoid Tumor of the Rectum Most hindgut carcinoids are found in the rectum. Rectal carcinoids differ from those found elsewhere in the gut due to the fact that they do not take up silver (i.e. are not argentaffinomas) and do not show histochemical evidence of serotonin production.12 On hemotoxylin and eosin stain, however, they take on a histologic appearance identical to other carcinoid tumors. Caldarola and colleagues have described three basic histologic patterns: scirrhous, “rosette,” and “ribbon.”29 These tumors, like the appendiceal carcinoids, are often discovered incidentally as asymptomatic lesions. Presentation with cancer symptoms is exceedingly rare, and because the tumor does not produce 5-HIAA, carcinoid syndrome as a result of rectal primaries is essentially nonexistent.12 Few descriptions of carcinoid tumor of the rectum in adolescents are available in the literature. Bates identified three cases in the 10- to 19-year-old age-group (1.6% of cases).30 Two additional cases have been described, both boys 10 and 15 years of age, who presented with rectal
751
bleeding, an unusual presentation even in adults.31,32 The tumor in the 10-year-old boy measured 3 × 7 cm; in the other boy it was a polypoid lesion measuring 1.5 cm. No description of significant clinical follow-up was provided for either case. As is the case for carcinoids of the appendix, size is the factor guiding surgical therapy of rectal carcinoids. Tumors less than 1 cm in greatest dimension are unlikely to metastasize, and may be treated with fulguration or local excision.12,25 Tumors between 1 and 2 cm should be evaluated for depth of invasion. In those with limited invasiveness, wide excision is often adequate treatment.12 An aggressive surgical approach is recommended for those tumors with invasion beyond the muscularis mucosa and/or with a maximum dimension of 2 cm or greater. A long period of clinical follow-up is necessary, since these tumors may present with recurrent or metastatic disease many years after the original resection.33 Carcinoids of the gastrointestinal tract may present as multifocal tumors. La Ferla and colleagues reported an 11year-old boy with multiple colonic primaries who, at the time of initial surgery, had involvement of the anterior rectum with rectal bleeding. After 3 months, the tumor had metastasized widely, and ultimately led to his demise.34 One may conclude from this limited number of cases that carcinoid tumors of the rectum, in contrast with those found in the appendix, present at a more advanced stage and are a greater threat to life in children and adolescents. Gastric and small intestinal carcinoids are rarely described in children.9,35
COLORECTAL CARCINOMA Carcinoma of the colon is one of the most common cancers in the Western hemisphere, with an incidence of approximately 35 cases per 100 000 individuals. Surveillance, Epidemiology, and End Results (SEER) data have documented an annual incidence of fewer than 1 per million young individuals under the age of 20.36,37 Of the nearly 150 000 newly diagnosed cases of colorectal carcinoma each year, only about 80 are diagnosed in the first two decades of life.38 The limited experience with this malignancy in the pediatric population complicates its detection and appropriate therapy. Colorectal carcinoma in the pediatric age-group is generally confined to the second decade of life, and can be regarded as a malignancy of adolescence.39 – 43 The median age at diagnosis ranges from 15 to 19 years, and there is a slight male predilection. Abdominal pain, changing bowel habits, rectal bleeding, and anorexia are the principal presenting symptoms.44 – 46 Although it is difficult to ferret out the etiology of common and nonspecific complaints in the adolescent age-group, the primary care provider must take such physical complaints seriously. Associations have been described between colorectal carcinoma and Gardner syndrome,47,48 Turcot syndrome,49,50 familial adenomatous polyposis,51,52 Peutz –Jeghers syndrome,53,54 and the Cronkhite –Canada syndrome;55 but most pediatric cases are sporadic in occurrence.47,56,57 An association with neurofibromatosis and polyposis coli has been noted.58 Carcinoma of the colon has been reported as
752
PEDIATRIC MALIGNANCIES
a second malignant neoplasm in a child who previously received radiotherapy for Wilms’ tumor.59 Parc and colleagues recently reviewed the outcome of patients with familial adenomatous polyposis who underwent restorative coloproctectomy. Although the risk of colorectal cancer was eliminated following this procedure, 13% of patients developed mesenteric desmoid tumors. Desmoid tumors caused death in two patients and made a significant functional impact on the survivors. Mesenteric desmoid tumors were the primary cause of death following surgery in this patient group.60 Much attention has been paid to the fact that pediatric cases of colorectal carcinoma present at a more advanced stage when compared with adult cases.3,45,46,49,61 – 68 Approximately 60% of pediatric cases reported stage IV (Duke D) carcinoma compared to 20% observed in adult cases (see Table 1).3,49,61 – 66 This phenomenon has been attributed to a lack of screening, or failure to recognize symptoms attributable to colon cancer. However, several important clinicopathologic distinctions appear to exist between carcinomas of the large intestine in adolescents versus those in adults. The location of the primary tumor in children tends to be right-sided,3,49,66,67,69,70 with over 45% of tumors occurring proximal to the splenic flexure versus 17% in adults. Another important difference is that 30–70% of colonic carcinomas in children have mucinous or signet ring features, with associated poor prognosis (see Figures 2 and 3). Most adenocarcinomas of the colon in adults are well- to moderately differentiated neoplasms, and only 10–15% of cases have a poorly differentiated mucinous or signet ring pattern.3,49,66,67,71,72 Special attention must be given to the fact that a young person presenting with colorectal carcinoma may represent the proband of a kindred afflicted with hereditary nonpolyposis colon cancer (HNPCC) syndrome. The Amsterdam criteria for identification of these kindreds are as follows: (i) at least three relatives should have histologically verified colorectal cancer, with at least two of them being first-degree relatives; (ii) at least two successive generations should be affected; (iii) in one of the relatives (or the pediatric index case), colorectal cancer should be diagnosed at under 50 years of age.73 There have been some molecular genetic studies that link the pathogenesis of sporadic colorectal carcinomas in children to HNPCC.74 Children with colorectal carcinomas share specific common features with adult HNPCC patients and sporadic colon cancer patients, having defects of DNA Table 1 Colorectal carcinoma staging criteria: adult and pediatric cases.
TNM staging66
Dukes staging61
Adult cases (%)64,65
Pediatric cases (%)3,49,62,63
A
15 – 19
0
B
33
11
C
28
26
D
20
63
Stage 0 Tis Stage IT1 T2 Stage II T3 T4 Stage III Any T N Any T Stage IV Any T M TNM, Tumor-Node-Metastasis.
Figure 2 Adenocarcinoma of the transverse colon in an 18-year-old female who presented with intestinal obstruction. The cross-section of this 4.4 × 5.3 cm exophytic sessile mass shows several mucinous pools at the base of the tumor.
Figure 3 Adenocarcinoma of the transverse colon in an 18-year-old female, showing infiltrating malignant glands with abundant mucin filling and distending the glands. H&E.
mismatch repair as reflected by microsatellite instability. HNPCC has been shown to result from a defect in enzymes responsible for DNA mismatch repair, a requirement for faithful replication of the genome.75,76 Loss of activity in one of these genes results in genomic instability and assayable expansion or contraction of microsatellites. HNPCC tumors show marked microsatellite instability at multiple loci.75,77,78 They also show a remarkable predilection for the right colon,78 as is true of pediatric colon carcinomas reported in the literature. The basic element of therapy for all patients with colorectal carcinoma is optimal resection of the tumor(s) by an experienced surgeon. Foglia has reviewed the surgical management of these tumors in adolescents; the approach is similar in most respects to resection of colon carcinomas in adults.79 Consultations with a colorectal surgeon is appropriate in all cases. Staging is accomplished in concert with the consulting pathologist. The requirement for adjuvant therapy is based on the stage of disease.80,81 Currently, Dukes A (stage I) disease (tumor confined to the bowel wall) is cured with surgery alone, as are the majority of Dukes B lesions (tumor extension through the bowel wall and into the mesentery). The chemotherapeutic agent 5-fluorouracil (5-FU) remains the
UNCOMMON TUMORS OF THE GASTROINTESTINAL TRACT IN CHILDREN
fundamental drug in the management of more advanced disease. The use of leucovorin in combination with 5-FU has improved therapeutic outcomes for patients with highrisk Dukes B and C disease.82,83 This success has not been extended to patients with distant metastases. Some benefit is experienced, however, with combination chemotherapy incorporating irinotecan hydrochloride (CPT-11), which acts as a specific inhibitor of DNA topoisomerase I, and has shown promise in patients with advanced-stage and recurrent colorectal cancer.84,85 This may represent a significant improvement over the response rate of 11% observed in patients with Dukes D colon carcinoma receiving 5-FU as a single agent or in combination, as first-line or salvage therapy.86 More recently, oxaliplatin has been employed with 5-FU and leucovorin for patients with advanced-stage colon cancer. The regimen is well tolerated in adults and contributed to a significant improvement in disease-free survival for patients with stage II and stage III colorectal cancer.87 Given the mode of action and toxicities of the abovementioned chemotherapeutic agents, regimens with increased dose intensity may be considered for younger patients, but have not been studied to date. It must be emphasized again that appropriate initial therapy for juvenile colorectal carcinoma depends on the skill and experience of the colorectal surgeon who undertakes surgery with curative intent. SEER-based survival rates for localized and regional colon carcinoma climbed steadily during the era (1940–1981) prior to proof of effective adjuvant therapy, no doubt as a result of improved and more aggressive surgical care benefiting these patients. Recent experience has been gained with aggressive resection of hepatic metastases. This may lead to improved survival in selected patients with metastatic disease.88 Increased participation in therapeutic trials for colorectal cancer would contribute to the development of improved treatment options for this disease.
OTHER GASTROINTESTINAL NEOPLASMS Other types of neoplasms in the gastrointestinal tract of children are generally reported as single-case reports or in small series. Carcinoma of the stomach is one such neoplasm.89 – 93 Stromal mesenchymal neoplasms include the smooth muscle tumors (leiomyoma and leiomyosarcoma), GIST and its variant, the gastrointestinal autonomic nerve tumor, inflammatory myofibroblastic tumor, and angiosarcoma. An epithelioid smooth muscle tumor of the stomach in an adolescent female should suggest the possibility of the Carney’s triad. Any young female presenting with gastric leiomyosarcoma should undergo an intensive search for the other components of the triad: extra-adrenal paraganglioma and pulmonary chondroma.94 Diffuse leiomyomas of the esophagus may be a manifestation of Alport syndrome.95 Solitary or multiple smooth muscle tumors in the gastrointestinal tract are found in association with the acquired immunodeficiency syndrome (AIDS).96
Gastrointestinal Stromal Tumor (GIST) GISTs typically present in the stomach or small intestine.97,98 DeMatteo and colleagues noted their occurrence also in
753
the esophagus, liver, peritoneum, colon, and rectum. GIST occurs predominantly in adults over 40 years of age, with a median age of 58 reported in this series.99 However, GISTs have been reported at all ages; the youngest child to be reported was diagnosed at 1 day of age with a primary tumor in the terminal ileum.100 The chief presenting symptoms associated with pediatric GISTs include abdominal pain, melena, and fatigue. The patient is frequently found to have a palpable abdominal mass and anemia, which may be severe.101 – 103 The two reported neonatal cases presented with abdominal distension secondary to intestinal obstruction.100,104 The tumor has been linked to the Carney’s triad.105,106 These tumors vary in size and histologic appearance.107 Even though the microscopic features are consistent with a smooth muscle or neurogenic neoplasm, the immunophenotype is often restricted to vimentin, CD34, and CD117 expressions. The prognosis is affected by the size (< or >5 cm) and mitotic activity. It is difficult in some cases to predict the prognosis on the basis of the pathologic findings, and the tumor is appropriately interpreted as one with an “indeterminant malignant potential.”98,108 All GISTs are now found to be positive for c-KIT expression (CD117). The c-KIT expression is generally a result of a mutation in exon 11 of KIT. Other mutations have been described in exons 9, 13, and 17.109 Interestingly, tumors in patients younger than 20 years of age rarely have a detectable mutation. Prakash and associates reported one of six patients under 20 years of age with an exon 9 mutation. The other five patients had wild-type KIT.106 Price et al. described six patients with GIST, three of whom underwent mutational analysis: only one had a mutation – at exon 9.110 Up-regulation of KIT expression is seen in all GISTs, whether or not a detectable mutation is present. Exploitation of this transmembrane tyrosine kinase receptor by adjuvant therapy with imatinib mesylate (Gleevec), an orally bioactive tyrosine kinase inhibitor, offers promising results.111 DeMetri et al. have reported the use of imatinib in the largest series to date, with an estimated 1-year survival of 88%; the median survival had not been reached at the time of their publication. Response or resistance to imatinib was readily detected with [18 F] fluoro-2-deoxy-D-glucose positron emission tomography (PET) imaging.112 Despite the first blushes of success with imatinib, it should be emphasized that the primary therapy for GIST is surgical extirpation of the tumor and all easily resectable metastatic sites.113 Gastrointestinal autonomic nerve tumors are found in the stomach, proximal small intestine, and retroperitoneum.114 These tumors are recognized in the second decade of life or beyond. Some of the histologic findings overlap with the GIST, but neural differentiation is apparent by electron microscopy and/or immunohistochemistry.115 – 117 These tumors are often large (>5 cm), and pursue an aggressive clinical course in most cases. Complete surgical resection is the principal method of treatment. IMT presents in several sites in the abdominal cavity with a predilection for the distal small intestine and/or contiguous mesentery.118 These tumors have the same pathologic features as their
754
PEDIATRIC MALIGNANCIES
pulmonary counterparts (see Chapter 67, Uncommon Pediatric Tumors of the Thorax). It is important to differentiate the IMT from a smooth muscle neoplasm of the gastrointestinal tract, since the size, degree of cellularity, and mitotic activity of an IMT are the features in common with a leiomyosarcoma. A mitotically active spindle cell neoplasm in the gastrointestinal tract of a neonate may also cause concern about a stromal sarcoma, but most of these tumors behave in a benign manner. These tumors are related to the congenital-infantile fibrosarcoma and cellular mesoblastic nephroma with the t(12;15)(p13;q26) translocation. We have also seen kaposiform hemangioendothelioma present as multiple colonic tumors in a neonate.
TUMORS OF THE PANCREAS Primary neoplasms of the pancreas, regardless of their specific pathologic type and behavior, are rare in children.119 Although Kopelman reports that less than 5% of all malignancies diagnosed in children under the age of 15 arise in the pancreas,120 it appears that the true incidence is less than 1%. Grosfeld and associates reported their 20-years experience with 13 pancreatic tumors in children, 6 of which were malignant, including one each of ductal carcinoma, acinar cell carcinoma, pancreatoblastoma, and solidpseudopapillary tumor.121 There were two additional cases of rhabdomyosarcoma.122 Six pancreatic neoplasms were seen at the Toronto Hospital for Sick Children over a 20-year period. It was estimated that only 1 in 18 000 general surgical cases was a pancreatic neoplasm, at that institution.122 The six tumors in that review were: three solid-pseudopapillary tumors, a pancreatoblastoma, an islet cell carcinoma, and an insulinoma. Six pancreatic neoplasms were identified in another study of 35 nonlymphoid gastrointestinal malignancies in children.3 One islet cell neoplasm was found by Al-Sheyyab and co-workers among 136 children who were diagnosed with malignant epithelial tumors in various anatomic sites.2 There were five “pancreatic carcinomas” in a registry-based study of 234 children with the generic diagnosis of “carcinoma” in a 10-year period in the United Kingdom; these five tumors were not further characterized pathologically.5 On the other hand, Lack and associates reported the clinicopathologic features of eight neoplasms of the exocrine pancreas in patients between the ages of 15 months and 18 years at the time of diagnosis.123 There were three ductal adenocarcinomas, two acinar cell carcinomas, and one solid-pseudopapillary tumor. The two benign tumors were acinar cell adenomas. Neoplasms of the pancreas are classified histogenetically into exocrine and endocrine tumors. Those arising from the exocrine pancreas include the ductal and acinar cell tumors.124 Islet cell tumors comprise the endocrine neoplasms of the pancreas. Two tumor types of importance in the pediatric age-group, pancreatoblastoma and solidpseudopapillary tumor, have been classified with the exocrine pancreatic neoplasms, though both may contain minor populations of tumor cells with neuroendocrine features by electron microscopy and/or immunohistochemistry as an implication of their primordial or dysontogenetic nature.
Ductal type adenocarcinoma is the most common primary malignancy of the pancreas in adults and accounts for 85–90% of all cases. This may also be the most common primary malignancy of the pancreas in children, but such a conclusion is a tentative one, since it is based on case studies rather than epidemiologic data125 (see Table 2).121,123,126 – 141 This tumor may present with regional lymph node and hepatic metastases at the time of diagnosis. Symptoms and signs most frequently noted include the presence of an epigastric mass (50%), abdominal pain, anorexia, icterus, vomiting, and weight loss.142 Acinar cell carcinoma is the other malignant exocrine neoplasm of the pancreas and constitutes only 1–2% of exocrine tumors.143 This tumor type has been recognized in children, although like ductal adenocarcinomas, most cases have been diagnosed in adults. There is a variant, the mixed acinar-endocrine carcinoma of the pancreas. The importance of these tumors is the overlapping pathologic similarity in some cases to pancreatoblastomas and solid-pseudopapillary tumors that occur in young children and adolescents/young adults, respectively. Pancreatoblastoma (infantile pancreatic adenocarcinoma) is the dysontogenetic neoplasm of the pancreas that is analogous in a histogenetic sense to Wilms’ tumor of the kidney and neuroblastoma of the adrenal gland.144 This tumor occurs almost exclusively in children who are typically less than 10 years of age at diagnosis, with an average age range of 2 to 4 years.120,145 There is a male predilection, and Asian children appear to be at a greater risk for the development of these tumors.134,146,147 Pancreatoblastoma has been reported in association with the Beckwith–Wiedemann syndrome,148 as have Wilms’ tumor, adrenal cortical neoplasms, and hepatoblastoma. Approximately 70 cases of pancreatoblastoma had been reported through 1997.146,148 Very few cases have been studied for cytogenetic abnormalities, but Wiley and associates have reported two translocations, t(13;22)(q10;q10) and t(13;13)(q10;q10), in a pancreatoblastoma from a 4-year-old boy.149 An epigastric mass with or without accompanying diarrhea is the usual clinical presentation. Obstructive jaundice is unusual. Like the hepatoblastoma and endodermal sinus tumor, the serum α-fetoprotein level may be markedly elevated. Grossly, these tumors are well circumscribed, measuring 6–15 cm in their greatest dimension, and are found throughout the pancreas without the regional preference of certain pancreatic neoplasms. The larger tumors may occupy the entire pancreas. Microscopically, these tumors have mixed histologic features with acinar, trabecular, and small ductal-like profiles of uniformappearing cells. The pancreatoblastoma may be mistaken pathologically for an islet cell tumor. A characteristic feature is the presence of the squamous foci as nodules of nonkeratinizing cells (see Figure 4). Although these tumors are malignant, they often have an indolent clinical course. Liver metastasis may be discovered at diagnosis in 30–35% of cases. The overall survival is 50–65% and tumor-related deaths are uncommon beyond a 4-year disease-free interval. Solid-pseudopapillary tumor (solid-cystic neoplasm of the pancreas, Frantz tumor) is an uncommon yet distinctive neoplasm of the pancreas that occurs almost exclusively
UNCOMMON TUMORS OF THE GASTROINTESTINAL TRACT IN CHILDREN
Figure 4 Pancreatoblastoma in a 7-year-old Asian female, showing a solid squamous-like nest of tumor and an adjacent normal islet of Langerhans. These tumors have a prominent fibrous background. H&E.
in adolescent and young adult females.150,151 As with the pancreatoblastoma, the incidence appears to be higher in Asian children. A large, often asymptomatic, epigastric mass is the clinical presentation. These tumors are well circumscribed, often encapsulated masses, measuring 5–15 cm in their greatest dimension. There is a predilection for the body and tail of the pancreas. One tumor has been reported in a 13-year-old girl, which arose in the ectopic pancreas of the mesocolon.152 Invasion into the stomach and/or duodenum may be encountered at surgery. The macroscopic appearance shows a grossly cystic and hemorrhagic mass with focal areas of solid grayish white tissue. Solid sheets of uniform vacuolated-to-eosinophilic tumor cells and discohesive, pseudopapillary profiles of cells are the principal microscopic findings. The solid areas may have a trabecular perithelial appearance, whereas other foci have pseudoglandular features with hyaline –myxoid stroma (see Figure 5). Tumor-related deaths occur in less than 25% of cases.153 The basic management of these three pancreatic neoplasms (ductal adenocarcinoma, pancreatoblastoma, and the solidpseudopapillary tumor) is surgical in nature, but the extent of resection varies with the clinical behavior of the particular tumor type. Pancreatoduodenectomy remains the procedure
755
Figure 5 Solid-pseudopapillary tumor of the pancreas, showing solid nests of uniform basophilic cells with pale staining mucoid foci among the tumor cells. H&E.
of choice in the treatment of ductal adenocarcinoma. The procedure is often precluded by the discovery of metastatic disease, as was the case in 14 of 16 children included in our review (see Table 2).127,128 Effective adjuvant therapy for advanced pancreatic carcinoma has not yet been developed. Radiotherapy may provide relief in selected cases and this effect may be enhanced with the use of 5-FU.154 Complete surgical resection is the treatment of choice for pancreatoblastoma. If the tumor is located in the body or tail and has not directly extended into contiguous structures or metastasized to the liver, a complete excision may be possible without a pancreatoduodenectomy. In those patients with hepatic metastases and/or unresectable tumor at presentation, chemotherapy with or without radiation therapy has yielded favorable results, with tumor shrinkage that allowed for total surgical resection.133 Adriamycin, cyclophosphamide, and cisplatin, as well as other agents, have been administered with variable success.133 – 135 Cystic-pseudopapillary tumors of the pancreas are best managed by complete resection, but the size and friability of these neoplasms, in addition to adhesions to surrounding
Table 2 Tumors of the pancreas.
Type (number)
Ages
Gender
Location
Adenocarcinoma (31 cases)121,123,126,127
17 months – 17 years; median: 11 years
Acinar cell carcinoma (5 cases)121,123,128 – 130 Pancreatoblastoma (23 cases)121,126,131 – 135
15 months – 13 years; median: 9 years 2 – 14 years; median: 4 years
M = 12 F = 18 Unknown = 1 M=4 F=1 M = 13 F = 10
Papillary (cystic)-solid type (12 cases)121,126,136 – 141 Islet cell carcinoma (4 cases)127
2 – 15 years; median: 12 years 3 – 10 years; median: 7 years
M=2 F = 10 M=1 F=3
Head = 10 Body = 4 Tail = 5 Head = 3 Tail = 1 Head = 9 Body = 1 Tail = 6 Head = 3 Tail = 5 Tail = 3
a
Follow-up based on available data.
Metastases at diagnosis
Outcomea
Yes = 14 No = 2
Alive = 4 Dead = 24
Yes = 1 No = 3 Yes = 13 No = 4
Alive = 1 Dead = 4 Alive = 12 Dead = 9
Yes = 1 No = 2 –
Alive = 8 Dead = 2 Alive = 2 Dead = 1
756
PEDIATRIC MALIGNANCIES
Table 3 Islet cell tumors of the pancreas.14,156
Hormone
Tumor
Syndrome
Symptoms
Glucagon
Glucagonoma
Diabetes-dermatitis syndrome
Insulin
Insulinoma
Somatostatin
Somatostatinoma
Hypoglycemia syndrome “Whipple’s triad”: (i) hypoglycemic attack (ii) decreased blood glucose (iii) relief with oral or IV glucose Inhibitory syndrome
Diabetes mellitus, dermatitis (necrotizing, migratory erythema), ileus, constipation, anemia, glossitis, cheilitis, venous thrombosis, mental aberrations Headache, blurred vision, incoherence, convulsions, sweating, weakness, hunger, palpitations
hPP Serotonin
Pancreatic polypeptide tumor Carcinoid-islet cell tumor
MEN 1 (MEA-I) Atypical carcinoid syndrome
Gastrin
Gastrinoma
Zollinger – Ellison syndrome
VIP
VIPoma
ACTH
Pancreatic corticotropinoma
Watery diarrhea syndrome (Verner – Morrison syndrome) (Ectopic-ACTH syndrome)
Parathyroid hormone
Pancreatic parathyroidoma
Ectopic-hypercalcemia syndrome
Diabetes mellitus, cholecystolithiasis, steatorrhea, hypochlorhydria, anemia Severe and prolonged facial flush, hypotension, periorbital edema, lacrimation Acid hypersecretion (hypergastrinemia, hyperchlorhydria) Diarrhea (pancreatic cholera), hypokalemia, hypochlorhydria Obesity, hyperpigmentation, hypertension, glucose intolerance Hypercalcemia, hypophosphatemia
IV, intravenous; hPP, human pancreatic polypeptide; MEN 1, multiple endocrine neoplasia type 1; MEA-I, multiple endocrine adenomatosis type I; VIP, vasoactive intestinal peptide; VIPoma, vasoactive intestinal peptide producing tumor; ACTH, adrenocorticotropin hormone.
structures, are the factors that may complicate the surgical resection. Since these tumors are often more distal in the pancreas, a segmental resection may be possible. Pancreatoduodenectomy may be reserved for recurrent or more advanced disease. The indolent clinical behavior of these neoplasms is manifested in some cases by locally recurrent tumors many years after the primary excision. It has been suggested that cystic-pseudopapillary tumors that are diagnosed in children may have an even more indolent course than those in adults.139 There is relatively limited experience with chemotherapy and radiation therapy in the management of this tumor, since these are so infrequently needed. Islet cell neoplasms include the adenomas and carcinomas whose pathologic discriminants from each other are not sharply defined in all cases. These tumors constitute a larger proportion of all pancreatic neoplasms in children, in comparison to the adult experience. However, islet cell carcinoma is one of the rare malignant endocrine neoplasms of childhood.155 The possibility of multiple endocrine neoplasia type I should be considered in a child with a non– β cell islet cell tumor.156 Some islet tumors synthesize a single hormonal peptide and others are multipeptide producing neoplasms.157 These generally small tumors may create physiologic perturbations far out of proportion to their size (see Table 3).14,158 Most insulinomas are adenomas and require only surgical resection. Most gastrin-producing islet cell tumors of the Zollinger –Ellison syndrome are carcinomas with regional lymph node and/or liver metastases at diagnosis.159 Some islet cell carcinomas are clinically nonfunctional, but the tumor cells may contain one or more stainable peptides by immunohistochemistry.160 These tumors are multifocal in the pancreas in some cases.
For cases in which surgical cure cannot be accomplished, medical management may be appropriate. Somatostatin, or its long-lasting analog Octreotide, has been used with beneficial results in adults with tumors which produce insulin, serotonin, and/or VIP, gastrin, and glucagon.161 Overall hormone response rates (partial and complete) ranged from 57 to 80%. Rarely, this therapy may provide the bonus of actual tumor regression as described by Moertel,14 and Kraenzlin and colleagues.162 Symptomatic therapy may be provided by agents specific to the hormonal effects of the tumor. Corticosteroids or cyproheptidine (Periactin) have also proven useful for adjuvant symptomatic therapy. Patients with Zollinger –Ellison syndrome benefit from therapy with specific H2 -antagonists such as cimetidine, ranitidine, or imiprazole. Total gastrectomy may be required in some cases. Occasionally, regression of metastatic gastrinoma has been observed following gastrectomy.14 Other neoplasms that may present in or around the pancreas of a child are Ewing’s sarcoma/primitive neuroectodermal tumor and neuroblastoma.163 The pancreas is also one of the common metastatic sites for alveolar rhabdomyosarcomas.
ACKNOWLEDGMENT We wish to thank Emily Place for her assistance, and to gratefully acknowledge support from the Pine Tree Apple Tennis Classic Oncology Research Funds.
REFERENCES 1. Jemal A, et al. Cancer statistics, 2003. CA Cancer J Clin 2003; 53: 5 – 26. 2. Al-Sheyyab M, et al. Malignant epithelial tumours in children: incidence and etiology. Med Pediatr Oncol 1993; 21: 421 – 48.
UNCOMMON TUMORS OF THE GASTROINTESTINAL TRACT IN CHILDREN 3. B¨uy¨ukpamuk¸cu M, et al. Carcinoma of the colon in children. Turk J Pediatr 1996; 38: 51 – 8. 4. Harms D, Schmidt D. Rare tumors in childhood: pathologic aspects. Experience of the Kiel Pediatric Tumor Registry. Med Pediatr Oncol 1993; 21: 239 – 48. 5. McWhirter WR, Stiller CA, Lennox EL. Carcinomas in childhood. A registry-based study of incidence and survival. Cancer 1989; 63: 2242 – 6. 6. Bethel CAI, et al. Alimentary tract malignancies in children. J Pediatr Surg 1997; 32: 1004 – 9. 7. Skinner MA, et al. Gastrointestinal tumors in children: an analysis of 39 cases. Ann Surg Oncol 1994; 1: 283 – 9. 8. Hameed R, et al. Paediatric malignant tumors of the gastrointestinal tract in the West Midlands, UK, 1957 – 2000: a large population based survey. Pediatr Blood Cancer 2004; 43: 257 – 60. 9. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999; 10: 745 – 50. 10. Stiller CA. International variations in the incidence of childhood carcinomas. Cancer Epidemiol Biomarkers Prev 1994; 3: 305 – 10. 11. Lewin KJ, Riddell RH, Weinstein WM. Gastrointestinal Pathology and its Clinical Implications. New York: Igaku-Shoin, 1992: 197 – 257. 12. Stringer DA, et al. Malignant carcinoid within a recurrent sacrococcygeal teratoma in childhood. Can Assoc Radiol J 1990; 41: 105 – 7. 13. Whelan T, et al. Primary carcinoid of the prostate in conjunction with multiple endocrine neoplasia IIb in a child. J Urol 1995; 153: 1080 – 2. 14. Moertel CG. An Odyssey in the land of small tumors. J Clin Oncol 1987; 5: 1503 – 22. 15. Bean WB, Olch D, Wienberg HB. The syndrome of carcinoid and acquired valve lesions of the right side of the heart. Circulation 1955; XII: 1 – 6. 16. Davies RP, Slavotinek JP, Dorney SF. VIP secreting tumours in infancy. A review of radiological appearances. Pediatr Radiol 1990; 20: 504 – 8. 17. Kimura N, et al. Multiple-hormone gene expression in ganglioneuroblastoma with watery diarrhea, hypokalemia, and achlorhydria syndrome. Cancer 1993; 71: 2841 – 6. 18. Field JL, Adamson LF, Stoeckle HE. Review of carcinoids in children. Functioning carcinoid in a 15-year-old male. Pediatr 1962; 29: 953 – 60. 19. Beger A. Ein Fall von (Krebs) des Wurmfortsatzer. Klin Wochenschr 1882; 19: 616 – 8. 20. Oberndorfer S. Karzinoide tumoren des D¨unndarms. Frankf Z Pathol 1907; 1: 426 – 30. 21. Corpron CA, et al. A half century of experience with carcinoid tumors in children. Am J Surg 1995; 170: 606 – 8. 22. Moertel CL, Weiland LH, Telander RL. Carcinoid tumor of the appendix in the first two decades of life. J Surg Pediatr Surg 1990; 25: 1073 – 5. 23. Parkes SE, et al. Carcinoid tumours of the appendix in children 1957 – 1986: incidence, treatment and outcome. Br J Surg 1993; 80: 502 – 4. 24. Stringer MD. Carcinoid tumours of the appendix in children 1957 – 1986; incidence, treatment and outcome. Br J Surg 1993; 80: 1217. 25. Burke AP, et al. Goblet cell carcinoids and related tumors of the vermiform appendix. Am J Clin Pathol 1990; 94: 27 – 35. 26. Dall’Igna P, et al. Carcinoid tumor of the appendix in childhood: the experience of two Italian institutions. J Pediatr Gastroenterol Nutr 2005; 40: 216 – 9. 27. Spunt SL, et al. Childhood carcinoid tumors: the St. Jude Children’s Research Hospital experience. J Pediatr Surg 2000; 35: 1282 – 6. 28. Moertel CG, et al. Carcinoid tumor of the appendix. Treatment and prognosis. N Engl J Med 1987; 317: 1699 – 701. 29. Caldarola VT, et al. Carcinoid tumors of the rectum. Am J Surg 1964; 107: 844 – 9. 30. Bates HR Jr. Carcinoid tumors of the rectum. Dis Colon Rectum 1962; 5: 270 – 80. 31. Root GT, Leddy W, MacDonald JL. Carcinoid tumors of the rectum. Am J Surg 1959; 98: 243 – 7.
757
32. Gold MS, Winslow PR, Litt IF. Carcinoid tumors of the rectum in children: a review of the literature and report of a case. Surgery 1971; 69: 394 – 6. 33. Janson ET, et al. Carcinoid tumors: analysis of prognostic factors and survival in 301 patients from a referral centre. Ann Oncol 1997; 8: 685 – 90. 34. La Ferla G, et al. Multiple colonic carcinoid tumours in a child. Br J Surg 1984; 71: 843. 35. Burke AP, et al. Carcinoids in the jejunum and ileum. An immunohistochemical and clinicopathologic study of 167 cases. Cancer 1997; 79: 1086 – 93. 36. Ries LAG, et al. (eds) SEER Cancer Statistics Review, 1973 – 1994, NIH publication No. 97 – 2789. Bethesda, Maryland: National Cancer Institute, 1997: 156. 37. Thomas RM, Sobin LH. Gastrointestinal cancer. Cancer 1995; 75: 154 – 70. 38. Pratt CB, Pappo AS. Management of infrequent cancers of childhood. In Pizzo PA, Poplack DG (eds) Principles and Practice of Pediatric Oncology, 4th ed. Philadelphia, Pennsylvania: Lippincott Williams & Wilkins, 2002: 1160. 39. Ford EG. Gastrointestinal tumors. In Andrassy RJ (ed) Pediatric Surgical Oncology. Philadelphia, Pennsylvania: WB Saunders, 1998: 289 – 304. 40. Gips M, Wolloch Y. Carcinoma of the colon in youngsters: a report of three cases. Isr J Med Sci 1993; 29: 730 – 2. 41. Kauffman WM, et al. Imaging features of ovarian metastases from colonic adenocarcinoma in adolescents. Pediatr Radiol 1995; 25: 286 – 8. 42. Leichtner AM, Hoppin AG. Intestinal neoplasms. In Walker WA, et al. (eds) Pediatric Gastrointestinal Disease. Pathophysiology. Diagnosis. Management, 2nd ed. St Louis, Missouri: CV Mosby, 1996: 922 – 936. 43. Taguchi T, et al. Carcinoma of the colon in children: a case report and review of 41 Japanese cases. J Pediatr Gastroenterol Nutr 1991; 12: 394 – 9. 44. Pratt CB, Pappo AS. Management of infrequent cancers of childhood. In Pizzo PA, Poplack DG (eds) Principles and Practice of Pediatric Oncology, 4th ed. Philadelphia, Pennsylvania: Lippincott Williams & Wilkins, 2002: 1161. 45. Karnak I, et al. Colorectal carcinoma in children. J Pediatr Surg 1999; 34: 1499 – 504. 46. Chantada GL, et al. Colorectal carcinoma in children, adolescents and young adults. J Pediatr Hematol Oncol 2005; 27: 39 – 41. 47. Gardner EJ. A genetic and clinical study of intestinal polyposis, a predisposing factor for carcinoma of the colon and rectum. Am J Hum Genet 1951; 3: 167 – 76. 48. Smith DB, Strand JA, Carter PL. Gardner’s syndrome: current concepts and use of a new genetic marker. Mil Med 1988; 153: 289 – 93. 49. Goldthorne JF, Powars D, Hays DM. Adenocarcinoma of the colon and rectum in the adolescent. Surgery 1983; 93: 409 – 14. 50. Turcot J, Despr´es J.-P, St Pierre F. Malignant tumors of the central nervous system associated with familial polyposis of the colon: report of two cases. Dis Colon Rectum 1959; 2: 465 – 8. 51. Sharma AK, Sharma SS, Mathur P. Familial juvenile polyposis with adenomatous-carcinomatous change. J Gastroenterol Hepatol 1995; 10: 131 – 4. 52. Church JM, et al. Teenagers with familial adenomatous polyposis. Dis Colon Rectum 2003; 45: 887 – 9. 53. Giardiello FM, et al. Increased risk of cancer in the Peutz – Jeghers syndrome. N Engl J Med 1987; 316: 1511 – 4. 54. Foley TR, McGarrity TJ, Abt AB. Peutz – Jeghers syndrome: a clinicopathologic survey of the ’Harrisburg family’ with a 49-year follow-up. Gastroenterology 1988; 95: 1535 – 40. 55. Malhotra R, Sheffield A. Cronkhite – Canada syndrome associated with colon carcinoma and adenomatous changes in c-c polyps. Am J Gastroenterol 1988; 83: 772 – 6. 56. Coffin CM, Pappin AL. Polyps and neoplasms of the gastrointestinal tract in childhood and adolescence. Perspect Pediatr Pathol 1997; 20: 127 – 71. 57. Lelli JL, Coran AG. Polyploid disease of the gastrointestinal tract. In O’Neill JA, et al. (eds) Pediatric Surgery, 5th ed. St Louis, Missouri: CV Mosby, 1998: 1283 – 1296.
758
PEDIATRIC MALIGNANCIES
58. Pratt CB, et al. Multiple colorectal carcinomas, polyposis coli, and neurofibromatosis. J Natl Cancer Inst 1988; 80: 1170 – 2. 59. Sabio H, et al. Adenocarcinoma of the colon following the treatment of Wilms tumor. J Pediatr 1979; 95: 424 – 6. 60. Parc Y, et al. Long-term outcome of familial adenomatous polyposis patients after restorative coloproctectomy. Ann Surg 2004; 239: 378 – 82. 61. Dukes CB. The classification of cancer of the rectum. J Pathol Bacteriol 1932; 35: 323 – 32. 62. Rao BN, et al. Colon carcinoma in children and adolescents. A review of 30 cases. Cancer 1985; 55: 1322 – 6. 63. Steinberg JB, Tuggle DW, Postier RG. Adenocarcinoma of the colon in adolescents. Am J Surg 1988; 156: 460 – 2. 64. Moertel CG. Trials, errors, and glimmers of success in the surgical adjuvant treatment of colorectal cancer. In Levin B (ed) 30th Annual Clinical Conference on Cancer, Gastrointestinal Cancer: Current Approaches to Diagnosis and Treatment. Austin, Texas: University of Texas Press, 1988: 3 – 17. 65. Gordon PH. Malignant neoplasma of the colon. In Gordon PH, Nivatongs S (eds) Principles and Practice of Surgery for the Colon, Rectum, and Anus. St. Louis, Missouri: Quality Medical Publishing, 1992: 501 – 590. 66. Schantz SP, Harrison LB, Hong WK. Cancer of the head and neck. In De Vita VT Hellman S, Rosenberg SA (eds) Cancer: Principles and Practice of Oncology. Philadelphia, Pennsylvania: JB Lippincott, 1993: 578 – 579. 67. Vastyan AM, et al. Colorectal carcinoma in children and adolescents – a report of seven cases. Eur J Pediatr Surg 2001; 11: 338 – 41. 68. Radhakrishnan CN, Bruce J. Colorectal cancers in children without any predisposing factors. A report of eight cases and review of the literature. Eur J Pediatr Surg 2003; 13: 66 – 8. 69. Mayo CW. Surgery of the colon. Selection of patients and operations for malignant lesions. Postgrad Med 1949; 5: 394 – 8. 70. Andersson A, Bergdahl L. Carcinoma of the colon in children: a report of six new cases and a review of the literature. J Pediatr Surg 1976; 11: 967 – 71. 71. Symonds DA, Vickery AL Jr. Mucinous carcinoma of the colon and rectum. Cancer 1976; 37: 1891 – 900. 72. Lanza G, et al. Prognostic significance of DNA ploidy in patients with stage II and stage III colon carcinoma. A prospective flow cytometric study. Cancer 1998; 82: 49 – 59. 73. Vasen HF, et al. The International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer (ICG-HNPCC). Dis Colon Rectum 1991; 34: 424 – 5. 74. Aaltonen LA, et al. Incidence of hereditary nonpolyposis colorectal cancer and the feasibility of molecular screening for the disease. N Engl J Med 1998; 338: 1481 – 7. 75. Thibodeau SN, et al. Altered expression of hMSH2 and hMLH1 in tumors with microsatellite instability and genetic alterations in mismatch repair genes. Cancer Res 1996; 56: 4836 – 40. 76. Honchel R, Halling KC, Thibodeau SN. Genomic instability in neoplasia. Cell Biol 1995; 6: 45 – 52. 77. Peltom¨aki P, et al. Microsatellite instability is associated with tumors that characterize the hereditary non-polyposis colorectal carcinoma syndrome. Cancer Res 1993; 53: 5853 – 5. 78. Lothe R, et al. Genomic instability in colorectal cancer: relationship to clinicopathological variables and family history. Cancer Res 1993; 53: 5849 – 52. 79. Foglia RP. Colorectal tumors. In O’Neill JA, et al. (eds) Pediatric Surgery, 5th ed. St Louis, Missouri: CV Mosby, 1998: 1461 – 1464. 80. O’Connell MJ, et al. Prospectively randomized trial of postoperative adjuvant chemotherapy in patients with high-risk colon cancer. J Clin Oncol 1998; 16: 295 – 300. 81. Moertel CG. Surgical adjuvant treatment of colorectal cancer. FORUM Trends Exp Clin Med 1992; 2: 426 – 37. 82. Porschen R, et al. Fluorouracil plus leucovorin as effective adjuvant chemotherapy in curatively resected stage III colon cancer : Results of the trial adjCCA-01. J Clin Oncol 2001; 19: 1787 – 94. 83. International Multicentre Pooled Analysis of Colon Cancer Trials (IMPACT) Investigators. Efficacy of adjuvant fluorouracil and folinic acid in colon cancer. Lancet 1995; 345: 939 – 44.
84. Pitot HC, et al. Phase II trial of irinotecan in patients with metastatic colorectal carcinoma. J Clin Oncol 1997; 15: 2910 – 9. 85. Saltz LB, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Engl J Med 2000; 343: 905 – 14. 86. Advanced Colorectal Cancer Meta-Analysis Project. Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: evidence in terms of response rate. J Clin Oncol 1992; 10: 896 – 903. 87. Andr`e T, et al. Oxaliplatin, fluorouracil and leucovorin as adjuvant treatment for colon cancer. N Engl J Med 2004; 350: 2343 – 51. 88. Jamison RL, et al. Hepatic resection for metastatic colorectal cancer results in cure for some patients. Arch Surg 1997; 132: 505 – 11. 89. Chatura KR, et al. Case report: gastric carcinoma as a complication of dyskeratosis congenita in an adolescent boy. Dig Dis Sci 1996; 41: 2340 – 2. 90. Katz S, et al. Pernicious anemia and adenocarcinoma of the stomach in an adolescent: clinical presentation and histopathology. J Pediatr Surg 1997; 32: 1384 – 5. 91. Munck A, et al. Carcinoma of the stomach in a child. J Pediatr Gastroenterol Nutr 1993; 16: 334 – 6. 92. Norman JT, Wagner ML, Chintagumpala M. Umbilical metastasis (Sister Mary Joseph’s nodule) in a child. Pediatr Radiol 1998; 28: 56 – 8. 93. Nottingham J. Signet-ring carcinoma of stomach in a child. Histopathology 1994; 24: 490 – 1. 94. Carney JA, et al. The triad of gastric leiomyosarcoma, functioning extra-adrenal paraganglioma and pulmonary chondroma. N Engl J Med 1977; 296: 1517 – 8. 95. Cochat P, et al. Diffuse leiomyomatosis in Alport syndrome. J Pediatr 1988; 113: 339 – 43. 96. van Hoeven KH, et al. Visceral myogenic tumors. A manifestation of HIV infection in children. Am J Surg Pathol 1993; 17: 1176 – 81. 97. Suster S. Gastrointestinal stromal tumors. Semin Diagn Pathol 1996; 13: 297 – 313. 98. Tworek JA, et al. Stromal tumors of the jejunum and ileum. Mod Pathol 1997; 10: 200 – 9. 99. DeMatteo RP, et al. Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 2000; 231: 51 – 8. 100. Shenoy MU, et al. Gastrointestinal stromal tumor: a rare cause of neonatal intestinal obstruction. Med Pediatr Oncol 2000; 34: 70 – 1. 101. Durham MM, et al. Gastrointestinal stromal tumors arising from the stomach: a report of three children. J Pediatr Surg 2004; 39: 1495 – 9. 102. Hughes JA, et al. Gastrointestinal stromal tumor of the duodenum in a 7-year-old boy. Pediatr Radiol 2004; 34: 1024 – 7. 103. Terada R, et al. Clinical and histopathological features of colonic stromal tumor in a child. J Gastroenterol 2000; 35: 456 – 9. 104. Geramizadeh B, et al. Neonatal gastrointestinal stromal tumor. Report of a case and review of literature. J Pediatr Surg 2005; 40: 572 – 4. 105. Cypriano MS, et al. Pediatric gastrointestinal stromal tumors and leiomyosarcoma. Cancer 2004; 101: 39 – 49. 106. Prakash S, et al. Gastrointestinal stromal tumors in children and young adults. J Pediatr Hematol Oncol 2005; 27: 179 – 87. 107. Erlandson RA, Klimstra DS, Woodruff JM. Subclassification of gastrointestinal stromal tumors based on evaluation by electron microscopy and immunohistochemistry. Ultrastruct Pathol 1996; 20: 373 – 93. 108. Carrillo R, et al. Prognostic significance of DNA ploidy and proliferative index (MIB-1 index) in gastrointestinal stromal tumors. Hum Pathol 1997; 28: 160 – 5. 109. Rubin BP, et al. KIT activation is a ubiquitous feature of gastrointestinal stromal tumor. Cancer Res 2001; 61: 8118 – 21. 110. Price VE, et al. Clinical and molecular characteristics of pediatric gastrointestinal stromal tumors (GISTs). Pediatr Blood Cancer 2005; 45: 20 – 4. 111. Corless CL, Fletcher JA, Heinrich MC. Biology of gastrointestinal stromal tumors. J Clin Oncol 2004; 22: 3813 – 25. 112. DeMetri GD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med 2002; 347: 472 – 80. 113. Wu PC, et al. Surgical treatment of gastrointestinal stromal tumors in the imatinib (STI-571) era. Surgery 2003; 134: 656 – 6.
UNCOMMON TUMORS OF THE GASTROINTESTINAL TRACT IN CHILDREN 114. Lauwers GY, et al. Gastrointestinal autonomic nerve tumors. A clinicopathological, immunohistochemical, and ultrastructural study of 12 cases. Am J Surg Pathol 1993; 17: 887 – 97. 115. Kodet R, Snajdauf J, Smelhaus V. Gastrointestinal autonomic nerve tumor: a case report with electron microscopic and immunohistochemical analysis and review of the literature. Pediatr Pathol 1994; 14: 1005 – 16. 116. Matsumoto K, et al. Gastrointestinal autonomic nerve tumors: immunohistochemical and ultrastructural studies in cases of gastrointestinal stromal tumor. Pathol Int 1997; 47: 308 – 14. 117. Ojanguren I, Ariza A, Navas-Palacios JJ. Gastrointestinal autonomic nerve tumor: further observations regarding an ultrastructural and immunohistochemical analysis of six cases. Hum Pathol 1996; 27: 1311 – 18. 118. Coffin CM, Humphrey PA, Dehner LP. Extrapulmonary inflammatory myofibroblastic tumor: a clinical and pathologic survey. Semin Diagn Pathol 1998; 15: 85 – 101. 119. Parham DM, Bugg MF, Pratt CB. Carcinomas, adenomas, precursor lesions, and second malignancies. In Parham DM (ed) Pediatric Neoplasia: Morphology and Biology. Philadelphia, Pennsylvania: Lippincott-Raven, 1996: 370 – 372. 120. Kopelman HR. Tumors of the pancreas. In Walker WA, et al. (eds) Pediatric Gastrointestinal Disease. Pathophysiology. Diagnosis. Management, 2nd ed. St Louis, Missouri: CV Mosby, 1996: 1494 – 1501. 121. Grosfeld JL, et al. Pancreatic tumors in childhood: analysis of 13 cases. J Pediatr Surg 1990; 25: 1057 – 62. 122. Jaksic T, et al. A 20-year review of pediatric pancreatic tumors. J Pediatr Surg 1992; 27: 1315 – 17. 123. Lack EE, et al. Tumors of the exocrine pancreas in children and adolescents. A clinical and pathology study of eight cases. Am J Surg Pathol 1983; 7: 319 – 27. 124. Solcia E, Capella C, Kloppel G. Tumors of the Pancreas, Third Series. Washington, District of Columbia: Armed Forces Institute of Pathology, 1997: 103 – 114. 125. Ivy EJ, Sarr MG, Reiman HM. Nonendocrine cancer of the pancreas in patients under age forty years. Surgery 1990; 108: 481 – 7. 126. Kissane JM. Tumors of the exocrine pancreas in childhood. In Humphrey GB, et al. (eds) Pancreatic Tumors in Children. The Hague, Netherlands: Martinus Nijhoff, 1982: 99 – 129. 127. Tsukimoto I, Tsuchida M. Pancreatic carcinoma in children in Japan – review of the Japanese literature. In Humphrey GB, et al. (eds) Pancreatic Tumors in Children. The Hague, Netherlands: Martinus Nijhoff, 1982: 149 – 157. 128. Osborne BM, et al. Acinar cell carcinoma of the pancreas in a 9-yearold child: case report with electron microscopic observations. South Med J 1977; 70: 370 – 2. 129. Mah P.-T, Loo DC, Tock EPC. Pancreatic acinar cell carcinoma in childhood. Am J Dis Child 1974; 128: 101 – 4. 130. Taxy JB. Adenocarcinoma of the pancreas in childhood. Report of a case and a review of the English language literature. Cancer 1976; 37: 1508 – 28. 131. Horie A. Pancreatoblastoma. Histopathologic criteria based upon a review of six cases. In Humphrey GB, et al. (eds) Pancreatic Tumors in Children. The Hague, Netherlands: Martinus Nijhoff, 1982: 159 – 166. 132. Iseki M, et al. Alpha-fetoprotein-producing pancreatoblastoma. A case report. Cancer 1986; 57: 1833 – 5. 133. Chun Y, et al. Pancreatoblastoma. J Pediatr Surg 1997; 32: 1612 – 5. 134. Capelle J. Pancr´eatoblastome. Aspects e´ chographiques et tomodensitom´etriques. A propos d’un cas clinique. J Radiol 1986; 67: 345 – 7. 135. Willnow U, et al. Pancreatoblastoma in children. Case report and review of the literature. Eur J Surg 1996; 6: 369 – 72. 136. Buchino JJ. Fine-needle aspiration of solid and papillary cystic tumor of the pancreas. Pediatr Pathol Lab Med 1996; 16: 235 – 42. 137. Reed DN Jr, Turcotte JG. Papillary epithelial neoplasm of the pancreas in the pediatric population. J Surg Oncol 1986; 32: 182 – 3.
759
138. Horisawa M, et al. Frantz’s tumor (solid and cystic tumor of the pancreas) with liver metastasis: successful treatment and long-term follow-up. J Pediatr Surg 1995; 30: 724 – 6. 139. Yagi M, et al. A solid and cystic tumor of the pancreas in a 10-yearold girl: report of a case and review of the literature. Surg Today (Jpn J Surg) 1994; 24: 826 – 8. 140. Fabbro MA, et al. Neoplatico cistico-papillare del pancreas: descrizione di un caso in et`a pediatrica. Pediatr Med Chir 1996; 18: 607 – 10. 141. W¨unsch LP, et al. Diagnosis and treatment of papillary cystic tumor of the pancreas in children. Eur J Pediatr Surg 1997; 7: 45 – 7. 142. Welch KJ. The pancreas. In Welch KJ, et al. (eds) Pediatric Surgery. Chicago, Illinois: Year Book Medical Publishers, 1986: 1097. 143. Klimstra DS, et al. Acinar cell carcinoma of the pancreas. A clinicopathologic study of 28 cases. Am J Surg Pathol 1992; 16: 815 – 37. 144. Kissane JM. Pancreatoblastoma and solid and cystic papillary tumor: two tumors related to pancreatic ontogeny. Semin Diagn Pathol 1994; 11: 152 – 64. 145. Klimstra DS, et al. Pancreatoblastoma. A clinicopathologic study and review of the literature. Am J Surg Pathol 1995; 19: 1371 – 89. 146. Hua C, Shu XK, Lei C. Pancreatoblastoma: a histochemical and immunohistochemical analysis. J Clin Pathol 1996; 49: 952 – 4. 147. Murakami T, et al. Pancreatoblastoma: case report and review of treatment in the literature. Med Pediatr Oncol 1996; 27: 193 – 7. 148. Drut R, Jones MC. Congenital pancreatoblastoma in Beckwith – Wiedemann syndrome: an emerging association. Pediatr Pathol 1988; 8: 331 – 9. 149. Wiley J, et al. Cytogenetic and flow cytometric analysis of a pancreatoblastoma. Cancer Genet Cytogenet 1995; 79: 115 – 8. 150. Pelosi G, et al. Solid and cystic papillary neoplasms of the pancreas: a clinicocytopathologic and immunocytochemical study of five new cases diagnosed by fine-needle aspiration cytology and a review of the literature. Diagn Cytopathol 1995; 13: 233 – 46. 151. Poustchi-Amin M, et al. Papillary-cystic neoplasm of the pancreas. Pediatr Radiol 1995; 25: 509 – 11. 152. Ishikawa O, et al. Solid and papillary neoplasm arising from an ectopic pancreas in the mesocolon. Am J Gastroenterol 1990; 85: 597 – 601. 153. Sclafani LM, et al. The malignant nature of papillary and cystic neoplasm of the pancreas. Cancer 1991; 68: 153 – 8. 154. Moertel CG, et al. Early evaluation of combined fluorouracil and leucovorin as a radiation enhancer for locally unresectable, residual, or recurrent gastrointestinal carcinoma. J Clin Oncol 1994; 12: 21 – 7. 155. Parlowsky T, et al. Malignant endocrine tumours in childhood and adolescence – results of a retrospective analysis. Klin Padiatr 1996; 208: 205 – 9. 156. Friesen SR. Tumors of the endocrine pancreas. N Engl J Med 1982; 306: 580 – 90. 157. Lobe TE, et al. Hepaticopancreaticogastroduodenectomy with transplantation for metastatic islet cell carcinoma in childhood. J Pediatr Surg 1992; 27: 227 – 9. 158. Moertel CG, et al. Streptozocin – doxorubicin, streptozocin – fluorouracil, or chloro-zotocin in the treatment of advanced islet-cell carcinoma. N Engl J Med 1992; 326: 519 – 23. 159. Wilson SD. Zollinger – Ellison syndrome in children: a 25-year followup. Surgery 1991; 110: 696 – 703. 160. Aszodi A, et al. Giant nonfunctioning islet cell tumor requiring pancreatoduodenectomy and complete liver revascularization. J Surg Oncol 1993; 53: 273 – 6. 161. Kvols LK, et al. Treatment of metastatic islet cell carcinoma with a somatostatin analogue (SMS 201 – 995). Ann Intern Med 1987; 107: 162 – 8. 162. Kraenzlin ME, et al. Long-term treatment of a VIPoma with somatostatin analogue resulting in remission of symptoms and possible shrinkage of metastases. Gastroenterology 1985; 88: 185 – 7. 163. Danner DB, et al. Primitive neuroectodermal tumor arising in the pancreas. Mod Pathol 1994; 7: 200 – 4.
Section 11 : Pediatric Malignancies
69
Uncommon Pediatric Genitourinary Tumors Barbara Bambach
INTRODUCTION Pediatric genitourinary tumors account for approximately 10% of pediatric malignancies, constituting 760 cases per year in the United States.1 The most common of these are Wilms’ tumor and rhabdomyosarcoma. Tumors in this location are a heterogeneous group, with many different histologies. For the purposes of this chapter, they are divided by the site of origin, that is, the urinary system (kidney, bladder, ureter, and urethra), male reproductive system (penis, prostate, and testes), and the female reproductive system (clitoris, perineum, cervix, vagina, uterus, ovary, and fallopian tube).
THE URINARY SYSTEM Kidney Renal Cell Carcinoma
The majority of these cases are sporadic, with <0.2% of all pediatric tumors and 1–3% of primary renal tumors in children being renal cell carcinoma.2,3 There are cases reported in association with tuberous sclerosis or von Hippel–Lindau disease and a handful of cases occur secondarily, mostly after therapy for neuroblastoma.4 It is uncertain whether these secondary renal cell carcinomas occur as a consequence of the previous therapy or are due to a genetic predisposition for malignancy. The clinical and pathologic features in children are similar to those of the adult form. There have been approximately 400 published cases in children, with the youngest reported as 3 months of age.5 The median age is 15.5 years. The distribution is equal between males and females, unlike in adults. Right-sided and left-sided tumors occur with equal frequency, with 1% occurring bilaterally.5 This tumor occurs in Caucasian children three times more often than in African-American children.5 Presenting symptoms include hematuria, palpable mass, and pain, and 25% will have renal calcifications. The hematuria and renal calcifications seen with renal cell carcinomas are in contrast to
Wilms’ tumor, in which these findings are unusual. In addition, hepatic dysfunction and coagulation abnormalities have been described.6 Histologically, they may be of the clear cell (most common in adults) (see Figure 1), tubopapillary, or anaplastic variant.7 There are some reports to suggest that the pediatric tumors are more likely to be high grade and papillary with some unique subtypes identified by immunohistochemical and cytogenetic features8 that are not seen in adults. A significant proportion of pediatric renal cell carcinomas are associated with Xp11.2 translocations and fusions involving the TFE3 transcription factor gene.9 Renal cell carcinoma will extend beyond the kidney at diagnosis approximately 60% of the time. Staging is identical to that of adults, following Robson’s classification (see Table 1). Metastasis is usually to the lungs, liver, bone, and, less commonly, the pleura. Survival is stage dependent, with close to 90% survival for stage I, 52–58% for stage II or III, but only 7% for stage IV.5 Age at presentation also appears to be a prognostic feature, with one series having 6 of 6 patients less than 11 years of age surviving, but only 4 of 14 patients older than 11 years surviving.10 Primary therapy is complete resection by radical nephrectomy. Radiation therapy, chemotherapy, or adoptive immunotherapy can be considered if a curative operation is not possible. Chemotherapy with cyclophosphamide, carmustine (CCNU), procarbazine, 5-fluorouracil (5-FU), and combinations have been tried, as well as progestin and testosterone.10 However, because of the rarity of this disease and the heterogeneity of the therapies tried, no comment can be made on the effectiveness of these regimens. The biologic response modifiers interferon α-2A and interleukin-2 have been shown in adults to have significant clinical activity,11 but their role in pediatric renal cell carcinoma remains unclear. Clear Cell Sarcoma
This tumor has also been called “bone metastasizing renal tumor of childhood,” and is now considered a distinct, separate entity from Wilms’ tumor. It accounts for 5% of
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
UNCOMMON PEDIATRIC GENITOURINARY TUMORS
761
rather than ifosfamide because of the potential renal toxicity of ifosfamide. Patients with pulmonary metastases should also receive whole lung irradiation. Follow-up should include chest X-ray, brain MRI, and bone scan for at least 3 years. Rhabdoid Tumor of the Kidney (RTK)
0.05 mm
Figure 1 Renal cell carcinoma with clear cell pattern. Irregular nests of cells with clear cytoplasm separated by delicate fibrovascular septae.
Table 1 Robson’s classification of renal cell carcinoma.
Stage I Stage II Stage III Stage IV
Confined to the kidney Infiltrating perirenal fat, not beyond Gerota’s fascia Involving renal vein or lymph nodes Distant metastases or involving adjacent structures
malignant renal neoplasms in children.12 The age distribution is similar to that of Wilms’ tumor. The prognosis is less favorable. These tumors cannot be distinguished from Wilms’ tumor by imaging studies, and typically present as a solid mass with areas of necrosis. Approximately one-half will have a cystic component.13 Generally, there is no vascular involvement, and calcifications may be seen. As the name implies, these tumors metastasize to the bones and also to the brain, although spread may not be present at diagnosis. Histopathologically, polygonal or stellate cells with clear cytoplasm and characteristic nuclei are seen. The staging system used is the same as for Wilms’ tumor. Imaging studies of the abdomen, chest X-ray, magnetic resonance imaging (MRI) of the brain, and bone scan are recommended for the workup. High relapse rates occur, even with lowstage disease. Aggressive therapy, however, with the addition of doxorubicin, has improved survival.14,15 The National Wilm’s Tumor Study Group recently reported improved relapse-free survival for patients treated with long (9 additional months) versus standard (6 months) chemotherapy with vincristine, dactinomycin, and doxorubicin. Overall survival rates, however, were similar (87.5% for long arm and 85.5% for standard arm at 8 years).16 In addition, the combination of ifosfamide and etoposide has been found to be highly active.15,17 The current therapy by the National Wilms’ Tumor Study is radical resection of the primary lesion, followed by abdominal irradiation to 1080 cGy (regardless of the stage) and chemotherapy with doxorubicin, etoposide, and cyclophosphamide. Cyclophosphamide is being tested
Two percent of malignant renal neoplasms in children are rhabdoid tumors.13 This was originally thought to be a rhabdomyosarcomatoid pattern of Wilms’ tumor, but there is no evidence of myoblastic differentiation and it is now classified as a separate entity. The median age of diagnosis is 13 months, with a range of 2 months to 5 years.2 Boys are five times more likely to be affected than girls.18 These tumors may metastasize to the lung, abdomen, liver, brain, bone, or lymph nodes. There is also an association with primary brain tumors of neuroectodermal origin, such as medulloblastoma, ependymoma, glioma, and primitive neuroectodermal tumor (PNET). Unlike Wilms’ tumor, there is no association with Beckwith-Wiedemann syndrome, aniridia, or hemihypertrophy. However, there have been cases reported in association with hypercalcemia.13 Rhabdoid tumor cannot be distinguished from Wilms’ tumor by imaging, although it may be suggested by a central location within the kidney, the presence of subcapsular fluid, and metastatic lesions at diagnosis. These tumors usually originate from the renal sinus as opposed to the cortical origin of Wilms’ tumor. Pathologically, a single cell line is present, in contrast to the triphasic pattern seen with Wilms’ tumor. There are usually abundant periodic acid Schiff (PAS) positive hyaline globular cytoplasmic inclusions. Ultrastructurally, large cytoplasmic whorls of intermediate filaments displacing nuclei are seen.18 Several microscopic subtypes have been reported, including a sclerosing variant with the potential for primitive osteochondral metaplasia, an epithelioid variant, a spindle cell variant, a paraganglioma-like form, a variant assuming a lymphoma-like configuration, and a variant with numerous cysts.18 Translocations involving chromosomes 11 and 22 are reported, but involve different segments than those seen in Ewing’s sarcoma and PNET.18 Prominent expression of the c-myc oncogene has also been reported.19 The molecular hallmark of rhabdoid tumor of the kidney (RTK) was discovered in 1998 by the Delattre group.9 A biallelic inactivation of the hSNF5/INI1 tumor suppressor gene on the long arm of chromosome 22 occurs via mutation, deletion, or whole chromosome loss.9 These tumors have a very poor prognosis, with a mortality rate of 80%.13 Aggressive therapy with surgical resection, radiation therapy, and experimental dose-intensive or new chemotherapy regimens is warranted because of the poor response to vincristine, dactinomycin, and doxorubicin.20 The current National Wilms’ Tumor Study is investigating a modified dosing of doxorubicin, etoposide, and carboplatin, which has been used in the treatment of other primitive neurogenic tumors in children. Congenital Mesoblastic Nephroma
This is the most common renal neoplasm of infancy, most often being detected by 6 months of age. It is usually benign. However, there have been a few reports of metastasis to the
762
PEDIATRIC MALIGNANCIES
lungs and subsequent death.21,22 Polyhydramnios is often a clue to the initial diagnosis. These lesions are usually unifocal and unilateral, and on ultrasound appear predominantly solid. The computed tomography (CT) appearance shows distortion of the renal parenchyma, but the tumor does not enhance with contrast and does not invade the renal vessels or collecting system. Local recurrences are rare. The treatment is nephrectomy. If there is no residual disease and the histology is not aggressive, no follow-up imaging is needed.13 However, those with atypical cellularity, increased mitoses, or aneuploidy may have aggressive behavior and are more likely to metastasize.23,24 A cellular variant of CMN is microscopically identical to infantile fibrosarcoma (IFS) and affects the same young age-group. Recent studies have shown cellular CMN to be molecularly distinct from the classic CMN.9 In fact, cellular CMN has the same chromosome translocation, t(12;15)(p13;q25), and gene fusion (ETV6 gene on chromosome 12 with NTRK3 gene on chromosome 15) as IFS.9 Thus, cellular CMN is now considered an IFS of the renal sinus.9 Consequently, chemotherapy used for IFS has been used for recurrent or unresectable cellular CMN with some success. This has included chemotherapy with vincristine/doxorubicin/dactinomycin, vincristine/doxorubicin/cyclophosphamide, or ifosfamide/ carboplatin/etoposide.25 Metanephric Stromal Tumor (MST)
A newly recognized entity, metanephric stromal tumor (MST), consists of what historically had been considered as CMN in children more than 3 years of age.9 It is a biphasic tumor with an epithelial component identical to metanephric adenoma. Typical presentation is an abdominal mass or hematuria. Rarely, there may be hypertension or hemorrhage. Grossly, it is a tan, lobulated fibrous mass, frequently in the renal medulla. It is unencapsulated but has a subtly infiltrative scalloped border.9 Most of these tumors result in angiodysplasia of entrapped arterioles. These tumors are benign, with no reports of metastasis or local recurrence.9 Excision is the only necessary therapy. Some patients have had morbidity or mortality from extrarenal angiodysplasia, such as aneurysms of the renal arteries or aortic rupture.9 Ossifying Renal Tumor of Infancy
Only a few cases of this tumor have been reported. It is speculated that it is a variant of CMN. Some believe it arises in the renal pelvis, and others believe it is related to an embryonic rest.26 It typically follows a benign course, with only local invasion reported. There have been no metastases or known recurrences.26 It usually presents as gross hematuria in infants less than 6 months of age, and is seen radiographically as a calcified renal mass with distortion of the intrarenal collecting system. Microscopically, there is a predominant osteoid component with plump, polygonal cells and eosinophilic cytoplasmic granules that merge with a spindle cell component. Most patients have undergone nephrectomy. Only two patients have been reported who were treated with chemotherapy consisting of dactinomycin
and vincristine.27 There is 100% survival in those patients in whom survival information is known.27 Primary Renal Lymphoma
The existence of this entity is controversial, particularly in the pediatric population. Many feel that renal lymphomas are not primary in the kidney, but rather represent dissemination.28 Extrarenal disease must be excluded by radiographic studies, bone marrow aspiration, and lumbar puncture. Some authors have argued that a staging laparotomy is necessary.29 Radiographically, they have a classic appearance of an enlarged kidney with a nodular pattern, so that percutaneous renal biopsy can be performed to confirm the diagnosis, rather than nephrectomy. Pediatric patients have had both lymphoblastic T cell and B cell lymphomas reported in the literature, and there is one reported case of an undifferentiated lymphoma.27 Angiocentric T cell lymphoma has been reported only in adult patients with acquired immunodeficiency syndrome (AIDS). However, with the increasing prevalence of AIDS in the pediatric population, this must also be considered. Renal lymphoma may present with pain, a palpable mass, and/or hypertension. However, the renal failure seen in adults does not appear to be as common in the pediatric population. Most patients have died of disease progression, so that systemic therapy rather than surgical resection is indicated. The chemotherapy should be based on the histologic type as well as the extent of disease, as with nonrenal lymphomas. Nephrectomy should be reserved for those cases with persistent macroscopic disease refractory to chemotherapy and radiation. Sarcomas
Malignant mesenchymal tumors of the kidney are extremely rare, comprising only 2.8% of renal tumors in adults.2 There have been no patient series specific to pediatric-aged patients. Leiomyosarcomas (LMS) and angioleiomyosarcomas are the most common variants. Initial presentation may be with a catastrophic hemorrhage (Wunderlich syndrome).2 Treatment is radical nephrectomy. Renin-Secreting Juxtaglomerular Cell Tumor
Twenty-one cases have been described in children and adolescents, with only three children younger than the teenage years.30 These patients have presented with headaches, hypertension, hyperreninemia, and secondary hyperaldosteronism with hypokalemia.30 These tumors have a benign course with no metastatic disease reported. Local excision is recommended (see Figure 2). Oncocytic Carcinoid
Only 12 cases with this diagnosis have been reported in the literature, with one case occurring in a 13-year-old girl.31 This presented as periodic Cushing’s syndrome, and pathologically resembled a renal oncocytoma. A renal lesion seen initially by renal angiography had been thought to represent a renal cyst. The patient continued to have a relapsing course for 3.5 years before a CT scan suggested the lesion to be solid. The renal mass was excised and the patient remained asymptomatic at follow-up of 2.5 years.31
UNCOMMON PEDIATRIC GENITOURINARY TUMORS
763
and prominent stromal desmoplasia with inflammation are seen.34 Radiographically, the tumor is central, infiltrating the renal sinus and shows peripheral nodules with heterogeneous enhancement.34 The disease is usually advanced at diagnosis, and the prognosis is poor. There is little response to chemotherapy or radiation. Therapy, as with most renal tumors, has been with radical nephrectomy. Mean survival reported is 15 weeks.34 Metastatic Osteosarcoma
0.05 mm
This has only rarely been detected premortem (seven cases in the literature), but has been noted in 11% of autopsy cases.35 Most patients are asymptomatic, although two patients presented with flank pain and hematuria. Renal Pelvis and Ureteral Tumors
Figure 2 Juxtaglomerular tumor of kidney. The cells have uniform round to polygonal appearance with granular cytoplasm.
Renal Hemangiopericytoma (HPC)
Fifteen cases of renal and perirenal hemangiopericytoma (HPC) are reported in the literature, with one reported case in a 2-year-old.32 In general, HPCs are benign, although HPCs located deep in tissue are more likely to be malignant. It is seen as a complex mass on sonogram, and is hypervascular on angiogram. It enhances strongly with welldemarcated margins on CT scan.32 Nephrectomy is the mode of treatment. Nephrogenic Adenofibroma
Five patients with this entity are reported in the literature. It was first described by Hennigar and Beckwith in 1992.33 The mean age is 13.3 years, ranging from 3.5 to 23 years, thus occurring later than Wilms’ tumor or mesoblastic nephroma. It has been postulated to arise from intralobar rests.33 Two patients had tumors that contained low-grade papillary adenocarcinoma resembling collecting duct carcinoma.33 In these patients, there was an association with polycythemia, which resolved after nephrectomy.33 No metastasis or recurrence has been seen 1–2 years after nephrectomy. Although this entity is felt to be benign, it is recommended that these patients be followed closely, particularly with the presence of ductal carcinoma.33 Renal Medullary Carcinoma
Renal medullary carcinoma is a newly described, highly aggressive malignant tumor that occurs in adolescent and young adult black patients with sickle cell (SC) trait or hemoglobin SC disease.34 Interestingly, it has not been described in patients homozygous for SC disease (hemoglobin SS). It has been called the seventh SC nephropathy.34 Mean age of presentation is 20 years with a range of 10–39 years. Presentation is with gross hematuria, abdominal or flank pain. Less commonly, weight loss or fever may occur.34 This tumor arises at the renal pelvis–mucosal interface, quickly fills the renal pelvis, and invades vascular and lymphatic structures. There are often intraparenchymal satellite nodules. Histologically, SCs, hemorrhage, necrosis,
Eight pediatric cases have been documented in the literature.36 Five patients had transitional cell carcinoma, two had papillary adenocarcinoma, and one had squamous cell carcinoma. Symptoms are related to calculi and pyelonephritis, with cystoscopy showing a hydronephrotic kidney and a filling defect. Urine cytology is usually negative. These tumors can metastasize to the liver, lungs, peritoneum, and lymph nodes. Most patients die within 1 year of diagnosis.36 The only known therapy is surgical resection, there being no available data on chemotherapy or radiotherapy. It is thought that the use of bowel segment in urinary reconstruction and diversion may increase the risk of malignancy because of exposure to urine or feces or both.37 Because of the young age at which many pediatric patients undergo this reconstructive surgery, there may be a significant risk with long-term exposure.
Bladder Transitional cell carcinoma in the pediatric population was initially thought to be low grade, noninvasive, and nonrecurrent. While this is generally the case, recurrences and deaths have now been reported. Approximately 100 known pediatric cases are reported in the literature, with fewer than 30 of them in children under 10 years of age.38,39 The most common presenting symptom is gross, painless hematuria, although bladder stones, recurrent urinary tract infections, or irritative voiding symptoms have occurred. There is a male preponderance. Urinary cytology is not often useful. Frequently, delays in diagnosis are due to failure to consider this in the differential diagnosis of hematuria in a pediatric patient. A urogram or cystogram may show a filling defect. Cystoscopy should be performed. These tumors are graded and staged as in adults, with most pediatric patients presenting with low-stage and low-grade tumors.38 Only 3% were found to be invasive through the lamina propia. Traditional therapy has been transurethral resection, although in one series, 25% had an open excision with or without fulguration.38 The recurrence rate has ranged from 2 to 5%.38 There is one reported case of intravesical therapy with thiotepa in a pediatric patient with recurrent multifocal disease.40 These patients need to have follow-up with urinalysis for hematuria and urine cytology on a regular basis, particularly if a high-grade tumor is
764
PEDIATRIC MALIGNANCIES
found. Routine cystoscopy is probably not warranted.38 One small series of five patients found ultrasound to be sensitive for the detection of intravesical tumors, making this potentially useful for diagnosis prior to cystoscopy, as well as for follow-up.38
THE MALE REPRODUCTIVE SYSTEM Penis Carcinoma
Carcinoma of the penis is a tumor that typically occurs in the sixth and seventh decades of life, although there have been five cases reported in children less than 15 years of age.41,42 Poor socioeconomic status and lack of personal hygiene are thought to contribute, and neonatal circumcision has been proven to nearly eliminate its occurrence.43 – 45 Patients present with ulcerative lesions on the glans or shaft of the penis and may have nodular or ulcerated lymph nodes. Surgical therapy for local disease is the only known effective therapy. Rhabdomyosarcoma
While rhabdomyosarcoma is one of the more frequent neoplasms in children, its origin in the penis is uncommon. Four pediatric patients were reported in the literature, with six cases in total.46 – 51 In one of these pediatric patients, the origin was actually in the urethra. Of the three remaining patients, two patients had dorsal peno-pubic masses, and one had a ventral penile mass associated with priapism. Therapy includes surgery and/or radiation as well as traditional chemotherapy for rhabdomyosarcoma, and is dependent on the stage of the disease, as with rhabdomyosarcomas originating elsewhere. Primary Penile Lymphoma
A total of 10 cases have been reported in the literature, of which two occurred in children (ages 4 and 18 years).52 – 59 Histology for the pediatric patients was a diffuse large cell immunoblastic lymphoma in one, and a small cell lymphosarcoma in the other. Systemic therapy should be based on the histology and stage of the disease.
Prostate Primary Carcinoid of the Prostate
This entity has been reported in a 7-year-old in conjunction with multiple endocrine neoplasia IIb.60 The patient had a several-year history of burning micturition and hematuria, and had an asymmetrically enlarged prostate on rectal examination. His metastatic workup was negative, and a nerve-sparing radical retropubic prostatectomy was performed. Follow-up beyond 3 months is unknown. In adults, these tumors are generally low grade and have an excellent prognosis if the surgical margins are clear.61 – 63 Adenocarcinoma of the Prostate
This is a very rare tumor in men less than 35 years, with 14 reports in children less than 18 years, the youngest being
20 months old.64 – 67 Children usually present with urinary retention and hematuria and tend to have a large tumor burden at diagnosis. The pediatric patient invariably has a poorly differentiated, aggressive tumor with negative tumor markers of prostatic acid phosphatase (PAP) and prostatespecific antigen (PSA).65,66 These tumors metastasize to regional and distant lymph nodes and lungs. Bony metastases usually develop late, and are osteolytic, unlike those in adults.65,67 The majority of pediatric patients described have had advanced disease at diagnosis not amenable to radical surgery, and have responded poorly to radiation and hormonal therapy. Chemotherapy may be of some benefit (unlike in adults), including vincristine, doxorubicin, and ifosfamide.64 However, most patients die within 1 year.
Testes One percent of childhood malignancies are testicular tumors, which occur in 1 in 1 000 000 Caucasian Americans.68 The majority are endodermal sinus tumors and yolk sac tumors. The common adult histologies, including seminomas, are unusual in the pediatric population. All solid tumors of the testes should be approached through an inguinal excision with a high inguinal orchiectomy for suspected malignancy. Testicular tumors can include germ cell tumors, gonadal stromal tumors, gonadoblastoma, tumors of the supporting tissues, lymphomas and leukemias, metastatic tumors, and tumors of the adnexa (see Table 2). A staging system used for pediatric testicular tumors has been designed by the Committee on Tumors, Section of Urology, Academy of Pediatrics (see Table 3).69 Gonadoblastoma
This rare tumor is seen in patients with gonadal dysgenesis and a Y-chromosome cell line. The majority of patients Table 2 Classification of prepubertal testicular tumors.
Germ cell tumors Yolk sac tumor Teratoma Seminoma Mixed germ cell tumor Gonadal stromal tumors Leydig cell Sertoli cell Granulosa cell Mixed gonadal stromal tumor Gonadoblastoma Tumors of supporting tissues Fibroma, fibrosarcoma, leiomyoma, leiomyosarcoma, hemangioma Lymphomas and leukemia Tumor like lesions Epidermoid cyst Hyperplastic nodule secondary to congenital adrenal hyperplasia Secondary tumors Tumors of the adnexa Rhabdomyosarcoma, fibroma, fibrosarcoma, leiomyoma, leiomyosarcoma, hemangioma, lipoma Source: Adapted from Coppes MJ, Rackley R, Kay R. Primary testicular and paratesticular tumors in childhood. Med Pediatr Oncol 1994; 22: 329 – 40. AFP = Alpha feto-protein
UNCOMMON PEDIATRIC GENITOURINARY TUMORS Table 3 Grouping of testicular tumors in children.
Group I
Group II
Group III
Group IV
Tumor confined to the testis, AFP levels normal within 1 month after orchiectomy; imaging of retroperitoneum and chest normal Similar to group I but retroperitoneal lymph node dissection reveals unsuspected nodal metastases Retroperitoneal lymph node metastases demonstrated on imaging studies, serum AFP levels persistently elevated Demonstrable metastases beyond the retroperitoneum
Source: Adapted from Coppes MJ, Rackley R, Kay R. Primary testicular and paratesticular tumors in childhood. Med Pediatr Oncol 1994; 22: 329 – 40.
are phenotypic girls with various degrees of virilization. The remainder are phenotypic boys with hypospadias and cryptorchidism. These tumors are usually benign, but do have significant potential for malignant transformation, often to dysgerminoma, and should thus have prophylactic gonadectomy. Scully reported the largest review,70 in which 57% had 46XY, 30% had 45X/46XY, one patient had 45X, and the remainder had other mosaicisms. One-third of these tumors are bilateral at diagnosis.71 Histologically, three distinct elements are seen: large germ cells similar to seminoma, sex cord nongerminal elements (Sertoli or granulosa) and mesenchymal or stromal elements (Leydig cells). Calcifications may be seen. One-half will have an overgrowth of the germ cell component, and 10% of these will metastasize.70 This entity has been described in the newborn period in a patient with dysgenetic testes,72 emphasizing the importance of early diagnosis and prophylactic gonadectomy. Seminoma
While this is the most common testicular tumor in adults, it is rarely seen in prepubertal boys.73 Recommended therapy is the same as for adults: orchiectomy for stage I, orchiectomy plus chemotherapy or radiation for stage II, and preoperative chemotherapy followed by surgery, radiation, and additional chemotherapy for stage III.74,75 Chemotherapy is with cisplatin and bleomycin. In situ seminoma in the testes has been seen in young males who are infertile as well as in dysgenetic gonads.76 Leydig Cell Tumor
While 10% of these tumors in adults have a malignant course, they are benign in children. The treatment is inguinal orchiectomy.71 Sertoli Cell Tumor
Less than 1% of testicular tumors are Sertoli cell tumors, and 15% of these occur in children.71 These usually present as a painless testicular mass. The most common associated endocrine disorder is gynecomastia. There has also been some association with Peutz-Jeghers syndrome.77,78 The majority of these are benign. However, malignant variants have been reported in children, one of which
765
metastasized.74,79 Treatment is inguinal orchiectomy, but retroperitoneal extension must be excluded. Retroperitoneal lymphadenectomy is indicated for sertoli cell tumors with malignant features (e.g. metastasis, retroperitoneal extension).80 Large Cell Calcifying Sertoli Cell Tumor
These tumors are often bilateral, and occur more frequently with endocrine disturbances such as gynecomastia secondary to estrogen elevation.81 They have been reported in association with cardiac myxoma, as well as endocrine disorders, including adrenocortical hyperplasia, pituitary adenoma, and sexual precocity.81 Histologically, large Sertoli cells with abundant eosinophilic cytoplasm and diffuse trabecular growth patterns with calcification are seen.81 Treatment is similar to that of other Sertoli cell tumors. Metastatic Disease to the Testes
Neuroblastoma and Wilms’ tumor metastases to the testes have been described.82,83 These patients usually have widely disseminated disease. Up to 4% of patients with neuroblastoma have been found to have testicular metastases.84 These patients usually present with an adrenal primary and metastatic disease to the retroperitoneal nodes, bone and bone marrow. The neuroblastoma metastases to the testes are usually discovered at autopsy.82 Wilms’ tumor metastases have been reported in two cases.83 One patient had involvement of all the veins throughout the testis and spermatic cord, suggesting retrograde venous extension, without other systemic metastasis. He was treated with orchiectomy, radiation, and adjuvant chemotherapy. The second patient had multiple metastases after orchiectomy and died within a year. Leukemia
Historically, 20% of boys with acute lymphocytic leukemia were found to have testicular disease grossly at diagnosis,85 although this is now estimated to be 5% because of improved early diagnosis. If routine testicular biopsies are performed, the incidence of testicular leukemia may be as high as 35%.85 Patients usually have painless testicular swelling, which may be unilateral or bilateral. Patients at highest risk for testicular disease at diagnosis as well as testicular relapse are those with T cell lymphoblastic leukemia and higher lymphoblast counts at diagnosis. Since isolated testicular relapses can lead to systemic relapse, therapy involves irradiation of both testes (18–24 Gy) plus subsequent chemotherapy.86 The importance of occult testicular leukemia is undetermined, and routine testicular biopsy is not justified. Non-Hodgkin’s Lymphoma
Overt infiltration of the testes with non-Hodgkin’s lymphoma (NHL) has been described either at diagnosis or relapse.87,88 Most reported series are of adult patients, with relatively few pediatric patients described. Kellie et al., in a review of the St. Jude experience,89 found 5% of all boys with NHL, and 7% of those with advanced stage disease had testicular involvement. Six of nine patients had disease at diagnosis and presented with painless testicular enlargement. In two,
766
PEDIATRIC MALIGNANCIES
the disease was bilateral. Four patients had diffuse undifferentiated non-Burkitt’s lymphoma and one patient had lymphoblastic lymphoma. All six patients responded to induction therapy with a clinical complete response; four of these patients remained disease-free 2.9–8.3 years after receiving high-dose fractionated cyclophosphamide and intermediate dose methotrexate. Two of these patients received radiation to the testes (2250–2400 cGy), and one patient had an orchiectomy. Two patients have relapsed. Three of nine patients had their testicular disease occur as isolated unilateral relapse, and all achieved a second complete response after reinduction; one of these is a survivor. Four patients were cured by chemotherapy alone without testicular radiation. Primary NHL of the testes was not seen. Angiocentric Lymphoma
Angiocentric immunoproliferative lesions (AIL) are a spectrum of benign to malignant disorders in which extranodal sites have a polymorphic cellular infiltrate centered around vessels, with a destructive and infiltrative pattern with necrosis.90 There is one reported case in the literature of a 6.5-year old diagnosed with T cell acute lymphoblastic leukemia (ALL) at age 5 years, in remission for 17 months, who developed a rapidly growing, painless testicular mass, pan-hypogammaglobulinemia, absence of circulating B cells, and an elevation of circulating activated T cells.89 This child had a grade III AIL histologically. He remained on his chemotherapy for ALL and received 3000 cGy external beam involved-field radiation to the testicle. It is hypothesized that his intensive chemotherapy led to immunosuppression, which allowed for the predisposition to AIL. This entity has been reported in children with ALL previously. However, it usually occurs in the lungs.89 Melanotic Neuroectodermal Tumor of Infancy in the Epididymis (MNTI)
This rare pigmented neoplasm of infants is usually seen in the maxilla in children less than 1 year of age. Twenty-three cases have been reported in the epididymis,91 three of which were malignant on the basis of lymph node involvement or on pathologic features.92 – 94 Most patients were less than 1 year of age. There is evidence of neural crest origin by biochemical, ultrastructural, and immunohistochemical studies.91 – 94 These tumors are usually benign, although local recurrence has occurred in up to 60%.95 Approximately 7% developed metastases, and postoperative radiation therapy with adjuvant chemotherapy has been given to patients whose tumors demonstrate malignant features,95 although no specific regimen is prescribed. Juvenile Granulosa Cell Tumor of the Testis
There are 24 cases reviewed in the literature;96 14 in the newborn period, 9 between the ages of 1 and 6 months, and 1 at 21 months. These usually present as enlargement of the scrotal testis without endocrine abnormalities. Five patients had ambiguous genitalia or abnormal karyotype. All 12 patients for whom follow-up is known were well 6 months to 7 years after unilateral orchiectomy.
Malignant Mesothelioma of the Tunica Vaginalis
A few pediatric cases of this tumor have been reported, with the youngest described patient being 5 years old.97 – 99 These tumors are usually discovered incidentally during a hydrocele repair. They may be locally invasive and/or metastasize to inguinal and retroperitoneal nodes as well as to the chest and abdomen. Grossly, they appear as studding of the hydrocele sac with papillary excrescences. Microscopically, they may be either epithelioid or biphasic, and the epithelial component may have a mixture of papillary, tubopapillary, and tubular patterns. The treatment has been primarily surgical with or without adjuvant radiation therapy. The favored surgical treatment is radical orchiectomy with hemiscrotectomy if the overlying scrotal skin is involved. Metastatic disease has been treated with doxorubicin-based chemotherapy combined with lymph node dissection or radiation therapy.99 The potential for long-term survival is dependent on an initial complete resection. Late recurrences have been reported, with local recurrence usually occurring before metastases develop. The median disease-free survival for recurrence is 20 months.98 Testicular ‘Tumor’ of the Adrenogenital Syndrome(TTAGS)
These benign testicular masses in men with congenital adrenal hyperplasia (CAH) were first described in 1901.100 They are dependent on an elevation of adrenocorticotropic hormone (ACTH) levels, and decrease in size with administration of exogenous corticosteroids to lower the ACTH level. When CAH is not identified beforehand, an orchiectomy is usually performed. Pathologically, the lesion may be misinterpreted as a Leydig cell tumor.101 Approximately one-third of cases are found in children.102 Symptoms are attributable to the adrenogenital syndrome and include vomiting, diarrhea, sudden collapse, virilization, and sexual precocity. Approximately 20% of the patients will not have the diagnosis of CAH made until after development of the testicular mass. The majority (65%) of patients have the salt-wasting variety of CAH, and 83% have bilateral masses.102 Surgery is not indicated, and therapy should be aimed at treating the underlying CAH. Epidermoid Cyst
Although these occur most frequently in adults, pediatric cases have been reported.103 – 105 They usually present as painless nodules or cysts, and clinically mimic germ cell tumors. Thus, they are usually not diagnosed prior to surgery. The absence of testicular intraepithelial neoplasia distinguishes them from teratoma. Typically, these tumors have a benign course, and testis-sparing excisions have been performed with no reports of local recurrences.103
THE FEMALE REPRODUCTIVE TRACT Clitoris and Vulva Hemangiopericytoma
HPC is a tumor of the pericytes, or rudimentary contractile cells adjacent to the capillary wall. This tumor may have
UNCOMMON PEDIATRIC GENITOURINARY TUMORS
a benign or malignant course. There are eight pediatric cases arising from the clitoris reported in the literature.106,107 Despite increased cellularity and mitoses consistent with malignant features, pediatric patients typically follow a benign course, although there are reports in the literature of local recurrences and metastases to the chest and brain in children with congenital lesions.108 Radiotherapy has been ineffective,107,109 but chemotherapy with dactinomycin, cyclophosphamide, and methotrexate has been successful in two patients.110 Clear Cell Eccrine Hidradenocarcinoma
Clear cell eccrine hidradenocarcinoma is a malignant tumor, which arises from eccrine sweat glands, most often in the head, trunk, and limbs. There are two reported cases of this tumor arising from the vulva in children.110,111 Both patients were adolescents and had clitoral lesions with inguinal metastases. They underwent radical vulvectomy with lymphadenectomy. Neither patient received adjuvant therapy, and both were reported alive with no evidence of disease at 11 months and 6 years. Metastases from these tumors arising from other sites have been to the lung, bone, and viscera, and usually occur within months, although late recurrences are reported.111 Most descriptions of these tumors are in the pathology literature, with little clinical information. However, systemic therapy for distant metastatic disease has been tried with doxorubicin, cyclophosphamide, vincristine, 5-FU, dactinomycin, and bleomycin with no response and no survivors.112,113 Aggressive Angiomyxoma
Aggressive angiomyxoma is a rare but well-described tumor with a predilection for female pelvic soft parts. Only 24 cases have been reported,114 with one being an adolescent (15 years old). These tumors are locally infiltrative with a tendency for multiple recurrences.114 They tend to grow slowly and may occupy the whole pelvic region. Microscopically, they are predominantly myxoid with small mesenchymal stellate and spindle shaped cells and prominent vascularization. The differential diagnosis is that of other benign or malignant mesenchymal tumors, but these other tumors do not typically have the prominent vascular component. Recurrences may be late, but metastatic disease is not reported. Most recurrences are related to inadequate excision. Wide surgical resection is the best treatment. Superficial Perineal Leiomyosarcoma
LMS are malignant tumors of smooth muscle origin, being divided according to tissue of origin (large vessel, bone, visceral, and somatic soft tissue), and classified as either superficial (dermal and subcutaneous) or deep seated (entire tumor is subfascial). Only a handful of cases can be found in the literature, the youngest patient being a 15-yearold female.115 – 119 The prognosis of this tumor is poor, with three of the four patients in one report dead at 17–99 months from diagnosis.115 Thus, they appear to have a less favorable outcome than those that arise in other sites. It has been postulated that the large amount of fat tissue in the ischiorectal fossa may allow for tumors to grow and spread
767
significantly prior to producing symptoms.115 The primary treatment is wide surgical excision, with resection attempted for also local recurrence and metastases.115 Other
There is one reported case of alveolar rhabdomyosarcoma arising from the clitoris in a pediatric patient.120 This patient had presented with a clitoral mass following trauma. She underwent a radical clitorectomy and was treated with doxorubicin and 4000 cGy of radiation to the tumor bed. She had no evidence of disease at 3 years. There is a single case report of malignant schwannoma in a 1-yearold with Von Recklinghausen’s neurofibromatosis, treated with radical clitorectomy, with no recurrence.121 A 2-yearold with an endodermal sinus tumor arising in the clitoris is also reported.122 She underwent wide resection and node sampling and was also reported free of disease.
Cervix Mullerian Adenosarcoma ¨
M¨ullerian adenosarcoma is a rare tumor composed of benign epithelial elements and malignant mesenchymal tissue. While usually arising in the uterine corpus of older women, five cases originating in the cervix are described, the youngest of which is a 14-year-old girl.123 This tumor has typically followed a benign course, with surgery being the only therapy in the majority of cases (local excision in two, total abdominal hysterectomy plus bilateral salpingooopherectomy in one). However, one adolescent girl was initially treated with chemotherapy consisting of vincristine, doxorubicin, cyclophosphamide, and dactinomycin with very little response. She eventually underwent tumor resection and radiation therapy, but 5 months later suffered a local recurrence with nodal spread. Additional chemotherapy with intraarterial cisplatin was given with partial response, but she ultimately developed diffuse metastatic disease and died.123 Alveolar Soft Part Sarcoma
This unusual tumor most often occurs in the head or neck of children. However, an 8-year-old child was reported in a series of eight patients with cervical origin.124 The patient received postoperative chemotherapy but no further clinical information was reported. Sarcoma Botryoides
Sarcoma botryoides is most typically seen in the vagina of infants. However, 19 cases arising from the cervix in females less than 18 years old have been reported.125 These patients have been treated with a variety of operative procedures, ranging from local excision or cervicectomy to radical operations with or without adjuvant chemotherapy or radiation. There is no available evidence from the literature to suggest that chemotherapy or radiation therapy has improved the prognosis of cervical tumors that can be successfully
768
PEDIATRIC MALIGNANCIES
0.05 mm
Figure 3 Alveolar soft part sarcoma. Nesting arrangement of tumor with detached central cells simulating an alveolar pattern.
125
removed surgically. However, additional therapy is justified for large, deeply infiltrating tumors.
Vagina Alveolar Soft Part Sarcoma
Less than 1% of all soft tissue sarcomas are alveolar soft part sarcomas (see Figure 3), usually located in the head and neck in children.126 However, a handful of these tumors have been reported arising from the vagina.127 – 130 Most patients presented with menometrorrhagia. Pathologically, large tumor cells in organoid arrangements with thin-walled sinusoidal spaces are seen. They characteristically have PASpositive diastase-resistant crystals. Recent advances have detected a translocation, t(X;17)(p11;q25), in many alveolar soft part sarcoma tumors.131,132 Ladanyi et al. has shown these results in a fusion of the TFE3 transcription factor gene on Xp11 to a normal gene on 17q25.133 Strong nuclear immunoreactivity for TFE3 can be seen.134 Four of five patients received radiation therapy ranging from 4500 to 8200 cGy, and one patient received chemotherapy with cisplatin and cyclophosphamide.128 Three patients are alive and free of disease 3–17 years from diagnosis. The status of the other two patients is unknown. Usual metastatic sites are lungs, brain, and bone. Extrapolating from what is known about these tumors at other sites, they are relatively slow growing, with many patients dying long after their initial diagnosis, emphasizing the need for prolonged follow-up.135 Because of their rarity, there is no consensus for therapy. Shen et al. have recommended wide local excision with lymph node dissection.136 The role of radiotherapy is still debated, particularly in young women and girls in whom the long-term complications of radiotherapy must be taken into consideration.126,129 Endodermal Sinus Tumor
Although this germ cell tumor usually arises in the ovary, there have been several reports of vaginal endodermal sinus tumors, primarily in children less than 2 years old.137 – 141 The
typical presentation is vaginal bleeding. On physical examination, polypoid tumors extending into the vaginal lumen are found. Prior to 1965, local therapy only with surgery and/or radiation was utilized, with most patients dying within 6 months. Vawter introduced the use of chemotherapy for this tumor in two patients with recurrent, metastatic disease,142 and chemotherapy is an integral part of the therapy today.140,141,143,144 Initial regimens included vincristine, dactinomycin, and cyclophosphamide. On the basis of the success of regimens utilized for ovarian primaries, current therapy consists of cisplatin, vinblastine, and bleomycin. Only primary chemotherapy has been used, with good results.140,141 Chemotherapy should be used as up-front therapy prior to exenterative surgery or radiation therapy.
Uterus Alveolar Soft Part Sarcoma
Eight patients have been reported in whom this tumor arose from the uterus, including the lower uterine segment or the uterine corpus (myometrium and endometrium).145 One of these patients was a 14-year-old. All patients had abnormal vaginal bleeding, and all underwent hysterectomy with or without bilateral salpingo-oopherectomy. No adjuvant therapy was given. No recurrences or metastatic disease were seen, with an average follow-up of 46 months (9–78 months). They seem to have a better prognosis than their counterparts arising outside of the female genital tract. However, this may be due to the relatively short follow-up period reported, as approximately one-third of reported recurrences have been 10 years or more after diagnosis.145 Mullerian Adenosarcoma ¨
Although this tumor typically occurs in the endometrial cavity of postmenopausal women, there have been several case reports in adolescent females.123,146 – 152 Of nine such cases, six had primary lesions in the lower uterine/cervical section. Five were initially treated with local resection only, and three had a recurrence. The four patients with primary lesions in the uterine body cavity underwent abdominal radical hysterectomy. One of these patients died; the remainder were disease-free at follow-up of 14 months to 15 years. The aggressiveness of the tumor is felt to be related to the depth of myometrial extension on histologic examination.148,153
Ovary and Fallopian Tube The most common type of ovarian malignancy in children is the germ cell tumor, with pure endodermal sinus tumor occurring most often, followed by mixed germ cell tumor.2 Ovarian tumors should be explored through a midline incision, allowing for tumor delivery and walling off of the operative field without risk of tumor rupture, which could complicate therapy. Once the mass has been externalized, it should be evaluated. If smooth, with a simple unilocular cyst, it can usually be enucleated and the ovary preserved. If it is felt to be malignant, a unilateral salpingo-oopherectomy should be performed, removing the infundibular-pelvic ligament, ovarian nerve, artery and vein, and any adjoining
UNCOMMON PEDIATRIC GENITOURINARY TUMORS Table 4 Ovarian staging, The Federation Internationale de Gynecologie et d’Obstetrique (FIGO).
Stage I Stage IA
Stage IB
Stage IC
Stage II Stage IIA Stage IIB Stage IIC Stage III
Stage IIIA Stage IIIB Stage IIIC
Stage IV
Tumor limited to the ovaries (one or both). Tumor limited to one ovary; capsule intact, no tumor on ovarian surface. No malignant cells in ascites or peritoneal washings. Tumor limited to both ovaries; capsules intact, no tumor on ovarian surface. No malignant cells in ascites or peritoneal washings. Tumor limited to one or both ovaries with any of the following: capsule rupture, tumor on ovarian surface, malignant cells in ascites or peritoneal washings. Tumor involves one or both ovaries with pelvic extension. Extension and/or implants on uterus and/or tube(s). No malignant cells in ascites or peritoneal washings. Extension to other pelvic tissues. No malignant cells in ascites or peritoneal washings. Pelvic extension (2a or 2b) with malignant cells in ascites or peritoneal washings. Tumor involves one or both ovaries with microscopically confirmed peritoneal metastasis outside the pelvis and/or regional lymph node metastasis. Microscopic peritoneal metastasis beyond the pelvis. Macroscopic peritoneal metastasis beyond pelvis 2 cm or less in greatest dimension. Peritoneal metastasis beyond pelvis more than 2 cm in greatest dimension and/or regional lymph node metastasis. Distant metastasis (excludes peritoneal metastasis).
769
very malignant tumors with only 30% survival in stage I patients,155 and most patients are dead within 6 months.154 Treatment is primarily surgical, with unilateral salpingooopherectomy. Only those patients without extraovarian spread have survived.154 The role of adjuvant therapy is unknown, although there are recent reports of successful therapy with an aggressive multiagent regimen.156,157 The combination of vinblastine, cisplatin, cyclophosphamide, bleomycin, doxorubicin, and etoposide has been described in detail158 and used with success in this tumor.159 Metastatic Tumors to the Ovary
There is very little literature on this phenomenon in children. The largest report comes from the Children’s Hospital Medical Center, Boston, in which 14 cases are described.154 The majority of these were diagnosed at autopsy, although five were symptomatic premortem, and were initially misdiagnosed as primary ovarian tumors. The most common tumor metastatic to the ovary is neuroblastoma, with autopsy specimens frequently showing an ovary diffusely replaced by tumor.160 This must be differentiated from primary neuroblastoma of the ovary, of which there are five reported cases (presumably arising from teratoma).160 The next most common tumor metastatic to the ovary is rhabdomyosarcoma. Of 11 reported cases, six were alveolar, three were embryonal, one mixed and one unknown. Other tumors that have been reported to spread to the ovary include Ewing’s sarcoma,161,162 retinoblastoma,161 and medulloblastoma.162 Desmoplastic small round cell tumor has been described in three patients, and Wilms’ tumor in one patient.160 There are very rare reports of colorectal carcinoma and gastric carcinoma metastatic to the ovary in adolescents.160 This is seen more frequently in adults. Bilaterality favors the suspicion of metastatic disease, rather than primary ovarian tumor. Borderline Mucinous Tumor
0.1 mm
Figure 4 Small cell carcinoma of the ovary.
lymph nodes along with the specimen.3 Staging of ovarian tumors is based on surgical findings (see Table 4), according to the International Federation of Gynaecology and Obstetrics (FIGO) classification. Small Cell Carcinoma with Hypercalcemia
The average age of presentation for this tumor (see Figure 4) is 23 years, with a reported range of 9–44 years.154 Symptoms are nonspecific, such as abdominal pain and swelling. These tumors are almost always unilateral. Hypercalcemia is reported in two-thirds of the patients.155 These are
Epithelial tumors of the ovary are rare in premenarchal women. The majority of these are mucinous tumors. There are a few reports of borderline mucinous tumors in children.163 These tend to have very low malignant potential, lacking stromal invasion. Ten-year survival is reported at 73% for all stages, with 100% survival for stage I. Seventyfive percent of patients present with stage I tumors.163 These patients should be treated conservatively to preserve hormonal and reproductive function. The tumor marker CA-125 was reportedly useful in one case. Serous Epithelial Carcinoma
There are fewer than 30 patients under 20 years of age reported, 2 of whom had advanced disease (stage III).164 Fourteen of these tumors were invasive, eight were borderline and seven unknown. The youngest child reported was aged 4 years.164 The younger patients tend to present with early stage disease. The tumor marker CA-125 may be useful in these patients. Treatment has consisted of debulking and platinum-based chemotherapy.164 Consideration should be given to preserving the uterus for reproductive capacity. The borderline serous tumors have prolonged survival of over 10 years, even at advanced stages.165
770
PEDIATRIC MALIGNANCIES
Acute Lymphocytic Leukemia
Primary Ovarian Sarcomas
Ovarian involvement usually presents as relapsed disease rather than being found at initial diagnosis. Up to 50% of autopsy specimens in females with ALL report disease in the ovaries.166 In a series from Emory University,166 nine patients diagnosed antemortem had bilateral disease, and 13 patients had unilateral disease, although the contralateral ovary was not always biopsied. In this series, approximately 20% had positive cerebrospinal fluid (CSF) cytology, and 20% had bone marrow relapse. Most of the cases have occurred late, often greater than 36 months after the initial diagnosis. Abdominal pain is the most frequent presenting symptom, and the majority of evaluable cases had palpable abdominal masses. Although the numbers are small, there appears to be no obvious advantage for salpingooopherectomy versus simple biopsy followed by adjuvant systemic therapy. One-third of the patients were reported in continuous complete remission with a median follow-up of 42 months. Earlier ovarian relapses had a higher failure rate than late relapses, similar to ALL relapse at other sites. Radiation therapy did not appear to increase survival. With the use of more intensive regimens including intermediate dose methotrexate, ovarian relapse may become less frequent.
Less than 3% of all ovarian tumors are sarcomas.171 Many different histologies have been reported, including rhabdomyosarcoma, stromal cell sarcoma, fibrosarcoma, and leiomyosarcoma.
Granulocytic Sarcoma or Chloroma
This localized tumor mass of myeloid leukemia cells has been reported rarely in the ovary. Drinkard et al. reported five cases in the literature.167 The French-American-British (FAB) subtype of these patients was not described, although Drinkard et al. described two patients with FAB M4 with eosinophilia and inversion of chromosome 16.167 Therapy is treatment of the underlying leukemia. Malignant Lymphomas
Lymphomatous involvement of the ovaries is frequently found at autopsy, but only rarely presents with ovarian enlargement antemortem. It may be primary or secondary. Although the existence of primary lymphoma of the ovary is controversial, there have been reported cases of lymphoma localized to the ovary, which have been treated successfully with surgical excision alone.168,169 It has been described in children as young as 2 years of age. Patients have presented with abdominal or back pain, swelling of the lower extremity, ‘B’ symptoms, irregular vaginal bleeding, ascites, and pleural effusions. In a large retrospective study, the primary tumors were distinguished by their detection inadvertently at the time of a surgical procedure for another disease.170 In this study, the tumors were all unilateral, involving the left ovary. The majority (54%) were small noncleaved Burkitt’s, with the others consisting of diffuse large cell (23%), follicular and diffuse large cell (8%), diffuse mixed small and large cell (8%), large cell immunoblastic (5%), and follicular and diffuse small cleaved cell (2%). Patients with bilateral disease had clinical symptoms and pathologic findings more consistent with secondary involvement. In this study, three out of four patients with primary ovarian lymphoma were free of disease 1–8 years later with only surgical excision.171
Rhabdomyosarcoma There have been very rare case reports of ovarian rhabdomyosarcoma in young people less than 20 years of age.172 They typically present with abdominal swelling and discomfort. The majority are pleomorphic (in children greater than 5 years of age) or embryonal (in children less than 5 years of age). The cases reported have been treated surgically, including hysterectomy and bilateral salpingo-oopherectomy. Adjuvant therapy with chemotherapy or radiation has been used only if resection was not possible. Survival was poor, with most patients dead within 12 months. On the basis of the experience of treating rhabdomyosarcoma at other sites, adjuvant treatment with chemotherapy, with or without radiation, may improve survival. Stromatosis (low-grade stromal sarcoma) Although this entity primarily occurs in the fifth and sixth decades of life, there are two reports of pediatric-aged patients.172 Symptoms are nonspecific. There are several histologic variants of this disorder, including a hyaline vascular variant and a cellular variant. These may be confused with HPCs because of the vascular elements seen.172 These often infiltrate into surrounding tissue without being clinically apparent. The treatment of choice is hysterectomy and bilateral salpingooopherectomy. Survival is excellent with appropriate surgical therapy, even when there has been spread beyond the ovary.172 Ovarian stromal sarcoma (endometrial type) This rare tumor usually occurs in women >35 years, but there is one reported case in a 14-year-old.172 Histologically, these tumors have abundant mitoses and lack differentiation, and rhabdomyoblasts are seen. Surgical removal is the therapy of choice, with little known about chemotherapy. The 5-year survival rate is 20–25%.172 Dissemination is both lymphatic and hematogenous, with frequent development of pulmonary metastases. Fibrosarcoma There is a single pediatric case report of this entity in an 8-year-old with nevoid basal cell carcinoma syndrome.172 These tumors are hypercellular with ‘herringbone’ and ‘storiform’ patterns. Treatment is surgical, with hysterectomy and bilateral salpingo-oopherectomy. Prognosis is poor; 13 of 15 reported patients have died.172 Leiomyosarcoma Prior to the report of a 12-year-old with this tumor,173 this entity was thought to arise in postmenopausal women only. Symptoms are nonspecific. The 12-year-old patient had a previous history of medulloblastoma treated with cranio-spinal radiation at 11 months of age. The total dose of gonadal exposure was estimated to be 200–1000 cGy. Surgery is the mainstay of treatment. The role of chemotherapy is undetermined. This pediatric patient was alive without recurrence 20 months later.
UNCOMMON PEDIATRIC GENITOURINARY TUMORS
Juvenile Granulosa Cell Tumor
This is a rare cause of gonadotropin-releasing hormone (GnRH)-independent sexual precocity in girls. The juvenile form accounts for 5% of all granulosa cell tumors, and comprises 90% of all prepubertal sex cord tumors.174 Historically, the mortality rate was as high as 25%, with most deaths occurring within 3 years of diagnosis.175 However, early detection and surgical excision has improved survival. Histologically, large luteinized cytoplasm with clear pleomorphic nuclei is seen, but the characteristic ‘grooving’ found in the adult variant is not found.175 Stage is the most important prognostic factor. Stage I tumors have <2% mortality, while stages II and III are almost uniformly fatal.176 Unilateral oophorectomy for stage I is sufficient because of the low rate of bilateral disease and favorable prognosis. The results of chemotherapy in aggressive, advanced disease have been poor.176 Inhibin and M¨ullerian inhibiting substance may be useful tumor markers.177,178 There have been reports of this tumor in association with other mesenchymally derived tumors such as osteochondroma alone,179,180 or with hemangioma (Maffucci syndrome).181,182 Squamous Cell Carcinoma
There is one pediatric patient reported in the literature with ovarian squamous cell carcinoma.183 The patient had pulmonary and bone marrow metastases, suggesting hematogenous spread, which is uncommon in ovarian squamous cell carcinomas seen in adults. There appear to be two major prognostic factors in adults: metastatic involvement and rupture of cysts before or during removal. Malignant Mixed Mullerian Tumor of Fallopian Tubes ¨
This tumor is more typically found in the endometrium, vagina, cervix, or ovary (see Figure 5). However, there have been 39 cases reported in the fallopian tubes, ranging in age from 14 to 79 years.184 The vast majority are in the postmenopausal age-group. Pain, atypical vaginal bleeding, and abdominal distension have been the presenting symptoms.
0.1 mm
Figure 5 Malignant mixed M¨ullerian tumor. Admixture of carcinomatous and poorly differentiated sarcomatous components. Foci of hyalinized areas resemble osteoid (right upper corner).
771
The one pediatric patient described had metastatic disease to the bone, lung, and peritoneum, and died of disease. The majority of adult patients have presented with stage I disease. Treatment is surgical, with the majority undergoing hysterectomy and bilateral salpingo-oopherectomy. The mean time to death is 16 months.185 Histologically, this is a carcinosarcomatous lesion with the carcinomatous component resembling primary adenocarcinoma. Stage of the tumor is the most important prognostic factor, with 9 of 13 surviving patients having stage I disease. Chemotherapy trials in adults have included vincristine, dactinomycin, doxorubicin, dacarbazine, cyclophosphamide, and cisplatin, without much success.185 Radiation therapy has been utilized in 13 patients, 8 of whom died.185
ACKNOWLEDGMENT The photographs were provided courtesy of the Department of Pathology, Roswell Park Cancer Institute, Buffalo, New York.
REFERENCES 1. Smyth TB, Dairiki Shortliffe LM. Pediatric genitourinary tumors. Curr Opin Oncol 1990; 2: 507 – 13. 2. La Quaglia MP. Genitourinary tract cancer in childhood. Semin Pediatr Surg 1996; 5(1): 49 – 65. 3. Kissane JM, Dehner LP. Renal tumors and tumor-like lesions in pediatric patients. Pediatr Nephrol 1992; 6: 365 – 82. 4. Donnelly LF, Rencken IO, Shardell K. Renal cell carcinoma after therapy for neuroblastoma. Am J Roentgenol 1996; 167: 915 – 7. 5. Fiori E, et al. Renal cell carcinoma in adolescents. A case report and review of the literature. Panminerva Med 1996; 38: 121 – 8. 6. Ward JS, Meddleton RG. Renal-cell carcinoma in children. Urology 1973; 2: 50 – 2. 7. Krigman HR, Bentley RC, Washington K. Anaplastic renal cell carcinoma following neuroblastoma. Med Pediatr Oncol 1995; 25: 52 – 9. 8. Renshaw AA, et al. Renal cell carcinomas in children and young adults: Increased incidence of papillary architecture and unique subtypes. Am J Surg Pathol 1999; 23(7): 795 – 802. 9.. Argani P, Ladanyi M. Recent advances in pediatric renal neoplasia. Adv Anat Pathol 2003; 10(5): 243 – 60. 10. Raney RB Jr, et al. Renal cell carcinoma in children. Med Pediatr Oncol 1983; 11: 91 – 8. 11. Figlin RA, Belldegrun A, Moldawer N. Concomitant administration of recombinant human interleukin-2 and recombinant interferon alpha2A: an active outpatient regimen in metastatic renal cell carcinoma. J Clin Oncol 1992; 10(3): 414 – 21. 12. Weeks DA, Mierau GW, Beckwith JB. Practical electron microscopy of pediatric renal tumors. Ultrastruct Pathol 1996; 20: 31 – 3. 13. Strouse PJ. Pediatric renal neoplasms. Radiol Clin North Am 1996; 34(6): 1081 – 100. 14. Green DM, et al. The treatment of children with clear cell sarcoma of the kidney (CCSK). A report from the National Wilms Tumor Study (NWTS). Proc Annu Meet Am Assoc Cancer Res 1994; 35: 1428. 15. Green DM, Breslow NE, D’Angio GJ. Treatment of children with clear cell sarcoma of the kidney. A report from the National Wilms Tumor Study Group. J Clin Oncol 1994; 12: 2132 – 7. 16. Seibel NL, et al. Effect of duration of treatment on treatment outcome for patients with clear-cell sarcoma of the kidney: A report from the National Wilms’ Tumor Study Group. J Clin Oncol 2004; 22(3): 468 – 73. 17. Miser J, Krailo M, Hammond GD. The combination of ifosfamide (IFOS), etoposide (VP-16) and MESNA (M): a very active regimen in the treatment of recurrent Wilms’ tumor (WT) (Abstract). Proc Am Soc Clin Oncol 1993; 12: 417.
772
PEDIATRIC MALIGNANCIES
18. Wick MR, Ritter JH, Dehner LP. Malignant rhabdoid tumors: a clinicopathologic review and conceptual discussion. Semin Diagn Pathol 1995; 12(3): 233 – 48. 19. Cowan B, Dairiki Shortliffe L. Pediatric genitourinary tumors. Curr Opin Oncol 1992; 4: 455 – 62. 20. D’Angio GJ, Breslow N, Beckwith JB. The treatment of Wilms’ tumor: results of the Third National Wilms’ Tumor Study. Cancer 1989; 64: 349 – 60. 21. Schlesinger AE, et al. Congenital mesoblastic nephroma metastatic to the brain: a report of two cases. Pediatr Radiol 1995; 25: S73. 22. Steinfeld AD, et al. Recurrent and metastatic mesoblastic nephroma in infancy. J Clin Oncol 1984; 2: 956. 23. Joshi VV, Kasznica J, Walters T. Typical mesoblastic nephroma: pathological characteristics of a potentially aggressive variant of congenital mesoblastic nephroma. Arch Pathol Lab Med 1986; 110: 100 – 6. 24. Barrantes JC, et al. Congenital mesoblastic nephroma: possible prognostic and management value of assessing DNA content. J Clin Pathol 1991; 44: 317 – 20. 25. Loeb DM, Hill DA, Dome JS. Complete response of recurrent cellular congenital mesoblastic nephroma to chemotherapy. J Pediatr Hematol Oncol 2002; 24(6): 478 – 81. 26. Middlebrook PF, Jimenez CL, Schillinger JF. Ossifying renal tumor of infancy: a case report. J Urol 1992; 147: 1337 – 9. 27. Sotela-Avila C, Beckwith JB, Johnson JE. Ossifying renal tumor of infancy: a clinicopathologic study of nine cases. Pediatr Pathol Lab Med 1995; 15: 745 – 62. 28. Arranz Arija JA, Carrion JR, Menarguez FJ. Primary renal lymphoma: report of 3 cases and review of the literature. Am J Nephrol 1994; 14: 148 – 53. 29. Harris GJ, Lager DJ. Primary renal lymphoma. J Surg Oncol 1991; 46: 273 – 7. 30. Abbi RK, McVicar M, Teichberg S. Pathologic characterization of a renin-secreting juxtaglomerular cell tumor in a child and review of the pediatric literature. Pediatr Pathol 1993; 13: 443 – 51. 31. Hannah J, Lippe B, Lai-Goldman M. Oncocytic carcinoid of the kidney associated with periodic Cushing’s syndrome. Cancer 1988; 61: 2136 – 40. 32. Wan Y-L, Chen W-J, Chen W-F. Renal hemangiopericytoma in childhood: noninvasive imaging. J Ultrasound Med 1992; 11: 237 – 9. 33. Hennigar RA, Beckwith JB. Nephrogenic adenofibroma: a novel kidney tumor of young people. Am J Surg Pathol 1992; 16(4): 325 – 34. 34. Lowe LH, et al. Pediatric renal masses: Wilms tumor and beyond. Radiographics 2000; 20(6): 1585 – 603. 35. Lockhart SK, et al. Osteosarcoma metastatic to the kidney. Clin Imaging 1989; 13: 154 – 6. 36. Moncino MD, et al. Papillary adenocarcinoma of the renal pelvis in a child: case report and brief review of the literature. Med Pediatr Oncol 1990; 18: 81 – 6. 37. Fichtner J, Dairiki Shortliffe LM. Pediatric genitourinary tumors. Curr Opin Oncol 1993; 5: 530 – 7. 38. Keetch DW, Manley CB, Catalona WJ. Transitional cell carcinoma of bladder in children and adolescents. Urology 1993; 42(4): 447 – 9. 39. Hoenig DM, McRae S, Caldamore AA. Transitional cell carcinoma of the bladder in the pediatric patient. J Urol 1996; 156: 203 – 5. 40. Moog RA. Tumours of the lower urinary tract in children. In 12th Congres de la Societe Internationale d’Urologie, Rio de Janeiro, Brazil, 1968. 41. Hemal AK, Kumar R, Wadhwa SN. Carcinoma penis in a young boy. A case report. Indian J Cancer 1996; 33(2): 108 – 10. 42. Narismharao KL, Chatterjee H, Veliath AJ. Penile carcinoma in the first decade of life. Br J Urol 1985; 57: 358. 43. Schelhammer PF, Jorden GH, Schlossberg SM. Tumors of the penis. In Walsh PC, et al. (eds) Campbell’s Urology. Philadelphia, Pennsylvania: WB Saunders, 1992: 1269 – 1276. 44. Persky L. Epidemiology of cancer of the penis. Res Cancer Res 1977; 60: 97. 45. Bissada NK, Marcos RR, El-Senoussi M. Post circumcision carcinoma of the penis (clinical aspects). J Urol 1986; 135: 283. 46. Dalkin B, Zaontz MR. Rhabdomyosarcoma of the penis in children. J Urol 1989; 141: 908 – 9.
47. Maresch R, Chiari H. Harnorgare mannliche Geschlechtsorgane Das Sarkom des Glides. In Henke F, Lubarsch O (eds) Handbuch der Speziellen Pathologischen Anatomic und Histologic. Berlin, Germany: Springer-Verlag, 1931: 369. 48. Ramos JZ, Pack GT. Primary embryonal rhabdomyosarcoma of the penis in a 2-year-old child. J Urol 1966; 96: 928. 49. Puhl H. Zur Kenntnis der Sarkome der Harnrohre (Mitteilungeines Falles von Myosarkom). Z Urol 1929; 23: 583. 50. Castellanos-Yodo U. Rhabdomyosarcoma primario del pene (relate de un cas). Arch Cuba Cancerol 1955; 14: 348. 51. Pak K, et al. Rhabdomyosarcoma of the penis. J Urol 1986; 136: 438. 52. Fairfax CA, Hammer CJ, Barry JM. Primary penile lymphoma presenting as a penile ulcer. J Urol 1995; 153: 1051 – 2. 53. Oomura J, et al. Primary reticulosarcoma of the penis: report of a case. Hinyokika Kiyo 1962; 8: 536. 54. Marks D, et al. Therapy of primary diffuse large cell lymphoma of the penis with preservation of function. J Urol 1988; 139: 1057. 55. Gough J. Primary reticulum cell sarcoma of the penis. Br J Urol 1970; 42: 336. 56. Dehner LP, Smith BH. Soft tissue tumors of the penis. A clinicopathologic study of 38 patients. Cancer 1970; 25: 1431. 57. Stewart AL, Grieve RJ, Banerjee SS. Primary lymphoma of the penis. Eur J Surg Oncol 1985; 11: 179. 58. Gonzalez-Campora R, et al. Lymphoma of the penis. J Urol 1981; 126: 270. 59. Yu GSM, Nseyo UO, Carson JW. Primary penile lymphoma in a patient with Peyronie’s disease. J Urol 1989; 142: 1076. 60. Whelan T, Gatfield CT, Schillinger JF. Primary carcinoid of the prostate in conjunction with multiple endocrine neoplasia IIb in a child. J Urol 1995; 153(3Pt 2): 1080 – 2. 61. Wasserstein PW, Goldman RL. Primary carcinoid of prostate. Urology 1979; 13: 318. 62. Wasserstein PW, Goldman RL. Diffuse carcinoid of prostate. Urology 1981; 18: 407. 63. Almagro UA. Argyrophilic prostatic carcinoma. A case report with literature review on prostatic carcinoid and carcinoid-like prostatic carcinoma. Cancer 1985; 55: 608. 64. Sandhu DPS, Munson KW, Benghiat A. Natural history and prognosis of prostate carcinoma in adolescents and men under 35 years of age. Br J Urol 1992; 69: 525 – 9. 65. Shimada H, et al. Carcinoma of the prostate in childhood and adolescence: report of a case and review of the literature. Cancer 1980; 46: 2534 – 42. 66. Briet S, et al. Adenocarcinoma of the prostate in adolescents and young adults. Apropos of a case in a 20 year old man. J Urol 1986; 92: 565 – 8. 67. Chiu CL, Weber DL. Prostatic carcinoma in young adults. JAMA 1974; 4: 724 – 6. 68. Connolly JA, Gearhart JP. Management of yolk sac tumors in children. Urol Clin North Am 1993; 20(1): 7 – 14. 69. Ablin A, Isaacs A Jr. Germ cell tumors. In Pizzo PA, Poplack DG (eds) Principles and Practice of Pediatric Oncology. Philadelphia, Pennsylvania: JB Lippincott, 1989: 713 – 731. 70. Scully R. Gonadoblastoma: a review of 74 cases. Cancer 1970; 25: 1340. 71. Coppes MJ, Rackley R, Kay R. Primary testicular and paratesticular tumors in childhood. Med Pediatr Oncol 1994; 22: 329 – 40. 72. Hung W, et al. Gonadoblastoma in dysgenetic testis causing male pseudohermaphroditism in newborn. Urology 1981; 17: 584. 73. Viprakasit D, et al. Seminoma in children. Urology 1977; 9: 568 – 70. 74. Kramer SA, Kelalis PP. Testicular tumors in children. In Javadpour N (ed) Principles and Management of Testicular Cancer. New York: Thieme, 1986: 362 – 377. 75. Kay R, Kaplan GW. Testicular tumors in infants and children. AUA Update Ser 1992; 11(15): 114 – 8. 76. Dehner L. Gonadal and extragonadal germ cell neoplasia of childhood. Hum Pathol 1983; 14: 493 – 511. 77. Dubois RS, et al. Feminizing sex cord tumor with annular tubules in a boy with Peutz – Jeghers syndrome. J Pediatr 1982; 101: 568 – 71. 78. Cantu JM, et al. Peutz – Jeghers syndrome with feminizing Sertoli cell tumor. Cancer 1980; 46: 223 – 8.
UNCOMMON PEDIATRIC GENITOURINARY TUMORS 79. Sharma S, Seam R, Kapoor H. Malignant Sertoli cell tumor of the testis in a child. J Surg Oncol 1990; 44: 129 – 31. 80. Henley JD, Young RH, Ulbright TM. Malignant sertoli cell tumors of the testis: A study of 13 examples of a neoplasm frequently misinterpreted as seminoma. Am J Surg Pathol 2002; 26(5): 541 – 50. 81. Proppe K, Scully R. Large-cell calcifying Sertoli cell tumor of the testis. Am J Clin Pathol 1980; 74: 607 – 19. 82. Kushner B, et al. Metastatic neuroblastoma and testicular involvement. Cancer 1985; 56: 1730. 83. Sauter E. et al. Wilms’ tumor with metastasis to the left testis. Am Surg 1990; 56: 260. 84. Cortez JC, Kaplan GW. Gonadal stromal tumors, gonadoblastomas, epidermoid cysts, and secondary tumors of the testis in children. Urol Clin North Am 1993; 20(1): 15 – 26. 85. Gutjhar P, Humpl T. Testicular lymphoblastic leukemia/lymphoma. World J Urol 1995; 13: 230 – 2. 86. Finkelstein JZ, Miller DR, Feisner J. Treatment of overt isolated testicular relapse in children on therapy for acute lymphoblastic leukemia – a report from the Children’s Cancer Group. Cancer 1994; 73: 219 – 23. 87. Jaffe N, et al. Role of staging in childhood non-Hodgkin’s lymphoma. Cancer Treat Res 1977; 61: 1001 – 7. 88. Al-Attar A, et al. Intensive chemotherapy for non-localized Burkitt’s lymphoma. Arch Dis Child 1986; 61: 1013 – 9. 89. Kellie SJ, Pui C-H, Murphy SB. Childhood non-Hodgkin’s lymphoma involving the testis: clinical features and treatment outcome. J Clin Oncol 1989; 7(8): 1066 – 70. 90. Hsueh C, Gonzalez-Crussi F, Murphy SB. Testicular angiocentric lymphoma of postthymic T-cell type in a child with T-cell acute lymphoblastic leukemia in remission. Cancer 1993; 72(5): 1801 – 5. 91. Kobayashi T, et al. Melanotic neuroectodermal tumor of infancy in the epididymis. Urol Int 1996; 57: 262 – 5. 92. Johnson RE, Scheithauer BW, Dahlin DC. Melanotic neuroectodermal tumor of infancy: a review of seven cases. Cancer 1983; 52: 661 – 6. 93. DeChiara A, Van Tornout JM, Hachitanda Y. Melanotic neuroectodermal tumor of infancy: a case report of paratesticular primary with lymph node involvement. Am J Pediatr Hematol Oncol 1992; 14: 356 – 60. 94. Jurincic-Winkler C, Metz KA, Klippel KF. Melanotic neuroectodermal tumor of infancy (MNTI) in the epididymis. A case report with immunohistological studies and special consideration of malignant features. Zentralbl Pathol 1994; 140: 181 – 5. 95. Pettinato G, et al. Melanotic neuroectodermal tumor of infancy: a reexamination of a histogenetic problem based on immunohistochemical, flow cytometric and ultrastructural study of 10 cases. Am J Surg Pathol 1991; 15: 233 – 45. 96. Groisman GM, Dische MR, Fine EM. Juvenile granulosa cell tumor of the testis: a comparative immunohistochemical study with normal infantile gonads. Pediatr Pathol 1993; 13: 389 – 400. 97. Jones MA, Young RH, Scully RE. Malignant mesothelioma of the tunica vaginalis. A clinicopathologic analysis of 11 cases with review of the literature. Am J Surg Pathol 1995; 19(7): 815 – 25. 98. Tjandra BS, Daemen MJAP, Weil EH. Papillary mesothelioma of the albuginea testis. Urology 1994; 43(1): 118 – 20. 99. Stein N, Henkes D. Mesothelioma of the testicle in a child. J Urol 1986; 135: 794. 100. Brutschy P. Hochgradige lipoidhyperplasie beider Nebenieren mit herdformigen kalkablagerungen bei einem Fall von Hypospadiasis Penisscrotalis und doppelseitigem Kryptorhismus mit unechter akzessorischer Nebenniere am rechten Hoden (Pseudohermaphroditismus masculinus externus). Frankf Z Pathol 1920; 24: 203 – 40. 101. Glenn JF, Boyce WH. Adrenogenitalism with testicular adrenal rests simulating interstitial cell tumor. J Urol 1963; 89: 456 – 63. 102. Rutgers JL, Young RH, Scully RE. The testicular ‘tumor’ of the adrenogenital syndrome. A report of six cases and review of the literature on testicular masses in patients with adrenocortical disorders. Am J Surg Pathol 1988; 12(7): 503 – 13. 103. Dieckmann KP, Loy V. Epidermoid cyst of the testis: a review of clinical and histogenetic considerations. Br J Urol 1994; 73: 436 – 41. 104. Price EB Jr. Epidermoid cysts of the testis: a clinical and pathologic analysis of 69 cases from the testicular tumor registry. J Urol 1969; 102: 708 – 13.
773
105. Malek RS, Rosen JS, Farrow GM. Epidermoid cyst of the testis: a critical analysis. Br J Urol 1986; 58: 55 – 9. 106. Brock JW III, Morgan W, Andersen T. Hemangiopericytoma of the clitoris. J Urol 1995; 153(2): 468 – 9. 107. Atkinson JB, et al. Hemangiopericytoma in infants and children. A report of six patients. Am J Surg 1984; 148: 372. 108. Morgan A, Evbuomwan I. Congenital hemangiopericytoma of the face with early distant metastasis. J Am Coll Surg 1983; 28: 123. 109. Friedman M, Egan JW. Irradiation of hemangiopericytoma of Stout. Radiology 1960; 74: 721. 110. Cohen Y, Liochtig C, Robinson E. Combination chemotherapy in the treatment of hemangiopericytoma. Oncology 1972; 26: 180. 111. Mambo NC. The significance of atypical nuclear changes in benign eccrine acrospiromas: a clinical and pathologic study of 18 cases. J Cutan Pathol 1984; 11: 35 – 44. 112. Lopez-Burbano LF, et al. Malignant clear-cell hidradenoma. Plast Reconstr Surg 1987; 80: 300 – 3. 113. Chow CW, Campbell PE, Burry AF. Sweat gland carcinomas in children. Cancer 1984; 53: 1222 – 7. 114. Simo M, Zapata C, Esquius J. Aggressive angiomyxoma of the female pelvis and perineum. Report of two cases and review of the literature. Br J Obstet Gynaecol 1992; 99: 925 – 7. 115. Grove A, Backman Nohr S. Superficial perineal leiomyosarcoma in an adolescent female and a review of the literature including vulvar leiomyosarcomas. APMIS 1992; 100: 1081 – 8. 116. Lavecchia G, et al. Cerebral metastases from perineal leiomyosarcoma. Ital J Neurol Sci 1985; 6: 351 – 4. 117. Phelan JT, Sherer W, Mesa P. Malignant smooth-muscle tumors (leiomyosarcomas) of soft tissue origin. N Engl J Med 1962; 226: 1027 – 30. 118. Stout AP, Hill WT. Leiomyosarcoma of the superficial soft tissue. Cancer 1958; 11: 844 – 54. 119. Aartsen EJ, Albus-Lutter CE. Vulvar sarcoma: clinical implications. Eur J Obstet Gynecol Reprod Biol 1994; 56: 181 – 9. 120. Bond SJ, Seibel N, Kapur S. Rhabdomyosarcoma of the clitoris. Cancer 1994; 73(7): 1984 – 6. 121. Thomas WJ, Bevan HE, Hooper DG. Malignant schwannoma of the clitoris in a 1 year old child. Cancer 1989; 63: 2216 – 9. 122. Castaldo TW, et al. Endodermal sinus tumor of the clitoris. Gynecol Oncol 1980; 9: 376 – 80. 123. Gal D, et al. Mullerian adenosarcoma of the uterine cervix. Gynecol Oncol 1988; 31: 445 – 53. 124. Moritsmu Y, Tanaka H, Iwanaga S. Alveolar soft part sarcoma of the uterine cervix. Acta Pathol Jpn 1993; 43: 204 – 8. 125. Daya DA, Scully RE. Sarcoma botryoides of the uterine cervix in young women: a clinicopathological study of 13 cases. Gynecol Oncol 1988; 29: 290 – 304. 126. Chang H-C, et al. Alveolar soft part sarcoma of the vagina. A case report. J Reprod Med 1994; 39(2): 121 – 5. 127. Kasai K, Yoshida Y, Okumara M. Alveolar soft part sarcoma of the vagina: clinical features and morphology. Gynecol Oncol 1980; 9: 227. 128. Chapman GW, Benda J, Williams T. Alveolar soft part sarcoma of the vagina. Gynecol Oncol 1984; 18: 125. 129. O’Toole RV, et al. Alveolar soft part sarcoma of the vagina: an immunochemical and electron microscopic study. Int J Gynecol Pathol 1985; 4: 258. 130. Carinelli SG, et al. Alveolar soft part sarcoma of the vagina. Tumori 1990; 76: 77. 131. Heinmann P, et al. Alveolar soft-part sarcoma: Further evidence by FISH for the involvement for chromosome band 17q25. Genes Chromosomes Cancer 1998; 23: 194 – 7. 132. Joyama S, et al. Chromosome rearrangement at 17q25 and Xp11.2 in alveolar soft-part sarcoma. Cancer 1999; 86: 1246 – 50. 133. Ladanyi M, et al. The der(17)t(X;17)(p11;q25) of human alveolar soft part sarcomafuses fuses the TFE3 transcription factor gene to ASPL, a novel gene at 17q25. Oncogene 2001; 20: 48 – 57. 134. Argani P, et al. Aberrant nuclear immunoreactivity for TFE3 in neoplasms with TFE3 gene fusions. Am J Surg Pathol 2003; 27: 750 – 61. 135. Enzinger FM, Weiss S. Soft Tissue Tumors, 2nd ed. St Louis, Missouri: CV Mosby, 1988.
774
PEDIATRIC MALIGNANCIES
136. Shen J, D’Ablaing G, Merrow CP. Alveolar soft part sarcoma of the vulva: report of first case and review of literature. Gynecol Oncol 1982; 13: 120. 137. SenGupta SK, et al. A rare case of endodermal sinus tumor of the vagina in an infant. Aust N Z J Obstet Gynecol 1991; 31(4): 381 – 2. 138. Young RH, Scully RE. Endodermal sinus tumor of the vagina. A report of nine cases and review of the literature. Gynecol Oncol 1984; 18: 380 – 2. 139. Kohorn EL, et al. Case reports: endodermal sinus tumour of the infant vagina. Gynecol Oncol 1985; 20: 196 – 203. 140. Andersen WA, et al. Endodermal sinus tumour of the vagina: the role of primary chemotherapy. Cancer 1985; 56: 1025 – 7. 141. Collins HS, et al. Endodermal sinus tumour of the infant vagina treated exclusively by chemotherapy. Obstet Gynecol 1989; 73: 507 – 9. 142. Vawter GF. Carcinoma of the vagina in infancy. Cancer 1965; 18: 1479 – 84. 143. Gonzalez-Crussi F. Atlas of Tumor Pathology, 2nd series, fascicle 18, Washington, DC: Armed Forces Institute of Pathology, 1982. 144. Norris HJ, Bagley GP, Taylor HB. Carcinoma of the infant vagina. Arch Pathol 1970; 90: 473 – 9. 145. Nielsen GP, et al. Alveolar soft-part sarcoma of the female genital tract: a report of nine cases and review of the literature. Int J Gynecol Pathol 1995; 14: 283 – 92. 146. Oda Y, Nakanishi I, Tateiwa T. Intramural mullerian adenosarcoma of the uterus with adenomyosis. Arch Pathol Lab Med 1984; 108: 798. 147. Fayemi AO, Ali M, Braun EV. Mullerian adenosarcoma of the uterine cervix. Am J Obstet Gynecol 1974; 130: 734. 148. Andrade LA, et al. Mullerian adenosarcoma of the uterus in adolescents. Int J Gynecol Obstet 1992; 38: 119 – 23. 149. Clement PB, Scully RE. Mullerian adenosarcoma of the uterus. A clinicopathologic analysis of 100 cases with a review of the literature. Hum Pathol 1990; 21: 363. 150. Gast MJ, et al. Mullerian adenosarcoma of the cervix with heterologous elements: diagnostic and therapeutic approach. Gynecol Oncol 1987; 32: 381. 151. Ostor AG, Fortune DW. Benign and low grade variants of mixed Mullerian tumor of the uterus. Histopathology 1980; 4: 369. 152. Roth LM, Pride GL, Sharma HM. Mullerian adenosarcoma of the uterine cervix with heterologous elements. Cancer 1976; 37: 1725. 153. Baker TR, et al. Stage I uterine adenosarcoma: a report of six cases. J Surg Oncol 1988; 37: 128. 154. Benrubi GI, Pitel P, Lammert N. Small cell carcinoma of the ovary with hypercalcemia responsive to sequencing chemotherapy. South Med J 1993; 86(2): 247 – 8. 155. Scully RE. Small cell carcinoma of hypercalcemic type. Int J Gynecol Pathol 1993; 12: 148 – 52. 156. Tewari K, et al. Advanced-stage small cell carcinoma of the ovary in pregnancy: long-term survival after surgical debulking and multi-agent chemotherapy. Gynecol Oncol 1997; 66: 531 – 4. 157. Powell JL, et al. Uterine and ovarian conservation in advanced small cell carcinoma of the ovary. Obstet Gynecol 1998; 91: 846 – 8. 158. Senekjian EK, et al. Vinblastine, cisplatin, cyclophosphamide, bleomycin, doxorubicin, and etoposide in the treatment of small cell carcinoma of the ovary. Cancer 1989; 64: 1183 – 7. 159. Sarosh R, Bretta W, Yamada D. Stage IIIC small cell carcinoma of the ovary: Survival with conservative surgery and chemotherapy. Obstet Gynecol 2004; 103: 1120 – 3. 160. Young RH, Kozakewich HPW, Scully RE. Metastatic ovarian tumors in children: a report of 14 cases and review of the literature. Int J Gynecol Pathol 1993; 12: 8 – 19. 161. Moore JG, Schifrin BS, Erez S. Ovarian tumors in infancy, childhood and adolescence. Am J Obstet Gynecol 1967; 99: 913 – 22.
162. Paterson E. Distant metastases from medulloblastoma of the cerebellum. Brain 1961; 84: 301 – 9. 163. Deprest J, Moerman P, Corneillie P. Ovarian borderline mucinous tumor in a premenarchal girl: review on ovarian epithelial cancer in young girls. Gynecol Oncol 1992; 45: 219 – 24. 164. Moen MD, Cliby W, Wilson TO. Stage III papillary serous cystadenocarcinoma of the ovary in a 15 year old female. Gynecol Oncol 1994; 53: 274 – 6. 165. Seoud M, El-Kak F, Khalil A. Premenarchal ovarian serous borderline tumor: a case report and review of the literature. Acta Obstet Gynecol Scand 1996; 75: 762 – 4. 166. Pais RC, et al. Ovarian tumors in relapsing acute lymphoblastic leukemia: a review of 23 cases. J Pediatr Surg 1991; 26(1): 70 – 4. 167. Drinkard LC, et al. Acute myelomonocytic leukemia with abnormal eosinophils presenting as an ovarian mass: a report of two cases and a review of the literature. Gynecol Oncol 1995; 56: 307 – 11. 168. Chorlton I, Norris HJ, King FM. Malignant reticuloendothelial disease involving the ovary as a primary manifestation: a series of 19 lymphomas and 1 granulocytic sarcoma. Cancer 1974; 34: 397 – 407. 169. Osborne BM, Robboy SJ. Lymphomas or leukemia presenting as ovarian tumors: an analysis of 42 cases. Cancer 1983; 52: 1933 – 43. 170. Paladugu RR, Bearman RM, Rappaport H. Malignant lymphoma with primary manifestation in the gonad: a clinicopathologic study of 38 patients. Cancer 1980; 45: 561 – 71. 171. Monterroso V, Jaffe ES, Merino MJ. Malignant lymphomas involving the ovary. A clinicopathologic analysis of 39 cases. Am J Surg Pathol 1993; 17(2): 154 – 70. 172. Shakfeh SM, Woodruff JD. Primary ovarian sarcomas: report of 46 cases and review of the literature. Obstet Gynecol Surv 1987; 42(6): 331 – 49. 173. Kraemer BB, Silva EG, Sneige N. Fibrosarcoma of ovary. A new component in the nevoid basal-cell carcinoma syndrome. Am J Surg Pathol 1984; 8: 231. 174. Monk BJ, Nieberg R, Berek JS. Primary leiomyosarcoma of the ovary in a perimenarchal female. Gynecol Oncol 1993; 48: 389 – 93. 175. Silverman LA, Gitelman SE. Immunoreactive inhibin, mullerian inhibitory substance, and activin as biochemical markers for juvenile granulosa cell tumors. J Pediatr 1996; 129(6): 918 – 21. 176. Raafat F, Klys H, Rylance G. Juvenile granulosa cell tumor. Pediatr Pathol 1990; 10: 617 – 23. 177. Powell JL. et al. Management of advanced juvenile granulosa cell tumor of the ovary. Gynecol Oncol 1993; 48: 119 – 23. 178. Lappohn RE. et al. Inhibin as a marker for granulosa cell tumors. N Engl J Med 1989; 321: 790 – 3. 179. Gustafson ML. et al. Mullerian inhibiting substance as a marker for sex-cord tumor. N Engl J Med 1992; 326: 466 – 71. 180. Tamimi HK, Bolen JW. Enchondromatosis (Ollier’s disease) and ovarian juvenile granulosa cell tumor. A case report and review of the literature. Cancer 1984; 53: 1605 – 8. 181. Vaz RN, Turner C. Ollier (enchondromatosis) associated with ovarian juvenile granulosa cell tumor and precocious pseudopuberty. J Pediatr 1986; 108: 945 – 7. 182. Kuzma JF, King JM. Dyschondroplasia with hemiangiomatosis (Maffucci’s syndrome) and teratoid tumor of the ovary. Arch Pathol 1946; 46: 74 – 82. 183. Tanaka Y. et al. Ovarian juvenile granulosa cell tumor associated with Maffucci’s syndrome. Am J Clin Pathol 1992; 97(4): 523 – 7. 184. Khanfar NT, Arndt CAS, Wold LE. Squamous cell carcinoma of the ovary of a 14-year old girl. Mayo Clin Proc 1996; 71: 380 – 3. 185. Imachi M. et al. Malignant mixed mullerian tumor of the fallopian tube: report of two cases and review of literature. Gynecol Oncol 1992; 47: 114 – 24.
Section 11 : Pediatric Malignancies
70
Uncommon Endocrine Tumors in Children and Adolescents
Raul C. Ribeiro, Carlos Rodriguez-Galindo, Gerald P. Zambetti, Bonald C. Figueiredo, Karel Pacak, Andrew Bauer and Constantine A. Stratakis
ADRENOCORTICAL TUMORS IN CHILDREN INTRODUCTION The first case of childhood adrenocortical tumor (ACT) was reported in 1865.1 Cushing described the classical features of hypercortisolism in 1912; however, the role of adrenal tumors in this syndrome was not well understood until 1934.2 Childhood ACT has peculiar clinical and biological features unlike those observed in other pediatric carcinomas. The incidence of most childhood carcinomas increases with age, whereas 65% of ACTs occur in children younger than 5 years.3,4 In fact, this age distribution resembles that of tumors of embryonic origin. In this section of the chapter, we summarize the clinical and biological characteristics of and treatment outcome in children with ACTs.
EPIDEMIOLOGY Only about 0.2% of all cases of childhood cancer are ACTs. The frequency of ACT is 0.4 per million during the first 4 years of life, and it decreases to 0.1 per million during the subsequent 10 years. It then rises to 0.2 per million during the late teen years and reaches another peak during the fourth decade of life.5 This pattern is consistent with the concept that pediatric ACTs comprise at least two distinct disease groups. From the United States Surveillance Epidemiology and End Results (SEER) program data, it is estimated that there are 19–20 new cases of adrenocortical carcinoma per year in the United States in children and adolescents. The estimates do not include those cases of adrenocortical carcinoma that are misdiagnosed as adenomas and therefore are not reported. If it is assumed that one-third of all ACTs are adenomas, 25–30 cases of ACTs are estimated to occur annually in patients under the age of 20 in the United States. The incidence of ACT differs across geographic regions (see Figure 1) The incidence per million in children less
than 14 years of age ranges from 0.1 in Hong Kong and Mumbai to 0.4 in Los Angeles to 3.4 in southern Brazil.5 – 9 There is overwhelming evidence of a clustering of pediatric ACT in southern Brazil (see Table 1).10,11 In the United States, in contrast, only 36 cases have been reported to the SEER Program in 20 years.5 Similarly, a recent report from Eurocare that included population-based cancer registries of 20 European countries (1983–1994) revealed that only 65 of 25 457 cases of pediatric solid malignancy (0.26%) were ACTs.7 Moreover, there is compelling evidence that the germline TP53 R337H mutation explains the high incidence of pediatric adrenocortical carcinoma in southern Brazil and is involved in its tumorigenesis.12,13 First, germline TP53 mutations, in general, increase the predisposition to pediatric ACT.14 Laboratory findings that strongly suggest that the TP53 R337H plays a role in adrenal tumorigenesis include the loss of heterozygosity (LOH) with retention of the mutant allele in tumor cells, the accumulation of TP53 protein in the nucleus, and the folding and other properties of the missense TP53 R337H protein in vitro. The penetrance of ACT was only about 10% in a large cohort of carriers of the TP53 R337H from families known to have children with ACT.15 Other genetic changes found with variable frequency in children with the TP53 R337H and ACT, including amplification of the 9q34 chromosomal region (detected in eight of nine tumor samples from children with ACT in Curitiba, Brazil),16 an increased copy number of the steroidogenic factor 1 (SF-1 ) gene in eight of the cases with 9q34 amplification,17 germline missense mutation in the inhibin α-subunit gene,18 and other yet-to-be-determined factors may contribute to adrenal cell transformation.
PATHOGENESIS The adrenal cortex arises from the coelomic mesoderm during approximately the sixth week of development. These mesothelial cells proliferate between the dorsal mesentery and the developing gonad, delineating two histologically
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
776
PEDIATRIC MALIGNANCIES
the lung, pancreas, and prostate, and gonadal germ cell tumors.26,28 In 1990, Malkin et al.28 screened five of these families and found germline mutations clustered in exon 7 of the TP53 gene in all the five. It is now well recognized that most of the constitutional genetic abnormalities in young children with ACT are the germline mutations in various exons of TP53. In fact, it is likely that more than 90% of young children with ACT have an inherited TP53 mutation. For example, Varley et al.14 found germline TP53 mutations in 9 of 13 cases of pediatric ACT selected without reference to the family’s history of cancer. The Beckwith-Wiedemann syndrome (BWS) is also a rare genetic disorder with a higher than expected incidence of ACT. It includes congenital umbilical hernia, macroglossia,
Brazil Great Britain Finland United States Spain Israel Canada Germany Australia China Italy
Males
Norway
Females
Sweden New Zealand 0
0.2
0.4 0.6 0.8 1.0 Incidence (per million)
1.2
1.4
Figure 1 Incidence of childhood adrenocortical carcinoma in different countries (per million).
distinct components: an outer zone from which the “adult cortex” originates and a more central zone called fetal cortex. The latter comprises the largest portion of the adrenal cortex at birth. It begins to undergo apoptosis by the last intrauterine month and disappears toward the end of the first year of life.23 The fetal cortex is responsible for 90% of the mother’s production of dehydroepiandrosterone (DHEA) and its sulfated derivative (DHEA-S).24 Predisposing constitutional genetic factors have been found in the majority of children and adolescents with ACT (see Table 2). Li and Fraumeni observed a remarkably high frequency of ACT (4 cases, or 10%) among 44 malignancies in children from families in which diverse cancers segregated in an autosomal-dominant pattern.25,26 In addition to childhood sarcoma and premenopausal breast cancer, members of these families are at increased risk of other malignancies, including leukemia, brain tumors, osteosarcomas, and adrenocortical carcinomas.27 Other possible tumors associated with this syndrome include melanoma, carcinomas of
Table 2 Constitutional genetic abnormalities associated with adrenocortical tumors (ACT).a
Condition
Tumor types
Observations
Li-Fraumeni syndrome and other germline P53 mutations BeckwithWiedemann syndrome
Adenomas, carcinomas
10% of these tumors are ACT
Adenomas, carcinomas
ACT is the second most common tumor (approx. 15% of children with this syndrome have ACT) 20% of these tumors are ACT Testicular tumors of heterotopic adrenal cortical tissue PPNAD occurs in approx. 25% of patients; common in children
Hemihypertrophy
Adenomas, carcinomas Adenoma, carcinoma (very rare) Primary pigmented nodular adrenocortical disease (PPNAD) Nodules, adenomas, carcinomas
Congenital adrenal hyperplasia Carney complex
Multiple endocrine neoplasia type I
a
Median age of patients with carcinomas is 40 years
Modified from Ribeiro et al.29 with permission from Elsevier, 2004.
Table 1 Comparison of cases of ACTs registered in select hospitals in the Brazilian states of S˜ao Paulo, Paran´a, and Bahia versus cases in the United States and Europe.a
Institution 11
HCACC SCSP11 CIDB19 IC11 HCFMSP11 HCC10 HCFMRP20 ONCO/HSR21 SEER5 Eurocare7
State
Period
Number of patients
Age (years)
TP53 R337H tested/positive (%)
S˜ao Paulo S˜ao Paulo S˜ao Paulo S˜ao Paulo S˜ao Paulo Paran´a S˜ao Paulo Bahia USA Europe
1978 – 2003 1985 – 2004 1980 – 2004 1977 – 2002 1982 – 2003 1966 – 2003 1985 – 2003 1981 – 2004 1975 – 1995 1983 – 1994
56 17 78 46 27 124 21 6 36 65
<21 <14 <18 <13 <16 <13 <12 <12 <20 <15
NA NA 20/20 (100) NA 21/27 (78) 61/65 (93.8) 12/16 (75) 0/5 (0) NA NA
Note: HCACC, Hospital do Cancer AC Camargo; SCSP, Santa Casa de S˜ao Paulo; IC, Instituto da Crian¸ca da Universidade de S˜ao Paulo; CIDB, Centro Infantil Domingos Boldrini, Universidade de Campinas; HCFMSP, Hospital das Cl´ınicas da Faculdade de Medicina da Universidade de S˜ao Paulo; HCC, Hospital de Cl´ınicas de Curitiba, Universidade Federal do Paran´a; HCFMRP, Hospital das Cl´ınicas da Faculdade de Medicina de Ribeir˜ao Preto; ONCO/HSR, Sociedade de Oncologia da Bahia Ltda/Hospital S˜ao Rafael; SEER, Surveillance Epidemiology and End Results; NA, not available. a Modified from Pianovski et al.22 with permission from John Wiley & Sons. Inc.
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS
and gigantism,30 but a wide phenotypic expression has been noted. Genetic linkage analysis has provided evidence that familial BWS is associated with regulatory abnormalities at the chromosome 11p15 region.31,32 The genes for insulinlike growth factor-II (IGF-II) and H19 have been mapped to this site. Adjacent genes have been shown to be paternally imprinted and maternally expressed.33 They appear to be important regulators of fetal adrenal growth.34,35 The loss of imprinting of these genes may cause their overexpression, thereby leading to oncogenesis.36 Other abnormalities involving both chromosomes 11 and 17, such as LOH,37 translocations,38 and mutations39,40 have been reported. Congenital hemihypertrophy, an unusual entity associated with an asymmetric overgrowth of a whole side or part of the body, has consistently been associated with several benign and malignant neoplasias, including ACT.41 – 45 Chronic stimulation of the adrenal gland by adrenocorticotropic hormone (ACTH) is thought to contribute to tumorigenesis.46 However, ACTs are described only rarely in patients with congenital adrenal hyperplasia.47 – 49 Carney complex and MEN150 – 55 are reviewed elsewhere in this chapter. The relationship between environmental factors and ACT is difficult to prove. Mann et al. reported the case of a child with ACT whose mother ingested hydroxyprogesterone hexanoate during pregnancy to prevent miscarriage.56 Hornstein et al. described a case of adrenocortical carcinoma in a child with fetal alcohol syndrome; however, the child also had hemihypertrophy and a familial history of cancer.57 Dedov and Norets administered [75 Se]selenomethionine to rats on days 11–13 of the gestation period. They found that prenatal exposure to the radionuclide was associated with the development of ACTs along with other endocrine and nonendocrine tumors in their offsprings.58 The origin of the TP53 R337H mutation in southern Brazil remains unexplained.
777
Table 3 Signs and symptoms of adrenocortical tumors in 58 children.
Feature
N
%
Pubic hair Hypertrophy Clitoris Penis Acne Deep voice Hypertension Facial hair Facial hyperemia Palpable tumor Weight gain Hirsutism Moon face Accelerated growth velocity Centripetal fat distribution Buffalo hump of the neck Seizures
53
91 84 62 22 72 55 55 50 48 48 38 36 33 29 24 19 12
36 13 42 32 32 29 28 28 22 21 19 17 14 11 7
CLINICAL FEATURES Among 254 children enrolled in the International Pediatric Adrenocortical Tumor Registry (IPACTR),4 the median age at diagnosis of ACT was 3.2 years. Fewer than 15% of patients were 13 years or older at diagnosis. The incidence was higher among girls, with the overall female : male ratio being 1.6 : 1; however, the ratio ranged from 1.7 : 1 in the 0- to 3-year-old group to 6.2 : 1 in adolescents (≥ 13 years). Although the reason(s) for this prevalence in females is not understood, there is evidence of gender-specific physiologic changes in the adrenal glands.59 Signs and symptoms of virilization were the most common presenting clinical manifestation (>80% of patients), as reported previously.3,60 – 62 Clinical manifestations of ACT can be present at birth.63 An acute abdomen due to spontaneous tumor rupture is rarely the presenting clinical manifestation of ACT.64 In a detailed review of the presenting features of 58 cases of childhood ACT,62 (see Table 3) features of virilization included pubic hair, facial acne, clitorimegaly, voice change, facial hair, hirsutism, muscle hypertrophy, growth acceleration, and increase in penis size (see Figure 2). Virilization was observed either alone (virilizing tumors, 40%
Figure 2 Virilizing adrenocortical carcinoma in a 10-year-old boy.
of patients) or accompanied by clinical manifestations of the overproduction of other adrenal cortical hormones, including glucocorticoids, androgens, aldosterone, or estrogens (mixed type, 45%) (see Figure 3). About 10% of patients showed no clinical evidence of an endocrine syndrome at presentation
778
PEDIATRIC MALIGNANCIES
Figure 3 Mixed type of adrenocortical carcinoma in a 6-year-old girl.
(nonfunctional tumors). Finally, overproduction of glucocorticoids alone (Cushing’s syndrome) was evident in only 3% of patients. Interestingly, patients with an excess of glucocorticoid did not show striae. Primary hyperaldosteronism (Conn’s syndrome) and pure feminization occur very rarely.59,65 – 67 None of the patients in the registry manifested either syndrome. The most common presenting clinical manifestations of hyperaldosteronism include headache, weakness of proximal muscle groups, polyuria, tachycardia with or without palpitation, hypocalcemia, and hypertension. The most frequent sign of feminization is gynecomastia. Tumor size varied widely among children for whom this information was available in the IPACTR. The tumor weight was greater than 200 g in 83 of 182 cases.4 There was no predominant tumor laterality. Bilateral tumors were not observed in patients enrolled in the IPACTR, but they have been occasionally reported.59,68,69 Ectopic ACT, which occurs very rarely, has been described in the spinal canal,70 paratesticular region,71 and intrathoracic cavity.72 Ectopic adrenal cortical tissue is commonly detected retroperitoneally in the celiac plexus, kidney, genitalia, broad ligaments, epididymis, and spermatic cord,73 and can act as a potential site for the development of ACT. Other relevant clinical presenting features of 58 cases of childhood ACT included elevated blood pressure in 55% of the patients, 12% of whom had hypertensive crisis (associated with seizures in one patient). Hypertension was common in patients with glucocorticoid-secreting tumors (Cushing’s or mixed type); it also occurred in half of the
patients with signs of virilization only and in three patients with nonfunctioning tumors. Treatment of hypertension in patients with ACT can be challenging. Fatalities due to hypertensive crisis have been reported.62 In general, patients have responded well to captopril, although in some cases it has been necessary to add other drugs, such as ketoconazole. Children and adolescents with functional ACT are subject to growth disturbances.74 Pure androgen and estrogen excess most often results in increased growth velocity and premature epiphyseal closure. In the Curitiba series, the height and weight of children with ACT often exceeded the 50th percentile at diagnosis. Patients with a greater than expected height for age included not only those with the virilizing form of ACT but also those with the mixed form. Bone age was advanced more than 1 year in 68% of the patients. Markers of growth and development have consistently remained within the normal range in long-term survivors.75 True precocious puberty is rarely noted. In many instances, the increased somatic growth of these children, their generally healthy appearance, and the lack of a palpable abdominal mass diverted pediatricians from the possibility of a malignancy. To avoid delaying the diagnosis of ACT, any child less than 4 years with pubarche should be considered to have ACT until proven otherwise. In addition, the presence of acne in an infant can be considered pathognomonic of an adrenocortical lesion. Finally, because Cushing’s syndrome is very rare in children, it should be considered highly indicative of ACT in children younger than 10 years.76
DIAGNOSIS The diagnosis of ACT is made on the basis of the gross and histologic appearance of tissue obtained surgically. The pathologic classification of pediatric ACT is troublesome. Even an experienced pathologist can find it difficult to differentiate adenoma from carcinoma (see Figures 4 and 5). Weiss et al.77,78 and Hough et al.79 formulated classification systems on the basis of macroscopic, microscopic, and clinical features present at diagnosis. Bugg et al.80 used a modified criteria of Weiss et al.77,78 to analyze a large series of pediatric ACT. In this study, the adrenal tumors were divided into three groups: adrenocortical adenomas, high-grade carcinomas, and low-grade carcinomas. Highgrade carcinoma and tumor weight were the most reliable predictors of outcome. Measurement of urinary 17-ketosteroids (17-KS) frequently provides the pivotal clue to the diagnosis of ACT. Most patients with ACT who are tested have elevated levels of 17-KS. In one review,62 48 of 49 patients tested positive for elevated 17-KS levels, whether their tumors caused Cushing’s syndrome or virilization. Plasma DHEA-S levels are abnormal in approximately 90% of cases, suggesting that increased DHEA-S level is the second most sensitive tumor marker. Abnormal urinary DHEA concentration is less sensitive, occurring in only two-thirds of cases. Urinary 17-hydroxycorticosteroid (17-OH) levels are
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS
Figure 4 Virilizing adrenocortical carcinoma with prominent nuclear pleomorphism.
elevated when there are clinical signs of excessive glucocorticoids. In our experience, a dexamethasone suppression test has rarely been necessary. Elevated glucocorticoid and androgen levels are strong indications of adrenal tumor. Routine laboratory evaluation for suspected ACT includes measurement of urinary 17-KS, 17-OH, and free cortisol, as well as of plasma cortisol, DHEA-S, testosterone, androstenedione, 17-hydroxyprogesterone, aldosterone, renin activity, deoxycorticosterone (DOC), and other 17-deoxysteroid precursors. This comprehensive panel of tests not only contributes to the diagnosis but also provides useful markers for the detection of tumor recurrence. Several different imaging modalities are used to establish the diagnosis of ACT. Computed tomography (CT), sonography, magnetic resonance imaging (MRI), and positron emission tomography (PET) scanning are most commonly used. Although ultrasound examination has its limitations, it is important for evaluating tumor extension into the inferior vena cava and right atrium.81 At the St. Jude Children’s Research Hospital, the use of MRI has steadily increased over the past few years. This modality has several advantages over CT, including the absence of ionizing radiation,
779
Figure 5 Virilizing adrenocortical adenoma. Note that the cells are arranged in alveolar clusters.
the capability of imaging multiple planes, and improved tissue contrast differentiation. On CT imaging, ACT is usually well demarcated with an enhancing peripheral capsule. Large tumors usually have a central area of stellate appearance caused by hemorrhage, necrosis, and fibrosis. This stellate zone is hyperintense on T2-weighted and Short-inversion recovery (STIR) MR images. Calcifications are common. Because ACT is metabolically active, fluorodeoxyglucose (FDG)-PET imaging is increasingly used in patients with ACT (see Figure 6).82 C-metomidate (MTO), a marker of sustained 11 − β-hydroxylase activity, has been investigated as an alternative PET tracer for adrenocortical imaging. The clinical indications of PET with this new tracer are still evolving.83 Although FDG/MTO-PET is unlikely to add information to that obtained with CT or MRI of the primary tumor and its regional extension, it can disclose distant metastases that are not readily detected on CT or MRI. PET imaging can also detect tumor recurrence in areas that may be missed on routine follow-up imaging. However, this technique has not been systematically evaluated in pediatric ACT. Presently, our recommendation is that, in addition to ultrasonography, all patients who have a suspected adrenal tumor
780
PEDIATRIC MALIGNANCIES
local relapse. An ipsilateral modified node dissection is performed, extending from the renal vein to the level of bifurcation of the common iliac vessel. If there are contralateral clinically enlarged lymph nodes, they should be removed as well. Surgical resection of recurrent local and distant disease is also important. In the latter case, multiple surgical resections may be necessary to render patients free of disease. This aggressive approach is associated with prolonged survival, particularly when combined with chemotherapy. Because of tumor friability, rupture of the capsule and tumor spillage are frequent (occurring in approximately 20% of cases during the initial procedure and in 43% after local recurrence).62 Infiltration of the vena cava may make radical surgery difficult in some cases, although successful complete resection of the tumor thrombus has been reported with cardiopulmonary bypass.84,85 Surgery requires careful and precise perioperative planning. All patients with a functioning tumor are assumed to have suppression of the contralateral adrenal gland, and therefore steroid replacement therapy is given. Special attention to electrolyte balance, hypertension, surgical wound care, and infectious complications is imperative.
(a)
(b)
Chemotherapy
(c) Figure 6 A 4-year-old girl with pulmonary metastatic adrenocortical carcinoma. (a) CT scan shows abnormal soft tissue in the right and left lower lobes. (b) Corresponding FDG-PET image shows increased accumulation of FDG within the lesions shown on the CT scan, consistent with metastatic tumor. (c) Fusion of CT and PET images in upper panels.
should be examined by CT or MRI. Because the liver and lungs are the most common sites of metastasis at the time of diagnosis, CT scans of the chest and abdomen are recommended for all newly diagnosed patients. The skeleton and central nervous system are involved in a few cases. Technetium bone scans are typically obtained in the initial evaluation of children with ACT. Imaging of the central nervous system is not routinely performed at presentation.
TREATMENT Surgery Surgery is the single most important procedure in the successful treatment of ACT. It is performed by a transabdominal approach, usually using an ipsilateral subcostal incision, which may be modified to a chevron or bilateral subcostal incision. En bloc resection, which may include the kidney, portions of the pancreas and/or liver, or other adjacent structures, may be necessary in rare cases of large, locally invasive tumors. A thoracoabdominal incision is indicated in rare cases. The role of regional lymph node dissection in pediatric ACT has not been evaluated, but it has been advocated because patients with large tumors commonly experience
The role of chemotherapy in the management of childhood ACT has not been established. Mitotane [1,1-dichloro-2-(ochlorophenyl)-2-(p-chlorophenyl)ethane, or o,p -DDD], an insecticide derivative that produces adrenocortical necrosis, has been used extensively in adults with ACT, but its efficacy in children is not known. Mitotane has been used to treat advanced metastatic ACT, given prior to surgery in cases of inoperable tumors or after surgery in patients at high risk of relapse (adjuvant chemotherapy), combined with other agents, and used to control symptoms associated with increased production of adrenal hormones. Objective responses to mitotane are obtained in 15 to 60% of treated adult patients.86 The wide variation in response rates may, in part, reflect the pharmacokinetics of mitotane. There has been evidence of greater tumor response when the plasma concentration of mitotane is above 14 mg L−1 .87 The most important common toxicities of mitotane are nausea, vomiting, diarrhea, and abdominal pain. Less-frequent reactions include somnolence, lethargy, ataxic gait, depression, and vertigo. Of interest, prepubertal patients can develop gynecomastia or thelarche. Another shortcoming of mitotane treatment is that it significantly alters steroid hormone metabolism, so that blood and urine steroid measurements cannot be used as markers of tumor relapse. Mitotane should be considered as an experimental agent in the treatment of children with ACT. Other chemotherapeutic agents, including 5-fluorouracil, etoposide, cisplatin, carboplatin, cyclophosphamide, doxorubicin, and streptozocin, have been used alone or in combination to treat ACT, with varied results.88,89 There has been no formal trial of conventional chemotherapy agents in pediatric ACT, but the available case reports and the experience of the IPACTR suggest that a subset of pediatric ACT is chemotherapy sensitive. The combination used most often in pediatrics consists of cisplatin and etoposide with or without doxorubicin given with mitotane.
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS
OUTCOME In the IPACTR series, of the 254 patients with known outcomes, 97 (38.2 %) died and 157 (61.8%) were alive at a median follow-up of 2 years and 5 months (range, 5 days to 22 years).4 The survival rate in this study is similar to that reported by others.90 Five patients died of causes unrelated to tumor progression (two died of infection, one of hypertensive complication, one of massive hemorrhage during surgery, and one of an unspecified complication). The 5-year event-free survival (EFS) and overall survival estimates were 54.2% (95% CI, 48.2–60.2%) and 54.7% (95% CI, 48.7–60.7%), respectively.
781
Table 4 Staging criteria for childhood adrenocortical tumor.
Stage
Description
I
Tumor totally excised, tumor weight/size <100 g or <200 cm3 , absence of metastasis, and normal hormone levels after surgery Tumor totally excised, tumor weight/size ≥ 100 g or ≥ 200 cm3 , absence of metastasis, and normal hormone levels after surgery Unresectable tumor, gross or microscopic residual tumor, tumor spillage during surgery, persistence of abnormal hormone levels after surgery, or retroperitoneal lymph node involvement Distant tumor metastasis
II
III
IV
PROGNOSTIC FACTORS Complete tumor resection is the single most important prognostic indicator. Patients who have residual disease after surgery have a dismal prognosis. Of 57 patients in the IPACTR who had distant or local, gross or microscopic residual disease after surgery, only 8 have remained free of disease. Conversely, the long-term survival rate is around 75% for children with completely resected tumors. Among the latter, tumor size has prognostic value. Registry data showed that among 192 such patients, those with tumors weighing more than 200 g had an EFS estimate of 39%, compared to 87% for those with smaller tumors. Tumor size has been consistently associated with prognosis in several studies of ACT.91,92 Children whose tumors produce excess glucocorticoid appear to have a worse prognosis than those who have pure virilizing manifestations. Classification schemes or disease staging systems to guide therapy for pediatric ACT are still evolving. A modification of the staging system of Sandrini and colleagues includes tumor volume and resectability (see Table 4).62 It is likely that prognostic factor analysis can be further refined by adding other predictive factors. For example, rupture of the tumor pseudocapsule during surgery and invasion of the vena cava were found to be associated with poor prognosis even among patients whose tumors were completely resected; but these variables are yet to be prospectively analyzed. Moreover, some histologic tumor features, such as vascular or capsular invasion, extensive necrosis, and marked mitotic activity, were independently associated with prognosis in a recent study.91
OTHER RARE PEDIATRIC ENDOCRINE TUMORS INTRODUCTION Endocrine-related cancer is extraordinarily rare in children. Most endocrine tumors in children are benign; pituitary, thyroid, and adrenal adenomas are by far the most common endocrine masses in childhood. Sporadic (nonfamilial) carcinomas in children are mostly thyroid and adrenal. Most other endocrine malignant tumors in children occur in the context of genetic conditions predisposing to multiple neoplasias: MEN 1 and MEN 2, McCune-Albright syndrome (MAS), CNC von Hippel–Lindau disease (VHLD), PeutzJeghers syndrome (PJS), Cowden disease (CD), hereditary hyperparathyroidism, and jaw tumor syndrome (HPJTS),
and other less-frequent conditions such as Burt-Hogg-Dub´e. Occasional sporadic parathyroid cancer, malignant testicular and ovarian tumors, and pheochromocytomas or other paragangliomas have also been reported, each in a handful of cases in the pediatric age-group. This section of the chapter reviews the main multiple endocrine neoplasias (MEN 1, MEN 2, and CNC) and some of the related syndromes (PJS, CD, and VHL). Table 5 provides a comprehensive list of endocrine tumors in childhood; few additional genetic conditions listed there are either very rare or affect the occasional child, being for the most part diseases of the young adult. Finally, this section reviews thyroid cancer and pheochromocytoma, both extraordinarily rare tumors in childhood.
MULTIPLE ENDOCRINE NEOPLASIAS AND RELATED SYNDROMES Introduction In the last 50 years, five MEN syndromes, all inherited in an autosomal-dominant manner, have been recognized as distinct clinical entities (see Table 5): MEN type 1 (MEN 1), which was first reported in a family with multiple parathyroid, pancreatic islet, and pituitary tumors,93 MEN 2, which was first identified in a patient with pheochromocytoma and carcinoma of the thyroid gland94 and has since been subdivided into three clinical syndromes,95 and CNC, first reported in a series of familial and sporadic cases with multiple endocrine and other tumors.96 In the last 10 years, genetic defects have been identified for all these syndromes.97 – 99 In addition to the “classic” MEN syndromes, several other genetic conditions associated with one or more endocrine tumors and also inherited as autosomal-dominant traits, have been described and their molecular defects elucidated. They include PJS and Cowden and von Hippel–Lindau diseases (CD and VHLD, respectively).100 – 103 Here we present, in brief, the newest information on the clinical and molecular genetics of the MENs (MEN 1, MEN 2, and CNC) for which we follow the suggested nomenclature104 and terminology used by Mendelian inheritance in man (MIM).105 Among the related syndromes (PJS, CD, VHLD), PJS is discussed somewhat more extensively, because of its similarities with CNC and the fact that it is relatively unknown for its association with endocrine tumors;
782
PEDIATRIC MALIGNANCIES
Table 5 Comprehensive list of endocrine tumors in childhood.
Disorder
Occurrence
Gene
Locus
Pituitary adenomas Thyroid tumors Adrenal tumors Bilateral adrenal hyperplasias Adrenal cancer MEN 1 MEN 2 McCune-Albright syndrome Carney complex Peutz-Jeghers syndrome von Hippel – Lindau disease Cowden disease Hyperparathyroidism/jaw tumors Burt-Hogg-Dub´e Pheochromocytomas/paragangliomas Carney triad Other endocrine cancers
∼ 0.1 per million per year ∼ 1% of all tumors Rare Rare Rare Rare Rare Rare Rare Rare Rare Rare Very rare Very rare Very rare Very rare Very rare
Various RET-PTC, other Various PRKAR1A, other TP53, other Menin RET GNAS PRKAR1A, other STK11/LKB1 VHL PTEN HPTJ BHD SDHB, -C, and -D Unknown Various
Several Several Several 17q, 2p, other 17p, other 11q13 10qcen 20q 17q, 2p, other 19p13 3p 10q2 1q 17p 11p, 11q, 1q Unknown Several
the other two conditions, CD and VHLD are only briefly mentioned; they have been extensively reviewed elsewhere, including endocrine journals.
than age 12 in MEN 1.113,114,116 MEN 1-related insulinoma has been described as early as age 6.119 Molecular Genetics of MEN 1
MEN 1 Clinical Presentation and Age at Presentation
Familial MEN 1 is an autosomal-dominant disorder with variable penetrance characterized by tumors of the parathyroids, pancreas and other locations of the gastrointestinal tract, anterior pituitary, and other tissues. The MEN1 gene, menin, which is located on chromosome 11q13, was recently identified.98 Individuals affected by MEN 1 inherit an MEN1 -inactivated allele; tumorigenesis in specific tissues follows inactivation of the remaining normal allele. MEN 1 is regarded as an endocrinopathy presenting in young adulthood.106 Like many hereditary tumors, MEN 1 typically presents earlier than sporadic tumors of the same tissue type. The age disparity at presentation is striking for MEN 1-related hyperparathyroidism (third vs sixth decade in sporadic cases), but less-pronounced for gastrinoma and insulinoma (fourth vs fifth decade in each).107 By contrast, there has been no apparent age difference for the onset of MEN 1-associated and sporadic prolactinoma (fourth decade).108 – 110 The age of clinical onset of treatable MEN 1-related tumors is an important factor in formulating biochemical and DNA screening recommendations. Hyperparathyroidism is usually the earliest and most common endocrine expression. In several studies, hypercalcemia was an almost universal finding at the time of MEN 1 diagnosis.109,111,112 Clinically evident hyperparathyroidism in MEN 1 has been reported at ages 5 and 7 years113 and several times at age 8.109,114 However, no morbidity has been reported from early hyperparathyroidism in MEN 1. MEN 1-associated prolactinomas have been reported previously in children 10 to 13 years old.115 – 117 Growth hormone (GH)producing adenomas in MEN 1 have been described in early adolescence,117 whereas a somatomammotroph adenoma was recently reported in a 5-year-old boy.118 Gastrinoma, the other defining feature of MEN 1, has not been seen earlier
Identification of the MEN1 gene was first achieved by positional cloning,98 and mutation testing is now available at several centers.106,107 Menin testing would not regularly result in major therapeutic interventions and timely medical benefit; by the time most manifestations have clinical significance, diagnosis can be made without the need for molecular testing. Furthermore, there have been more than 150 mutations in menin; they are spread almost equally along the length of the gene without any apparent genotype –phenotype correlation.98,106,107,118 Some mutations occur more frequently than others, but there are no obvious “hot-spots” in the nine coding exons of the gene. The mutations that are more frequent involve exons 83, 84, 209–211, and 514–516, which contain unstable DNA sequences [e.g. dinucleotide repeats or poly (C) tracts]. A small number of MEN 1 kindred do not have mutations in the MEN1 gene, but there is no obvious genetic heterogeneity in the syndrome; it is likely that mutations in these families and sporadic cases are in the noncoding regions of the gene.120 Because of the relatively late clinical manifestations of the disease, the lack of genotype –phenotype correlation, and the possibility that menin mutation may not be found in patients affected with the disease, MEN1 gene testing has not been routinely advocated in young individuals or in lieu of clinical diagnosis.106,107,118 On the other hand, the recent recognition of significant morbidity from MEN 1 at young ages,118 the need for carrier testing in large families with known mutations, and advances in molecular testing may lead to more frequent genetic testing in the future.
The MEN 2 Syndromes Clinical Presentation and Age at Presentation
The three MEN 2 syndromes are the phenotypic expressions of variants of activating mutations in the RET protooncogene, a tyrosine kinase receptor.121 Thus, all clinical
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS
manifestations of the MEN 2 syndromes reflect the inappropriate transmission of this growth- and survival-promoting signal in the neural crest–derived tissues that naturally express RET.95,122 The primary and most common lesion of the MEN 2 syndromes is medullary thyroid cancer (MTC); it is present in up to 95% of patients with MEN 2. It is preceded by hyperplasia of the calcitonin-secreting thyroid parafollicular C cells.95,122,123 Despite the presence of histologic evidence of MTC in more than 95% of the obligate gene carriers by age 35, calcitonin hypersecretion is not as specific as was initially believed. A small fraction of people identified on the basis of calcitonin-provocative tests as not having MEN 2A or 2B have been shown to be obligate noncarriers of their kindred’s characteristic RET mutation. Conversely, those classified as being affected on the basis of C cell hyperplasia alone should be reevaluated with RET testing to allow for accurate counseling about their children’s risk for MEN 2.95,122,123 Pheochromocytoma typically affects 50 to 60% of MEN 2A kindred with a high rate of bilaterality, but a low rate of extra-adrenal sites or malignancy. The tumor is commonly present in MEN 2B and absent by definition in familial MTC. A yearly program for pheochromocytoma that uses plasma metanephrines to screen can effectively identify tumors at an early stage (usually around 1 cm), before the development of hypertension or other adverse sequelae. Laparoscopic adrenalectomy now affords a less-invasive approach to small pheochromocytomas. Parathyroid disease is detected clinically in 10 to 15% of patients with MEN 2A. Recently, two variants of MEN 2A with distinctive nonendocrine manifestations were recognized–MEN 2A with cutaneous lichen amyloidosis, which is characterized by pruritic lesions composed of subepidermal keratin deposits over the scapular region;124 and MEN 2A with partial or extensive Hirschsprung’s syndrome, which represents another distinct clinical syndrome.125 Patients with this disease exhibit evidence of both RET hyperfunction and hypofunction in a tissue-specific fashion.126 Expanding the clinical spectrum of the MEN 2 syndromes are the distinctive skeletal findings found in virtually all MEN 2B patients: elongated facies with a long, relatively thin nose; proliferation of corneal nerves; mucosal neuromas of the lips and tongue; and gastrointestinal ganglioneuromas.126 In addition, these patients frequently have aggressive tumors, both MTC and pheochromocytoma. All these manifestations appear to stem from abnormal proliferation of neural crest elements during fetal and postnatal life and be orchestrated by a hyperfunctioning RET tyrosine kinase receptor.127 Molecular Genetics and Screening for the MEN 2 Syndromes
Analyzing the RET gene in the MEN 2A and familial MTC syndromes revealed that most of the patients had mutated one of five cysteine codons in exons 10 and 11; however, isolated familial MTC was also caused by mutations in the first tyrosine kinase domain of the receptor.127 The cysteineto-arginine change in codon 634 of the RET protein appears to be the most common mutation in the MEN 2 syndromes and is also found in the few kindred with MEN 2A that have
783
pheochromocytoma and hyperparathyroidism but no other clinical manifestations. In MEN 2A/Hirschsprung’s disease, most kindred have mutations in codons 609, 618, and 620. In MEN 2B, the mutations are in the tyrosine kinase domain. Over the past 35 years, there has been a substantial improvement in survival in MEN 2 families, an improvement largely attributable to the success of family screening programs, first with provocative biochemical tests and, more recently, with detection of RET gene mutations.121 – 123 RET gene analysis has superseded older methods because of its high sensitivity and specificity, utility in younger children, and much higher degree of patient acceptance.127 A typical screening program for known MEN 2A and familial MTC kindred is to initiate testing for RET mutations at ages 4 to 6. In the case of MEN 2B, it is generally possible to recognize the characteristic mucosal neuroma phenotype within the first 2 years of life. Prophylactic thyroidectomy at 5 years is now recommended for all patients who test positive for disease-causing mutations of the RET gene. In addition to established MEN 2 kindred, 6 to 8% of MTC patients with no apparent family history of the disorder harbor germline RET mutations and may thus have offsprings at risk. On the basis of this figure, it appears prudent to offer germline RET analysis to everyone with apparent sporadic MTC. Finally, mutations of RET and some of its ligands, including glial-derived nerve growth factor (GDNF ), have been found in approximately half of the 10% of familial cases of Hirschsprung’s disease, and its variants.128,129 However, screening for RET gene mutations (and mutations of its ligands) for familial Hirschsprung’s disease does not seem to affect the clinical care for this disease, which usually presents in infancy or early childhood.
Carney Complex (CNC) Clinical Presentation and Age at Presentation
CNC was first reported in 1985; the syndrome described the association of heart myxomas and lentigines with pituitaryindependent Cushing’s syndrome caused by an unusual adrenal pathology.96,130 – 135 The latter was characterized by multiple, small, pigmented, adrenocortical nodules and internodular cortical atrophy; the disease was shown to be primary and bilateral and is now commonly referred to as PPNAD. In the late 1980s, several patients were described with PPNAD and various combinations of myxomas affecting multiple organs (heart, skin, and breast), spotty skin pigmentation (lentigines and blue nevi), and tumors of three endocrine organs (adrenal, pituitary, and testis).131 – 133 Subsequently, the syndrome was shown to be transmitted in a manner consistent with dominant inheritance.134,135 The association of heart myxomas, spotty skin pigmentation, and other lesions in young patients that did not “fit” the profile of the usual patients with this tumor (most commonly older females) have been described before.136,137 Indeed, heart myxomas in CNC are often multiple, affect any or all cardiac chambers, occur at a relatively young age, and are equally distributed between the sexes.138 Other myxomas, with a somewhat peculiar distribution, were later found to be part of the syndrome: cutaneous myxomas, which have
784
PEDIATRIC MALIGNANCIES
a predilection for the eyelids, breast nipple, and external ear canals;139,140 and mammary myxoid fibroadenomas,141 which are often associated with other multiple foci of myxomatous changes.141,142 Centrofacial spotty pigmentation in patients with CNC, like that in those with PJS, involves the vermilion border of the lips and the conjunctiva. The conjunctival pigmentation typically affects the lacrimal caruncle and the conjunctival semilunar fold and may involve the sclera. Other pigmented skin lesions include blue and other types of nevi and caf´e au lait spots (CALS).130,143,144 Pigmented lesions are often present in the vaginal and penile mucosa of affected female and male patients, respectively. Cushing’s syndrome in CNC is the most common endocrine manifestation and is always caused by PPNAD. PPNAD is a form of bilateral adrenal hyperplasia; however, the glands are most commonly normal sized or even small and peppered with black or brown nodules set in a cortex that is usually (although not always) atrophic. Despite their small size (less than 6 mm), the nodules are visible on CT or MRI of the adrenal glands.145 Patients with CNC often present with a variant Cushing’s syndrome called atypical Cushing’s syndrome (ACS),146 which is characterized by an asthenic, rather than obese, body habitus caused by severe osteoporosis, short stature, and muscle and skin wasting. ACS was recognized as early as 1956 and has since been described in several cases of patients with Cushing’s syndrome.147 – 149 More recently, this condition was associated with CNC150,151 including the case of a patient who presented 27 years after unilateral adrenalectomy.150 Patients with ACS tend to have normal or near-normal 24-hour cortisol production, but this is characterized by the absence of the normal circadian rhythmicity of this hormone.146 – 152 Occasionally patients present with “periodic Cushing’s syndrome” (PCS), a variant of Cushing’s syndrome that is frequently found in children with the complex.151,152 All patients with PPNAD and classic CS, and most patients with ACS or PCS respond to dexamethasone with a paradoxical rise of cortisol production.150,152 In a recent study of the largest series reported to date,152 all patients with PPNAD responded to the graded administration of dexamethasone during the classic Liddle’s test with a rise in both urinary-free cortisol and 17-hydroxycorticosteroid production. The test may be used diagnostically for the identification of PPNAD, even in patients who have normal baseline cortisol levels and no clinical stigmata of CS. About 10% of patients with CNC have a GH-secreting pituitary adenoma that results in acromegaly. Although most of the known patients with this condition had macroadenomas, a number of recently investigated cases show that abnormal 24-hour GH and prolactin secretion can precede the development of a pituitary tumor in CNC.153,154 The disorder, therefore, provides the unusual opportunity for prospective screening of affected patients without clinical acromegaly. Hyperplasia appears to be present in the pituitary gland of all patients with the complex, acromegaly, and a GH-producing microadenoma, who have undergone surgeries till date.155 Endocrine involvement in CNC also includes three types of testicular tumors: large cell, calcifying Sertoli cell tumor (LCCSCT), adrenocortical rests, and Leydig cell tumor.156
More than three-fourths of affected male patients have one or more of these masses. LCCSCT, as in PJS, may secrete estrogens and cause precocious puberty, gynecomastia, or both.156 It is our clinical impression that LCCSCTs in CNC are more benign than those in PJS, at least as far as aromatization is concerned. Also, unlike PJS, the ovaries in CNC may be affected by cystic formation and occasionally epithelial cancer, but not by germ cell tumors.157 Since 1985, three new components of the syndrome have been identified: psammomatous melanotic schwannoma, epithelioid blue nevus, and ductal adenoma of the breast.158 – 160 Thyroid follicular neoplasms, both benign and malignant, have also been found in a number of patients.161 Molecular Genetics of CNC
Two genetic loci have been determined for CNC by linkage analysis of polymorphic markers from likely areas of the genome.162,163 Initially, positive lod scores were obtained for nine markers on the short arm of chromosome 2, identifying an approximately 4-cm-long area in the cytogenetic band 2p16 (CNC2 locus), which was likely to contain the gene(s) responsible for the complex.162,164 This region was where another genetic syndrome, hereditary nonpolyposis colorectal cancer (HNPCC), had been mapped.162,164 The gene for HNPCC (hMSH2 ) codes for a protein that plays a direct role in DNA mismatch repair, increasing microsatellite stability and enhancing mutation avoidance in human cells – this gene was excluded from being a candidate for the complex.164 Genomic instability in the form of telomeric associations and dicentric chromosomes is a frequent feature of fibroblasts derived from the myxoid and PPNAD tumors excised from patients with CNC.165 – 168 Earlier investigations had found chromosomal instability in cultured skin fibroblasts, peripheral blood lymphocytes, and adenomatous polyp cells established from patients with familial polyposis coli and PJS.169 – 171 In these studies, no specific chromosomal breaks or exchange points were revealed, although several sites were involved in three or more rearrangements. Microsatellite analysis of the tumors excised from patients with CNC confirmed the significant genomic instability that accompanies tumorigenesis in this syndrome.168 One of the first genes to be screened for mutations in CNC, was the gsp proto-oncogene (GNAS1 ); one study showed that gsp mutations were not present in CNC tumors,172 and the locations of several genes that code for components of the guanine nucleotide-binding proteins (G proteins) were excluded by linkage analysis;162 it seemed likely that the gene or genes responsible for CNC participate in G protein–controlled or – related signaling systems, because of the similarities of CNC with MAS.109 In the last 2 years, several families with CNC were described, which did not map to chromosome 2.173,174 Genetic heterogeneity was confirmed in this syndrome when a second locus on 17q22-24 was identified.163 This was followed by the recent identification of the PRKAR1A gene coding for the type Iα regulatory subunit (RIα) of protein kinase A (PKA) as the gene responsible for CNC in most families that mapped to chromosome 17 and some sporadic cases of the disease.99,175 So far, all CNC-responsible
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS
mutations lead to truncation of the RIα protein. The LOH observed at the 17q22-24 locus in CNC tumors suggests that the PRKAR1A gene may function as a tumor-suppressor gene in affected tissues.99,176 A defective cyclic nucleotide-dependent pathway has long been considered a candidate mechanism for the various manifestations of CNC130,172 including tumors similar to those of MAS154,155 and paradoxical responses to hormonal stimuli.152 However, measurements of baseline and poststimulation intracellular cyclic adenosine 3’ : 5’-monophosphate (cAMP) levels in cultured tumors from patients with CNC (Stratakis CA, unpublished) and mutation analysis of the GNAS1 gene172 gave negative results. Thus the defect in CNC was placed downstream from cAMP activation; thus, the PKA complex, a critical step in cAMP-dependent signaling, was a likely candidate for the identification of mutations in patients with CNC. How does one combine the cAMP-signaling defect with the identified LOH and possible tumor suppression function of PRKAR1A? One possibility is that truncated or absent RIα may inhibit the normal formation of the multimeric PKA enzymatic complex leading to dysregulated PKA activity.175 Alternatively, the truncated RIα may interfere with the function of the normal components of the PKA complex in a dominant-negative manner (assuming that normal RIα has primarily an inhibitory role in CNC tissues) or, perhaps, RIα deficiency leads to overexpression of the genes coding for other regulatory subunits of the PKA complex, as has been demonstrated in other contexts.175 Other possible mechanisms include an uninhibited catalytic subunit of the PKA complex, which by itself when mutated leads to unregulated PKA activity, or a reduced turnover of the cAMP molecule because both severe mutations interrupt the cAMPbinding domains of the RIα.99,176 The identification of the gene causing CNC on chromosome 17 left a group of families that appear to map collectively to chromosome 2 (although none of them with lod score over 3), for which the syndrome has not been molecularly elucidated.162,176 There are also families that seem to map neither to chromosome 2 nor to chromosome 17, allowing for a third possible locus harboring gene(s) responsible for CNC. In tumors from patients with CNC, genetic changes of the chromosome 2p16 locus, including both copy number gain and loss, have been identified.177 Interestingly, these changes are shared by both chromosome 2–and chromosome 17–mapping families, indicating, perhaps, a common molecular mechanism.177 Given the substantial clinical overlap between syndromes like Bannayan-Riley-Ruvalcaba (BRR), PJS, and CNC it is not unlikely that some patients, especially sporadic cases of CNC, will end up being diagnosed with one of these overlapping conditions. It is characteristic of the potential diagnostic errors in patients with one of the lentiginosis, that one of the first families identified with CNC and the first one that was found to harbor PRKAR1A mutations, were at first misdiagnosed with PJS.178 It remains to be seen whether functional characterization of the genes responsible for all these disorders will provide molecular basis for the clinical overlap.
785
Clinical and biochemical screening for CNC remains the gold standard for the diagnosis of CNC. Molecular testing for PRKAR1A mutations is not recommended at present for all patients with CNC, but may be advised for detection of affected patients in families with known mutations of that gene to avoid unnecessary medical surveillance of noncarriers.
Peutz-Jeghers Syndrome (PJS) Clinical Features
The syndrome was first described by Peutz179 and Jeghers et al.180 The hallmark of the syndrome is the presence of pigmented spots on the lips, which first present in early childhood. These lesions are associated with gastrointestinal hamartomatous polyps.181 Gastrointestinal cancers are frequent in PJS; they may arise from genetic changes known to occur in colorectal carcinoma and other tumors, because the PJS-responsible gene(s) function as tumor suppressor(s).181,182 PJS patients are also at an increased risk for breast, ovarian, testicular, uterine, and cervical cancers, as well as nonmalignant lesions in these tissues.181,183 – 186 A 49-year follow-up of the “Harrisburg family”, the kindred originally described by Jeghers et al., revealed that PJS is a premalignant condition associated with significant morbidity and increased mortality.187 Among the 12 affected family members, 10 underwent 75 polypectomies, and 2 developed gastric cancer and duodenal carcinoma, respectively. Other investigators found that cancer developed in 15 of 31 patients from 13 unrelated families; the overall incidence of carcinoma in patients with PJS varies from 20 to 50%, and it appears at a relatively early age.183 – 191 Among the nongastrointestinal neoplasms associated with PJS, endocrine tumors are the most frequent.183 – 186,189 These include thyroid nodules and cancer189 and genital tract neoplasms.191,192 The latter include, in female patients with PJS, ovarian neoplasms from both the epithelium and stromal cells, and also adenoma malignum of the cervix and adenocarcinoma of the endometrium. Male patients with PJS often have Leydig cell tumors, or a Sertoli cell tumor that is uniquely found in PJS and CNC156 –LCCSCT.193 LCCSCT in PJS, like in CNC, may be associated with increased aromatization of adrenal or testicular androgens, which produces estradiol and other estrogens (estrone, in particular) that may lead to precocious puberty, prepubertal or peripubertal gynecomastia.193 Molecular Genetics
One of the first reports on the genetics of PJS, was that of a patient with PJS and a pericentric inversion of chromosome 6.194 However, other patients with similar chromosomal changes do not manifest symptoms of the disease. More recently, three families with PJS were studied by linkage analysis and were mapped to the short arm of chromosome 1 (1p);195 however, the finding ended up being spurious (most likely because of the small number of patients investigated) and was not confirmed in a more recent investigation.196 A study of comparative genomic hybridization (CGH) and LOH analysis defined a susceptibility locus for PJS on
786
PEDIATRIC MALIGNANCIES
19p, around the microsatellite repeat marker D19S886.197 LOH of 19p13 markers suggested that the gene functions as a tumor-suppressor gene. Subsequently, most PJS families were linked to 19p13.3,197 although some mapped to a second locus at 19q13.4.198 The PJS gene at 19p13.3, STK11 (for serine-threonine kinase 11) or LKB1 was then identified by two research groups.100,101 More than half of the families with PJS have mutations in this gene, although the percentage varies greatly from one study to the other.199 – 204 STK11/LKB1 is a novel serine –threonine kinase containing 9 exons; it shows high homology, almost 84%, with XEEK1 (named for Xenopus egg and embryo kinase 1), a Xenopus cytosolic serine –threonine protein kinase.205 A mouse homolog was also recently identified.206 The kinase domain of the STK11/LKB1 gene is highly conserved between mouse and human. Although it has been suggested that mouse Lkb1 is a nuclear protein, wild-type STK11/LKB1 shows both nuclear and cytoplasmic localization. A number of recent studies have elucidated the effects of inherited STK11/LKB1 mutations in PJS kindred, based on the functional domains of the protein; in most cases, elimination of the kinase activity underlies the molecular cause of the phenotype.207 Genetic heterogeneity in PJS appears to be not accompanied by clinical heterogeneity, as there are no known differences between the families that map to 19p13.3 and have STK11/LKB1 mutations and those that map elsewhere or do not have mutations in that gene.198,207
Cowden Disease (CD) CD is associated with hamartomas and tumors of ecto-, meso-, and endodermal origin affecting multiple organs.102,208,209 Thyroid and breast masses, hamartomatous polyps, and mucocutaneous lesions (oral papillomatosis, acral keratosis, and multiple fibromas) occur consistently in this syndrome, which are also associated with anomalies of the skeletal and nervous systems.209 CD was first mapped to chromosome 10;208 mutations in the tumor-suppressor gene PTEN on 10q were found shortly after that.210 – 212 CD is allelic to Ruvalcaba-Myhre-Smith, BannayanZonana, or BRR syndrome which is also associated with hamartomatous intestinal polyposis, lentiginosis of the genitalia, developmental defects (macrocephaly, eye, and skeletal anomalies), and myopathy.209 Both syndromes are caused by mutations of the PTEN gene on chromosome 10,210 – 212 although genetic heterogeneity may exist.
Von Hippel–Lindau Disease (VHLD) VHLD, with an estimated birth incidence of 1 in 36 000 live births, is a relatively frequent multiple tumor syndrome that affects the pancreas (pancreatic cysts) and the adrenal medulla (pheochromocytoma).213 However, the main components of the syndrome are nonendocrine and include renal carcinoma and cysts, and tumors of the cerebellum, spinal cord, eyes, and epididymis.213 The VHL gene has tumor-suppressor functions and is located on chromosome 3p.103,213,214 Sporadic pheochromocytomas may harbor VHL
mutations, although both RET and VHL gene defects are rare in the sporadic form of this tumor.215
Pediatric Thyroid Nodules and Cancer Carcinoma in pediatric patients (<21-years old) is an uncommon event. Of the 1050 reported childhood and adolescent carcinoma cases per year, 35.5% are thyroid carcinomas, the majority of cases presenting in the adolescent female population.216 The female:male ratios are age-specific with a 1 : 6 (F:M) ratio for ages 5–9 years, a 1 : 1 (F:M) ratio for ages 10–14, and a 5 : 2 (F:M) ratio for ages 15–19, with the latter being the peak incidence of disease.216 In children, the most common presentation is an asymptomatic nodule noted on physical exam, estimated to be found in 1.5% of children.217 Unfortunately, while nodules are less-common in children, the risk of malignant disease is much higher, with 20–50% of pediatric nodules found to have carcinoma compared to 10–14% of that in adults.218 In the subgroup of children with a history of previous radiation exposure (environmental accidents or therapeutic), the incidence of thyroid nodules and malignancy is even greater, with relative risks reported as high as 27 and 53 times respectively, when compared to controls.219,220 Differentiated thyroid carcinoma (DTC) and its papillary (PTC) and follicular (FTC) variants are the most common forms of thyroid cancer in both children and adults, with PTC comprising 70–80% of these tumors.221 – 223 The histology of these tumors is the same; however, the clinical behavior and molecular signature of these carcinomas differ considerably. Clinically, compared to adults, children with thyroid cancer are more likely to have invasive disease at diagnosis with up to 90% having regional lymph node metastasis221 – 234 and up to 20% having lung metastasis.222 – 224,226,235 – 237 In addition, current data suggests that despite aggressive surgical and radioiodine therapy, up to 25% of cancers will recur, at times many years after apparently negative studies.221,223 – 227,237 Elucidation of the molecular mechanisms involved in pediatric thyroid cancers has also revealed differences when compared to adults. In adults, BRAF and ras mutations are commonly found and ret/PTC 1 mutations, when found, are often associated with more-aggressive disease.238 – 244 In children, ret/PTC are not uncommon, with ret/ PTC 1 mutations more commonly found in sporadic tumors and ret/PTC 3 mutations more commonly found in radiationinduced PTC.240,245 The inherited forms of pediatric thyroid cancer comprise approximately 5% of reported PTC cases246 and include familial nonmedullary thyroid carcinoma,247,248 Cowden syndrome (AD, autosomal dominant; multiple hamartomas, breast cancer, colon cancer; PTEN gene),249 – 251 Gardner syndrome (AD inheritance; familial adenomatoid polyposis, desmoid tumors, lipomas; APC gene)252,253 and CNC (AD inheritance; lentiginosis, pituitary adenomas, PPNAD, testicular tumors; PRKAR1A and CNC2 genes).254 The involvement of PRKAR1A as a tumor suppressor in thyroid tumorigenesis has been explored more stringently. In an animal model, 11% of PRKAR1A/+ mice develop thyroid neoplasms255 and PRKAR1A mutations have also been found in sporadic thyroid cancer samples.256 To date, however, no
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS
unifying molecular mechanism has been found to explain the differences in the clinical behavior of the pediatric thyroid cancer forms. Evaluation of a child with a thyroid nodule should include thyroid ultrasonography and fine-needle aspiration biopsy (FNAB). FNA, a technique that has proven efficacy in the adult population, has been used increasingly in children. The quality, sensitivity, and specificity of the technique varies between institutes and if the FNA results are equivocal, repeat FNA or surgery should be considered because of the increased risk of malignancy.218,257,258 Routine ordering of thyroid function testing and thyroid scans are not warranted as the results of either test cannot be relied upon to rule in or rule out malignancy.259 – 262 Chest X ray and chest CT scans are relatively insensitive and are often normal despite the presence of pulmonary disease detected on postablation or posttherapy 131 I scanning.229,236,263,264 Treatment of thyroid cancer in children is similar to that of adults, with surgical resection followed by radioiodine therapy. Controversy exists concerning the extent of surgery;221 – 225,235,237,265 – 276 however, in the hands of an experienced pediatric thyroidectomy surgeon, near-total or total thyroidectomy with removal of local lymph nodes is associated with a low risk of complications266,267 and decreased risk of persistent or recurrent disease.222,224 – 226,233,235 In addition, removal of the entire thyroid and suspicious lymph nodes provides distinct advantages for follow-up, most importantly allowing for the most efficient use of radioiodine therapy and improving the sensitivity of detecting residual or persistent disease on posttreatment scans.222 – 224,226,235 All children, irrespective of the extent of surgery, are placed on suppressive doses of thyroid hormone, with a goal of maintaining TSH at 0.01–0.02µU mL−1 .277 Release from suppressive doses of thyroid hormone to physiologic replacement is individualized, based on when the patient appears to be in remission. For a more thorough discussion on treatment and follow-up please see the excellent reviews of Hung and Sarlis277 or refer to the pediatric specific chapters in Thyroid Cancer; A Comprehensive Guide to Clinical Management, edited by Wartofsky.278 Prognosis for children diagnosed with differentiated thyroid cancer (PTC or FTC) is good, with low disease-specific mortality.228 – 234,279 – 282 However, prospective studies to individualize therapy based on patient age, extent of surgery, and the use of 131 I do not exist. In addition, there is a paucity of information on children followed up for 20 years or more posttreatment. Age less than 10 years, male gender, and tumor size have all been examined for prognostic significance with mixed results.18,24,225 – 227 With these limitations in mind, and the realization that thyroid cancer in children often presents with locally invasive disease and distant metastasis, total thyroidectomy with ablation appears to be the most prudent approach to initial treatment.221,222,226,235,237 With this approach a 10-fold increase in disease-free survival with total versus nontotal thyroidectomy and a fivefold reduction in risk of recurrence with the use of 131 I ablation has been reported.235 Even with aggressive initial therapy 15–35% of patients will experience recurrence of disease,
787
often 10 or more years after diagnosis, with pregnancy appearing to be a time of increased risk.221,222,226,235,237,283 With these issues in mind, differentiated thyroid cancers in children require life-long monitoring. In the future, it is hoped that individualized recommendations regarding treatment may evolve as further characterization of molecular markers are discovered and examined in regard to tumor behavior.
Pheochromocytoma Definition of Pheochromocytoma
Pheochromocytomas are chromaffin cell tumors that produce, store, metabolize, and secrete catecholamines. The metabolism of catecholamines is a more consistent process than that of catecholamine secretion.284,285 Clinical Presentation of Pheochromocytoma
Pheochromocytoma has been labeled ‘the great mimic’ since there are numerous reports in the literature of unusual presentations of benign or metastatic pheochromocytomas that require emergency intervention. Emergency situations listed below are either the result of organ-specific actions of catecholamines produced in high quantity by the tumor or the consequence of complications related to tumor localization (see Table 6). Paroxysmal hypertension represents a frequent clinical dilemma, particularly when these bouts are of abrupt onset and severe (blood pressure>200/110). Although severe paroxysmal hypertension should always raise suspicions of pheochromocytoma, it can also reflect a clinical entity called pseudopheochromocytoma. Pseudopheochromocytoma refers to the disease in a large majority of individuals (most often women) with severe paroxysmal hypertension, whether normotensive or hypertensive between episodes, in whom pheochromocytoma has been ruled out.288,289 Recent evidence indicated that pseudopheochromocytoma is a heterogeneous clinical condition divided into a primary and a secondary form. In contrast to the primary form, the secondary form is associated with various pathologies (e.g. hypoglycemia, epilepsy, baroreceptor failure), medications, or drug abuse (see Table 7). The most common clinical characteristics of this syndrome might in many cases be attributable to a short-term activation of the sympathetic nervous system. Paroxysmal hypertension is usually associated with tachycardia, palpitations, nervousness, tremor, weakness, excessive sweating, pounding headache, feeling hot, facial paleness, or rarely redness. In contrast to patients with pheochromocytoma, those with pseudopheochromocytoma more often present with panic attacks or anxiety, flushing, nausea, and polyuria. Another important feature distinct from pheochromocytoma is the circumstances under which the episode occurs. In pheochromocytoma, symptoms are usually unprovoked, while in pseudopheochromocytoma they usually follow some identifiable events. It is important, therefore, in questioning these patients, to search for specific provocative factors that may have precipitated these episodes. Similar to pheochromocytoma, episodes may last from few minutes to several hours and may occur daily or once every few months.
788
PEDIATRIC MALIGNANCIES
Table 6 Emergency situations related to catecholamine excess released from pheochromocytoma.
Clinical setting
Symptoms
Pheochromocytoma multisystem crisis (PMC)
Hypertension and/or hypotension, multiple organ failure, temperature = 40 ◦ C, encephalopathy Collapse Hypertensive crisis ∼ upon induction of anesthesia ∼ medication induced or other mechanisms Shock or profound hypotension Acute heart failure Myocardial infarction Arrhythmia Cardiomyopathy Myocarditis Dissecting aortic aneurysm Limb ischemia, digital necrosis, or gangrene Acute pulmonary edema Adult respiratory distress syndrome Abdominal bleeding Paralytic ileus Acute intestinal obstruction Severe enterocolitis and peritonitis Colon perforation Bowel ischemia, plus generalized peritonitis Mesenteric vascular occlusion Acute pancreatitis Cholecystitis Megacolon Hemiplegia Limb weakness Acute renal failure Acute pyelonephritis Severe hematuria Diabetic ketoacidosis Lactic acidosis
Cardiovascular
Pulmonary Abdominal
Neurological Renal
Metabolic Adapted from Brouwer et al.286
Table 7 Differential diagnosis of pheochromocytoma.
Endocrine
– adrenal medullary hyperplasia – hyperthyroidism, thyroid storm – carcinoid – hypoglycemia (often due to the presence of insulinoma) – medullary thyroid carcinoma – mastocytosis – menopausal syndrome Cardiovascular – heart failure – arrhythmias – ischemic heart disease, angina pectoris – baroreflex failure – syncope – orthostatic hypotension – labile hypernoradrenergic essential hypertension – renovascular disease Neurologic – migraine or cluster headaches – stroke – diencephalic autonomic epilepsy – meningioma – postural orthostatic tachycardia syndrome (POTS) – Guillain-Barre syndrome – encephalitis Psychogenic – anxiety or panic attacks – factitious use of drugs – somatization disorder – hyperventilation Pharmacologic – tricyclic antidepressant – cocaine – alcohol withdrawal – drugs stimulating adrenergic receptors – abrupt clonidine withdrawal – dopamine antagonists – monoamine oxidase inhibitors – ephedrine-containing drugs – factitious use of various drugs including catecholamines Miscellaneous – neuroblastoma, ganglioneuroma, ganglioneuroblastoma – acute intermittent porphyria – mastocytosis – unexplained flushing spells – recurrent idiopathic anaphylaxis – lead and mercury poisoning Adapted from Lenders et al.287
Between episodes, blood pressure is normal or may be mildly elevated. Pseudopheochromocytoma is usually successfully treatable with antihypertensive drugs or psychotherapy. Incidence of Familial Pheochromocytoma
Pheochromocytomas may occur sporadically or as part of a hereditary syndrome. According to the latest studies, in patients with nonsyndromic pheochromocytoma, up to 24% of tumors may be inherited.290 – 292 Hereditary pheochromocytoma is associated with multiple endocrine neoplasia type 2 (MEN 2A or MEN 2B), neurofibromatosis type 1 (NF1), VHL syndrome, and familial paragangliomas and pheochromocytomas because of germline mutations of genes encoding succinate dehydrogenase (SDH) subunits B, C, and D (see Table 8). In general, the traits are inherited in an autosomaldominant pattern.290 Patients with MEN 2-related pheochromocytoma often lack hypertension or symptoms (occurs only in about 50%).293 MEN 2-related pheochromocytomas are characterized by production of epinephrine and norepinephrine and are therefore best detected by elevated levels of plasma or urinary metanephrine, usually but not always in association
with elevated levels of normetanephrine and parent catecholamines. MEN 2-related pheochromocytomas are almost always intra-adrenal, often bilateral and they are rarely malignant (<5%).294,295 In addition, as with most epinephrinesecreting pheochromocytomas, hypertension when present is more likely to be paroxysmal than sustained. For these reasons, it is easy to miss the diagnosis. Overall, less than 30% of patients with VHL germline mutation develop a pheochromocytoma.296,297 Pheochromocytomas as part of the VHL syndrome have an exclusively noradrenergic phenotype, reflecting lack of production of epinephrine. Biochemical diagnosis is therefore best achieved from elevations of plasma or urinary normetanephrine. These tumors are mainly located intra-adrenally and are bilateral in about 50% of patients with a less than 5% incidence of metastases. Since pheochromocytomas in VHL patients do not express glucagon receptors, glucagon tests are unlikely to be useful for diagnosis. These tumors are commonly found during periodic annual screening or during searches for other tumors that are part of this syndrome. Therefore, these
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS
789
Table 8 Familial pheochromocytoma.
Syndrome
Genetic abnormalities
Phenotype
Multiple endocrine neoplasia syndromes (see Table 5 ) Neuroectodermal syndromes Neurofibromatosis (von Recklinghausen’s disease) type 1 (NF1) Cerebelloretinal hemangioblastomatosis (von Hippel – Lindau syndrome, VHL) type II
Chromosome 17 (17q11) mutations affect NF1, tumor-suppressor gene Chromosome 3 (3p25-26) missense mutations affect VHL, tumor-suppressor gene
Peripheral neurofibromas Retinal angiomas
Cerebellar and spinal cord hemangioblastomas Renal cell cancer Pancreatic, renal, epididymal, and endolymphatic cysts/tumors Succinate dehydrogenase gene family syndromes SDHB (Paraganglioma (PG)/ type IV)
Chromosome 1 (1p36) missense, nonsense, extra- and frameshift mutations
SDHC PG type III
Chromosome 1 (1q21)
SDHD PG type I
Chromosome 11 (11q23) missense, nonsense, and frameshift mutations; paternal transmission
tumors are commonly small and often fail to be detected by nuclear imaging methods. Furthermore, about 80% of pheochromocytomas found in VHL patients during screening are asymptomatic and not associated with hypertension. Genetics of Pheochromocytoma
Recently, pheochromocytoma susceptibility has been associated with germline mutations of the SDH gene family.291,292,298 The SDH genes (SDHA, SDHB, SDHC, and SDHD) encode the four subunits of complex II of the mitochondrial electron transport chain.299,300 Intra-adrenal pheochromocytomas and extra-adrenal catecholamine-producing paragangliomas are associated with mutations of SDHD and SDHB genes. Nonfunctional paragangliomas that do not produce, metabolize, or secrete catecholamines and that appear to arise from parasympathetic tissue may also occur secondary to mutations of SDHB, SDHD, or rarely, SDHC genes. Mutations of the SDHA gene do not appear to cause pheochromocytomas or paragangliomas. In recent studies, it has been found that about 4–12% of apparently sporadic pheochromocytomas and up to 50% of familial pheochromocytomas have either SDHD or SDHB mutation.301 Currently, there is a strong association of SDHB mutations with the presence of extra-adrenal multifocal or metastatic pheochromocytoma.290,302 – 304 Recently, it has been suggested that all patients younger than 50-years with either solitary extra-adrenal or multifocal pheochromocytoma (not initially found in the adrenal gland) should undergo genetic testing to search for mutations of SDH genes.290,291 The biochemical phenotype of pheochromocytomas in patients with SDH mutations is currently unknown. It has also been suggested that these patients have periodic follow-up including yearly measurement of plasma free metanephrines.290
Parasympathetic paraganglioma, adrenal, often metastatic pheochromocytoma Parasympathetic paraganglioma, extra-adrenal pheochromocytoma Parasympathetic paraganglioma, extra-adrenal pheochromocytoma
Metastatic Pheochromocytoma
Depending on the genetic background and tumor localization, the incidence reported for malignant pheochromocytoma ranges from 3 to 36% of pheochromocytoma patients.305 Patients with SDHB mutations have a particularly high incidence of metastatic pheochromocytoma.304 Malignant disease has an overall 5-year survival rate of 50 %.305,306 Patients with malignant pheochromocytoma, particularly those with metastases occurring secondary to paragangliomas, more often present with dopamine-producing tumors than do patients with solitary benign neoplasms.310 In some of these patients, specific imaging studies (e.g. 6[18 F]fluorodopamine PET or [123/131 I]metaiodobenzylguanidine, MIBG) are negative. Biochemical Diagnosis of Pheochromocytoma
The diagnosis of a pheochromocytoma depends on the biochemical evidence of excessive catecholamine (norepinephrine and epinephrine) production by the tumor. Such evidence is best obtained during initial testing by measurements of urinary fractionated metanephrines or plasma free metanephrines.285,307 Plasma free metanephrines have higher specificity than urinary fractionated metanephrines and may offer the preferred test, but this also depends on the proficiency of the testing laboratory in providing accurate results, free of analytical interference. High sensitivity indicates a low rate of false-negative results; high specificity indicates a low rate of false-positive results. Since missing the diagnosis of pheochromocytoma (due to low diagnostic sensitivity) can have catastrophic consequences, we recommend measurements of plasma free metanephrines as the initial biochemical test to confirm or rule out pheochromocytoma, with measurements of urinary fractionated metanephrines providing the next best test. Measurements of plasma free metanephrines
790
PEDIATRIC MALIGNANCIES
for plasma free metanephrines; this is in contrast to conjugated metanephrines that are cleared by the kidneys and show large increases associated with renal failure.309
can also help predict tumor size and location.308 This additional information may be useful for guiding diagnostic decision-making before and after tumor localization procedures. The conditions under which blood or urine samples are collected can be crucial to the reliability and interpretation of the test results (see Table 9). Blood samples for measurements of plasma free metanephrines or catecholamines should be collected with patients supine for at least 15–20 minutes before sampling. To avoid any stress associated with the needle stick, samples should ideally be collected through a previously inserted i.v. Patients should have refrained from nicotine and alcohol for at least 12 hours, and to minimize analytical interference should have fasted overnight before blood sampling. Often there is also a need for patients to avoid acetaminophen for at least 3–5 days before sampling, but this requirement depends on the laboratory method used for measurements of plasma free metanephrines. Tricyclic antidepressants and phenoxybenzamine increase plasma and urinary norepinephrine and normetanephrine levels and represent the most common causes of medication-associated false-positive results in patients tested for pheochromocytoma.
Dopamine-secreting Pheochromocytomas Although measurements of plasma free normetanephrine and metanephrine provide a sensitive test for diagnosis of pheochromocytoma, these measurements may fail to detect tumors that produce predominantly dopamine (patients with these tumors usually do not present with any cardiovascular symptoms that are normally seen in tumors secreting epinephrine or norepinephrine). Such tumors are usually very rare and they are found extra-adrenally. In such patients, measurements of plasma free methoxytyramine (metabolite of dopamine) or dopamine can be used to detect tumors; in contrast, measurements of urinary dopamine are much less reliable because of the urinary amine mainly derived from circulating dihydroxyphenylalanine, and not dopamine.310 Update on Imaging of Pheochromocytoma
The imaging test of choice is CT. Using of this imaging method localizes about 95% of all pheochromocytomas. MRI is also a very reliable imaging method that identifies over 95% of pheochromocytomas. This imaging technique is sometimes preferred for the localization of extra-adrenal pheochromocytomas, especially for
Renal Failure In patients with renal failure, pheochromocytoma can be reliably excluded on the basis of normal values
Table 9 Drugs that may cause false-positive elevations of plasma and urinary catecholamines or metanephrines.
Catecholamines
Tricyclic antidepressants Amitriptyline (Elavil), Imipramine (Topfranil), Nortriptyline (Aventyl) α-Blockers (nonselective) Phenoxybenzamine (Dibenzyline) α-Blockers (α1 -selective) Doxazosin (Cardura), Terazosin (Hytrin), Prazosin (Minipress) ß-Blockers Atenolol (Tenormin), Metoprolol (Lopressor), Propranolol (Inderal), Labetolol (Normadyne)a Calcium channel antagonists Nifedipine (Procardia), Amlodipine (Norvasc), Vasodilators Hydralazine (Apresoline), Isosorbide (Isordil, Dilatrate), Minoxidil (Loniten) Monoamine oxidase inhibitors Phenelzine (Nardil), tranylcypromine (Parnate), Selegiline (Eldepryl) Sympathomimetics Ephedrine, Pseudoephedrine (Sudafed), Amphetamines, Albuterol (Proventil) Stimulants Caffeine (coffeea , tea), Nicotine(tobacco), Theophylline Miscellaneous Levodopa, Carbidopa (Sinemet)a Cocaine
Metanephrines
NE
E
NMN
MN
+++
−
+++
−
+++
−
+++
−
+
−
−
−
+
+
+
+
+
+
−
−
+
−
−
−
+++
+++
++
++
++
++
++
++
Unknown
++ ++
− ++
Unknown Unknown
Unknown
Note: NE, norepinephrine; E, epinephrine; NMN, normetanephrine; MN, metanephrine. + + +, substantial increase; ++, moderate increase; +, mild increase if any;, little or no increase. a Indicates a drug that can also cause direct analytical interference with some methods.
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS
those in unusual locations. MRI should also be the initial imaging method of choice in children, pregnant women, patients with a documented allergy to contrast dye, or in situations where no additional radiation exposure is desired.311 There is some debate about whether functional imaging should be used as a second step after CT or MRI to confirm that the tumor is indeed a pheochromocytoma. Several important considerations impact the choice of additional functional imaging studies. First, although this tumor is most often localized in the adrenal gland, the adrenal gland is also the site of many benign adrenal tumors (adenomas); in the general population, between 5 and 10% may be expected to have such masses, this dependent on the age. Second, about 50% of adrenal pheochromocytomas produce near exclusively norepinephrine, this representing the same pattern as in extra-adrenal pheochromocytomas. Thus, while production of epinephrine (best detected by an increase in metanephrine) indicates an adrenal location, exclusive production of norepinephrine (best indicated by increases of normetanephrine with normal metanephrine) may reflect either an adrenal or extra-adrenal location. Third, about 10% of patients have metastatic pheochromocytoma at initial diagnosis; those with primary tumors larger than 5 cm are at particular risk. Fourth, in patients with previous surgeries (especially in the abdomen) the presence of postsurgical tissue changes (e.g. tissue fibrosis, adhesions) and surgical clips often precludes correct localization of recurrent or metastatic pheochromocytomas using CT or MRI. Fifth, up to 24% of pheochromocytomas are familial and these tumors are often multiple. On the basis of the above, we advise additional use of functional imaging studies for localization of most cases of biochemically proven pheochromocytoma. Exceptions may include small (less than 5 cm) adrenal masses associated with elevations of plasma or urine metanephrine levels (practically all epinephrine-producing pheochromocytomas are found in the adrenal gland or are recurrences of previously resected adrenal tumors). The functional imaging test of choice is [123 I]MIBG. A new method, PET using 6-[18 F]fluorodopamine, has been recently introduced.312 This method offers an excellent visualization of primary and metastatic pheochromocytoma and is also superior to [123 I]MIBG and Octreoscan, especially for the localization of metastatic pheochromocytoma or those tumors that are very difficult to localize (e.g. in patients with previous surgical procedure).313 In patients with rapidly progressing metastatic pheochromocytoma, MIBG and 6-[18 F]fluorodopamine functional imaging studies can sometimes be negative, and in these situations [18 F]fluorodeoxyglucose or Octreoscan may be useful.314 Otherwise these two methods should not be used for initial functional imaging.
ACKNOWLEDGEMENTS This work was supported in part by grants P30 CA-21765 and PO1 CA-20180 from the National Cancer Institute and by the American Lebanese Syrian Associated Charities (ALSAC).
791
REFERENCES 1. Pitman. General melasma and short hair over the entire body of a child of three years, with conversion of the left supra-renal capsule into a large malignant tumor; the external organs of generation resembling that of adult life. Lancet 1865; 1: 175. 2. Walters W, Wilder R, Kepler E. The suprarenal cortical syndrome with presentation of ten cases. Ann Surg 1934; 100: 670. 3. Ribeiro RC, et al. Adrenocortical carcinoma in children: a study of 40 cases. J Clin Oncol 1990; 8: 67 – 74. 4. Michalkiewicz E, et al. Clinical and outcome characteristics of children with adrenocortical tumors: a report from the International Pediatric Adrenocortical Tumor Registry. J Clin Oncol 2004; 22: 838 – 45. 5. Bernstein L, Gurney JG. Carcinomas and other malignant epithelial neoplasms. In Ries LAG, et al. (eds) Cancer incidence and survival among children and adolescents: United States SEER program 1975 – 1995. Bethesda, Maryland: National Cancer Institute, SEER Program, 1999: 139 – 147. 6. Drut R, Hernandez A, Pollono D. Incidence of childhood cancer in La Plata, Argentina, 1977 – 1987. Int J Cancer 1990; 45: 1045 – 7. 7. Gatta G, et al. Childhood cancer survival trends in Europe: a EUROCARE Working Group study. J Clin Oncol 2005; 23: 3742 – 51. 8. Young JL Jr, et al. Cancer incidence, survival, and mortality for children younger than age 15 years. Cancer 1986; 58: 598 – 602. 9. Stiller CA. International variations in the incidence of childhood carcinomas. Cancer Epidemiol Biomarkers Prev 1994; 3: 305 – 10. 10. Pereira RM, et al. Childhood adrenocortical tumors. Arq Bras Endocrinol Metabol 2004; 48: 651 – 8. 11. Latronico AC, Mendonca BB. Adrenocortical tumors – new perspectives. Arq Bras Endocrinol Metabol 2004; 48: 642 – 6. 12. Ribeiro RC, et al. An inherited p53 mutation that contributes in a tissue-specific manner to pediatric adrenal cortical carcinoma. Proc Natl Acad Sci USA 2001; 98: 9330 – 5. 13. DiGiammarino EL, et al. A novel mechanism of tumorigenesis involving pH-dependent destabilization of a mutant p53 tetramer. Nat Struct Biol 2002; 9: 12 – 6. 14. Varley JM, et al. Are there low-penetrance TP53 Alleles? Evidence from childhood adrenocortical tumors. Am J Hum Genet 1999; 65: 995 – 1006. 15. Figueiredo BC, et al. Penetrance of adrenocortical tumors associated with the germline TP53 R337H mutation. J Med Genet 2006; 43(1): 91 – 6; [Epub ahead of print]. 16. Figueiredo BC, et al. Comparative genomic hybridization analysis of adrenocortical tumors of childhood. J Clin Endocrinol Metab 1999; 84: 1116 – 21. 17. Figueiredo BC, et al. Amplification of the steroidogenic factor 1 gene in childhood adrenocortical tumors. J Clin Endocrinol Metab 2005; 90: 615 – 9. 18. Longui CA, et al. Inhibin alpha-subunit (INHA) gene and locus changes in paediatric adrenocortical tumours from TP53 R337H mutation heterozygote carriers. J Med Genet 2004; 41: 354 – 9. 19. Hirose PF, et al. Adrenocortical tumors in children: the Brazilian experience in a single institution. J Clin Oncol 2005; 23: A8506. 20. Sandrini F, et al. Inheritance of R337H TP53 gene mutation in children with sporadic adrenocortical tumor. Horm Metab Res 2005; 37: 231 – 5. 21. Barreto JHS, D´orea MDF, Mendon¸ca N. Carcinoma do c´ortex adrenal na infˆancia. Arq Bras Endocrinol Metabol 2001; 45: S383. 22. Pianovski MAD, et al. Mortality rate of adrenocortical tumors in children under 15 years of age in Curitiba, Brazil. Pediatr Blood Cancer 2005; Sep 30; [Epub ahead of print]. 23. Coulter CL. Fetal adrenal development: insight gained from adrenal tumors. Trends Endocrinol Metab 2005; 16: 235 – 42. 24. Buster JE. Fetal adrenal cortex. Clin Obstet Gynecol 1980; 23: 803 – 24. 25. Li FP, Fraumeni JF Jr. Rhabdomyosarcoma in children: epidemiologic study and identification of a familial cancer syndrome. J Natl Cancer Inst 1969; 43: 1365 – 73. 26. Li FP, Fraumeni JF Jr. Prospective study of a family cancer syndrome. JAMA 1982; 247: 2692 – 4.
792
PEDIATRIC MALIGNANCIES
27. Garber JE, et al. Follow-up study of twenty-four families with LiFraumeni syndrome. Cancer Res 1991; 51: 6094 – 7. 28. Malkin D, et al. Germline p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science 1990; 250: 1233 – 8. 29. Ribeiro RC, Figueiredo B. Childhood adrenocortical tumours. Eur J Cancer 2004; 40: 1117 – 26. 30. Weksberg R, Shuman C, Smith AC. Beckwith-Wiedemann syndrome. Am J Med Genet C Semin Med Genet 2005; 137: 12 – 23. 31. Koufos A, et al. Familial Wiedemann-Beckwith syndrome and a second Wilms tumor locus both map to 11p15.5. Am J Hum Genet 1989; 44: 711. 32. Ping AJ, et al. Genetic linkage of Beckwith-Wiedemann syndrome to 11p15. Am J Hum Genet 1989; 44: 720. 33. Lustig O, et al. Expression of the imprinted gene H19 in the human fetus. Mol Reprod Dev 1994; 38: 239 – 46. 34. Liu J, et al. H19 and insulin-like growth factor-II gene expression in adrenal tumors and cultured adrenal cells. J Clin Endocrinol Metab 1995; 80: 492 – 6. 35. Voutilainen R, et al. Parallel regulation of parentally imprinted H19 and insulin-like growth factor-II genes in cultured human fetal adrenal cells. Endocrinology 1994; 134: 2051 – 6. 36. Prawitt D, et al. Microdeletion and IGF2 loss of imprinting in a cascade causing Beckwith-Wiedemann syndrome with Wilms’ tumor. Nat Genet 2005; 37: 785 – 6. 37. Yano T, et al. Genetic changes in human adrenocortical carcinomas. J Natl Cancer Inst 1989; 81: 518. 38. Limon J, et al. Translocation t(4;11)(q35;p13) in and adrenocortical carcinoma. Cancer Genet Cytogenet 1987; 28: 343. 39. Henry I, et al. Molecular definition of the 11p15.5 region involved in Beckwith-Wiedemann syndrome and probably in predisposition to adrenocortical carcinoma. Hum Genet 1989; 81: 273 – 7. 40. Herrmann ME, et al. Chromosomal aberrations in two adrenocortical tumors, one with a rearrangement at 11p15. Cancer Genet Cytogenet 1994; 75: 111 – 6. 41. Benson RF, Vulliamy DG, Taubman JO. Congenital hemihypertrophy and malignancy. Lancet 1963; 1: 468 – 9. 42. Fraumeni JF Jr, Miller RW. Adrenocortical neoplasms with hemihypertrophy, brain tumors, and other disorders. J Pediatr 1967; 70: 129 – 38. 43. Muller S, et al. Wilms’ tumor and adrenocortical carcinoma with hemihypertrophy and hamartomas. Eur J Pediatr 1978; 127: 219 – 26. 44. Tank ES, Kay R. Neoplasms associated with hemihypertrophy, Beckwith-Wiedemann syndrome and aniridia. J Urol 1980; 124: 266. 45. van Seters AP, et al. Adrenocortical tumour in untreated congenital adrenocortical hyperplasia associated with inadequate ACTH suppressibility. Clin Endocrinol (Oxf) 1981; 14: 325 – 34. 46. Shinohara N, et al. Adrenocortical adenoma associated with congenital adrenal hyperplasia. Nippon Hinyokika Gakkai Zasshi 1986; 77: 1519 – 23. 47. Varan A, et al. Adrenocortical carcinoma associated with adrenogenital syndrome in a child. Med Pediatr Oncol 2000; 35: 88 – 90. 48. Daeschner GL. Adrenal cortical adenoma arising in a girl with congenital adrenogenital syndrome. Pediatrics 1965; 36: 140. 49. Pang S, et al. Adrenocortical tumor in a patient with congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Pediatrics 1981; 68: 242. 50. Carney JA, et al. Dominant inheritance of the complex of myxomas, spotty pigmentation, and endocrine overactivity. Mayo Clin Proc 1986; 61: 165 – 72. 51. Stratakis CA, et al. Carney complex, a familial multiple neoplasia and lentiginosis syndrome. Analysis of 11 kindreds and linkage to the short arm of chromosome 2. J Clin Invest 1996; 97: 699 – 705. 52. Groussin L, et al. Mutations of the PRKAR1A gene in Cushing’s syndrome due to sporadic primary pigmented nodular adrenocortical disease. J Clin Endocrinol Metab 2002; 87: 4324 – 9. 53. Skogseid B, et al. Clinical and genetic features of adrenocortical lesions in multiple endocrine neoplasia type 1. J Clin Endocrinol Metab 1992; 75: 76 – 81. 54. Chandrasekharappa SC, et al. Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 1997; 276: 404 – 7.
55. Langer P, et al. Adrenal involvement in multiple endocrine neoplasia type 1. World J Surg 2002; 26: 891 – 6. 56. Mann JR, et al. Transplacental carcinogenesis (adrenocortical carcinoma) associated with hydroxyprogesterone hexanoate. Lancet 1983; 2: 580. 57. Hornstein L, Crowe C, Gruppo R. Adrenal carcinoma in child with history of fetal alcohol syndrome. Lancet 1977; 2: 1292 – 3. 58. Dedov VI, Norets TA. State of the endocrine system in the progeny of female rats treatment with 75 Se-selenomethionine. Radiobiologiia 1982; 22: 409 – 12. 59. Kafrouni G, et al. Aldosteronoma in a child with localisation by adrenal vein aldosterone: collective review of the literature. J Pediatr Surg 1975; 10: 917 – 24. 60. Lee PDK, Winter RJ, Green OC. Virilizing adrenocortical tumors in childhood. Eight cases and a review of the literature. Pediatrics 1985; 76: 437 – 44. 61. Lack EE, et al. Adrenal cortical neoplasms in the pediatric and adolescent age group. Clinicopathologic study of 30 cases with emphasis on epidemiological and prognostic factors. Pathol Annu 1992; 27(Pt 1): 1 – 53. 62. Sandrini R, Ribeiro RC, DeLacerda L. Childhood adrenocortical tumors. J Clin Endocrinol Metab 1997; 82: 2027 – 31. 63. Artigas JL, et al. Congenital adrenal cortical carcinoma. J Pediatr Surg 1976; 11: 247 – 52. 64. Leung LY, et al. Ruptured adrenocortical carcinoma as a cause of paediatric acute abdomen. Pediatr Surg Int 2002; 18: 730 – 2. 65. Halmi KA, Lascari AD. Conversion of virilization to feminization in a young girl with adrenal cortical carcinoma. Cancer 1971; 27: 931 – 5. 66. Itami RM, et al. Prepubertal gynecomastia caused by an adrenal tumor. Am J Dis Child 1982; 136: 584 – 6. 67. Leditschke JF, Arden F. Feminizing adrenal adenoma in a five-yearold boy. Aust Paediatr J 1974; 10: 217 – 21. 68. Ranew RB, Meadows AT, D’Angio GJ. Adrenocortical carcinoma in children: experience at the Children’s Hospital of Philadelphia, 1961 – 1980. In Humphrey GB, et al. (eds) Adrenal and Endocrine Tumors in Children, 1st ed. Boston, Massachusetts: Martinus Nijhoff Publishers, 1983: 303 – 305. 69. Loridan L, Senior B. Cushing’s syndrome in infancy. J Pediatr 1969; 75: 349 – 59. 70. Kepes JJ, et al. Adrenal cortical adenoma in the spinal canal of an 8-year-old girl. Am J Surg Pathol 1990; 14: 481 – 4. 71. McWhirter WR, Stiller CA, Lennox EL. Carcinomas in childhood. A Registry-based study of incidence and survival. Cancer 1989; 63: 2242. 72. Medeiros LJ, et al. Virilizing adrenal cortical neoplasm arising ectopically in the thorax. J Clin Endocrinol Metab 1992; 75: 1522 – 5. 73. Page DL, DeLellis RA, Hough AJ. Embryology and postnatal development. In Page DL, DeLellis RA, Hough AJ (eds) Tumors of the Adrenal, 2nd Series edition. Washington, District of Columbia: Armed Forces Institute of Pathology, 1986: 25 – 35. 74. Hauffa BP, et al. Growth in children with adrenocortical tumors. Klin Padiatr 1991; 203: 83 – 7. 75. Schmit-Lobe MC, et al. Patterns of growth and development in 26 children operated for adrenocortical carcinoma (ACC) and diseasefree for more than one year. Pediatr Res 1990; 38: 622. 76. Gilbert MG, Cleveland WW. Cushing’s syndrome in infancy. Pediatrics 1970; 46: 217 – 29. 77. Weiss LM, Medeiros LJ, Vickery AL Jr. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989; 13: 202 – 6. 78. Weiss LM. Comparative study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 1984; 8: 163 – 9. 79. Hough AJ, et al. Prognostic factors in adrenal cortical tumors: a mathematical analysis of clinical and morphologic data. Am J Clin Pathol 1979; 72: 390 – 9. 80. Bugg MF, et al. Brazilian Group for Treatment of Childhood Adrenocortical Tumors. Correlation of pathologic features with clinical outcome in pediatric adrenocortical neoplasia. A study of a Brazilian population. Am J Clin Pathol 1994; 101: 625 – 9. 81. Kikumori T, et al. Intracaval endovascular ultrasonography for large adrenal and retroperitoneal tumors. Surgery 2003; 134: 989 – 93.
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS 82. Ahmed M, et al. Whole-body positron emission tomographic scanning in patients with adrenal cortical carcinoma: comparison with conventional imaging procedures. Clin Nucl Med 2003; 28: 494 – 7. 83. Minn H, et al. Imaging of adrenal incidentalomas with PET using (11)C-metomidate and (18)F-FDG. J Nucl Med 2004; 45: 972 – 9. 84. Godine LB, et al. Adrenocortical carcinoma with extension into inferior vena cava and right atrium: report of 3 cases in children. Pediatr Radiol 1990; 20: 166 – 8; discussion 169. 85. Chesson JP, Theodorescu D. Adrenal tumor with caval extension – case report and review of the literature. Scand J Urol Nephrol 2002; 36: 71 – 3. 86. Icard P, et al. Adrenocortical carcinomas: surgical trends and results of a 253-patient series from the French Association of Endocrine Surgeons study group. World J Surg 2001; 25: 891 – 7. 87. Baudin E, et al. Impact of monitoring plasma 1,1-dichlorodiphenildichloroethane (o,p DDD) levels on the treatment of patients with adrenocortical carcinoma. Cancer 2001; 92: 1385 – 92. 88. Berruti A, et al. Italian Group for the Study of Adrenal Cancer. Mitotane associated with etoposide, doxorubicin, and cisplatin in the treatment of advanced adrenocortical carcinoma. Cancer 1998; 83: 2194 – 200. 89. Khan TS, et al. Streptozocin and o,p DDD in the treatment of adrenocortical cancer patients: long-term survival in its adjuvant use. Ann Oncol 2000; 11: 1281 – 7. 90. Chudler RM, Kay R. Adrenocortical carcinoma in children. Urol Clin North Am 1989; 16: 469 – 79. 91. Wieneke JA, Thompson LD, Heffess CS. Adrenal cortical neoplasms in the pediatric population: a clinicopathologic and immunophenotypic analysis of 83 patients. Am J Surg Pathol 2003; 27: 867 – 81. 92. Michalkiewicz EL, et al. Clinical characteristics of small functional adrenocortical tumors in children. Med Pediatr Oncol 1997; 28: 175 – 8. 93. Wermer P. Genetic aspects of adenomatosis of endocrine glands. Am J Med 1954; 16: 363 – 71. 94. Sipple JH. The association of pheochromocytomas with carcinoma of the thyroid gland. Am J Med 1961; 31: 163 – 6. 95. Eng C. The RET proto-oncogene in multiple endocrine neoplasia type 2 and Hirschsprung’s disease. N Engl J Med 1996; 335: 943 – 51. 96. Carney JA, et al. The complex of myxomas, spotty pigmentation, and endocrine overactivity. Medicine (Baltimore) 1985; 64: 270 – 83. 97. Mulligan LM, et al. Germ-line mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature 1993; 363: 458 – 60. 98. Chandrasekharappa SC, et al. Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 1997; 276: 404 – 7. 99. Kirschner LS, et al. Mutations in the gene encoding the type Ia regulatory subunit of the protein kinase A (PRKARIA) in patients with Carney complex. Nat Genet 2000; 26: 89 – 92. 100. Hemminki A, et al. A serine/threonine kinase gene defective in PeutzJeghers syndrome. Nature 1998; 391: 184 – 7. 101. Jenne DE, et al. Peutz-Jeghers syndrome is caused by mutations in a novel serine threonine kinase. Nat Genet 1998; 18: 38 – 43. 102. Liaw D, et al. Germline mutations of the PTEN gene in Cowden disease, an inherited breast and thyroid cancer syndrome. Nat Genet 1997; 16: 64 – 7. 103. Latif F, et al. Identification of the von Hippel-Lindau disease tumor suppressor gene. Science 1993; 260: 1317 – 20. 104. Gagel RF, et al. Multiple endocrine neoplasia type 2 syndromes: nomenclature recommendations from the workshop organizing committee. Henry Ford Hosp Med J 1989; 37: 99. 105. Online Mendelian Inheritance in Man, OMIM (TM). Bethesda, MD: Center for Medical Genetics, Johns Hopkins University, and National Center for Biotechnology Information, National Library of Medicine, 1999: World Wide Web URL: http://www.ncbi.nlm.nih.gov/omim. 106. Marx SJ, et al. Multiple endocrine neoplasia type 1: clinical and genetic features of the hereditary endocrine neoplasias. Recent Prog Horm Res 1999; 54: 397 – 438. 107. Marx SJ. Multiple endocrine neoplasia type 1. In Bilezikian JP, Levine MA, Marcus R (eds) The Parathyroids. New York: Raven Press, 2000. 108. Trump D, et al. Clinical studies of multiple endocrine neoplasia type 1 (MEN 1). Q J Med 1996; 89: 653 – 69.
793
109. Vasen HF, Lamers CB, Lips CJ. Screening for the multiple endocrine neoplasia syndrome type I. A study of 11 kindreds in The Netherlands. Arch Intern Med 1989; 149: 2717 – 22. 110. Skarulis MC, et al. Multiple endocrine neoplasia type 1: clinical and genetic topics. Ann Intern Med 1998; 129: 484 – 94. 111. Lamers CB, Froeling PG. Clinical significance of hyperparathyroidism in familial multiple endocrine adenomatosis type I (MEA I). Am J Med 1979; 66: 422 – 4. 112. Marx SJ Jr, et al. Multiple endocrine neoplasia type I: assessment of laboratory tests to screen for the gene in a large kindred. Medicine (Baltimore) 1986; 65: 226 – 41. 113. Ballard HS, Frame B, Hartsock RJ. Familial multiple endocrine adenoma-peptic ulcer complex. Medicine 1964; 43: 481 – 516. 114. Betts JB, O’Malley BP, Rosenthal FD. Hyperparathyroidism: a prerequisite for Zollinger-Ellison syndrome in multiple endocrine adenomatosis Type 1 – report of a further family and a review of the literature. Q J Med 1980; 49: 69 – 76. 115. Scheithauer BW Jr, et al. Pituitary adenomas of the multiple endocrine neoplasia type I syndrome. Semin Diagn Pathol 1987; 4: 205 – 11. 116. Shepherd JJ. The natural history of multiple endocrine neoplasia type 1. Highly uncommon or highly unrecognized? Arch Surg 1991; 126: 935 – 52. 117. O’Brien T, et al. Results of treatment of pituitary disease in multiple endocrine neoplasia, Type I. Neurosurgery 1996; 39: 273 – 8; [discussion 278 – 279]. 118. Stratakis CA, et al. Pituitary macroadenoma in a 5-year-old: an early expression of multiple endocrine neoplasia type-1. J Clin Endocrinol Metab 2000; 85(12): 4776 – 80. 119. Giraud S, et al. Germ-line mutation analysis in patients with multiple endocrine neoplasia type 1 and related disorders. Am J Hum Genet 1997; 63: 455 – 67. 120. Thakker RV. Multiple endocrine neoplasia-syndromes of the twentieth century. J Clin Endocrinol Metab 1998; 83: 2617 – 20. 121. Gagel RF. RET protooncogene mutations and endocrine neoplasiaa story intertwined with neural crest differentiation. Endocrinology 1996; 137: 1509 – 11. 122. Eng C, et al. The relationship between specific RET proto-oncogene mutations and disease phenotype in multiple endocrine neoplasia type 2. International RET mutation consortium analysis. JAMA 1996; 276: 1575 – 9. 123. Edery P, et al. RET in human development and oncogenesis. Bioessays 1997; 19: 389 – 95. 124. Gagel RF, et al. Multiple endocrine neoplasia type 2A associated with cutaneous lichen amyloidosis. Ann Droit Int Med 1989; 111: 802 – 6. 125. Borrego S, et al. Molecular analysis of the ret and GDNF genes in a family with multiple endocrine neoplasia type 2A and Hirschsprung disease. J Clin Endocrinol Metab 1998; 83: 3361 – 4. 126. Eng C, Mulligan LM. Mutations of the RET proto-oncogene in the multiple endocrine neoplasia type 2 syndromes, related sporadic tumours, and Hirschsprung disease. Hum Mutat 1997; 9: 97 – 109. 127. Hoff A, Cote J, Gagel RF. Multiple endocrine neoplasias. Annu Rev Physiol 2000; 62: 377 – 411. 128. Angrist M, et al. Germline mutations in glial-derived neurotrophic factor (GDNF) and RET in a Hirschsprung disease patient. Nat Genet 1996; 14: 341 – 4. 129. Decker RA, Peacock ML, Watson P. Hirschsprung disease in MEN 2A: increased spectrum of RET exon 10 genotypes and strong genotype-phenotype correlation. Hum Mol Genet 1998; 7: 129 – 34. 130. Stratakis CA. Genetics of Carney complex and related familial lentiginoses, and other multiple tumor syndromes. Front Biosci 2000; 5: d353 – 66. 131. Salomon F, Froesh ER, Hedinger CE. Familial Cushing syndrome (Carney complex) [letter]. N Engl J Med 1990; 322: 1470. 132. Shenoy BV, Carpenter PC, Carney JA. Bilateral primary pigmented nodular adrenocortical disease. Rare cause of the Cushing syndrome. Am J Surg Pathol 1984; 8: 335 – 44. 133. Schweizer-Cagianut M, Froesh ER, Hedinger CE. Familial Cushing’s syndrome with primary adrenocortical microadenomatosis (primary adrenocortical nodular dysplasia). Acta Endocrinol (Copenh) 1980; 94: 529 – 35.
794
PEDIATRIC MALIGNANCIES
134. Carney JA, et al. Dominant inheritance of the complex of myxomas, spotty pigmentation and endocrine overactivity. Mayo Clin Proc 1986; 61: 165 – 72. 135. Carney JA, Young WF. Primary pigmented nodular adrenocortical disease and its associated conditions. Endocrinologist 1992; 2: 6 – 21. 136. Atherton DJ, et al. A syndrome of various cutaneous pigmented lesions, myxoid neurofibromata and atrial myxoma: the NAME syndrome. Br J Dermatol 1980; 103: 421 – 9. 137. Rhodes AR. et al. Mucocutaneous lentigines, cardiomucocutaneous myxomas, and multiple blue nevi: the “LAMB” syndrome. J Am Acad Dermatol 1984; 10: 72 – 82. 138. Carney JA. Differences between nonfamilial and familial cardiac myxoma. Am J Surg Pathol 1985; 9: 53 – 5. 139. Kennedy RH Jr, et al. The Carney complex with ocular signs suggestive of cardiac myxoma. Am J Ophthalmol 1991; 111: 699 – 702. 140. Ferreiro JA, Carney JA. Myxomas of the external ear and their significance. Am J Surg Pathol 1994; 18: 274 – 80. 141. Carney JA, Toorkey BC. Myxoid fibroadenoma and allied conditions (myxomatosis) of the breast. A heritable disorder with special associations including cardiac and cutaneous myxomas. Am J Surg Pathol 1991; 15: 713 – 21. 142. Courcoutsakis NA, et al. Breast imaging findings in the complex of myxomas, spotty pigmentation, endocrine veractivity, and schwannomas (Carney complex). Radiology 1997; 205: 221 – 7. 143. Carney JA, Ferreiro JA. The epithelioid blue nevus. A multicentric familial tumor with important associations, including cardiac myxoma and psammomatous melanotic schwannoma. Am J Surg Pathol 1996; 20: 259 – 72. 144. Carney JA. Carney complex: the complex of myxomas, spotty pigmentation, endocrine veractivity, and schwannomas. Semin Dermatol 1995; 14: 90 – 8. 145. Doppman JL, et al. Cushing syndrome due to primary pigmented nodular adrenocortical disease: findings at CT and MR imaging. Radiology 1989; 172: 415 – 20. 146. Mellinger RC, Smith RW. Studies of the adrenal hyperfunction in 2 patients with atypical Cushing’s syndrome. J Clin Endocrinol Metab 1955; 16: 350 – 66. 147. Kracht J, Tamm J. Bilaterale kleiknotige Adenomatose der Nebennierenrinde bei Cushing-Syndrom. Virchows Arch Pathol Anat 1960; 333: 1 – 9. 148. Levin ME. The development of bilateral adenomatous adrenal hyperplasia in a case of Cushing’s syndrome of eighteen years’ duration. Am J Med 1966; 40: 318 – 24. 149. De Moor P, et al. Unusual case of adrenocortical hyperfunction. J Clin Endocrinol Metab 1965; 25: 612 – 20. 150. Sarlis NJ, et al. Primary pigmented nodular adrenocortical disease (PPNAD): re-evaluation of a patient with Carney complex 27 years after unilateral adrenalectomy. J Clin Endocrinol Metab 1997; 82: 2037 – 43. 151. Gomez-Muguruza MT, Chrousos GP. Periodic Cushing’s syndrome in a short boy: usefulness of the ovine corticotropin releasing hormone test. J Pediatr 1989; 115: 270 – 3. 152. Stratakis CA, et al. Paradoxical response to dexamethasone assists with the diagnosis of primary pigmented nodular adrenocortical disease (PPNAD). Ann Intern Med 1999; 131: 585 – 91. 153. Watson JC, et al. Neurosurgical implications of Carney complex. J Neurosurg 2000; 92: 413 – 8. 154. Raff SB, et al. Prolactin secretion abnormalities in patients with the “syndrome of spotty skin pigmentation, myxomas, endocrine overactivity and schwannomas” (Carney complex). J Pediatr Endocrinol Metab 2000; 13: 373 – 9. 155. Pack S, et al. Pituitary tumors in patients with the “complex of spotty skin pigmentation, myxomas, endocrine overactivity and schwannomas” (Carney complex): evidence for progression from somatomammotroph hyperplasia to adenoma. J Clin Endocrinol Metab 2000; 85: 3860 – 5. 156. Premkumar A, et al. Testicular ultrasound in Carney complex. J Clin Ultrasound 1997; 25: 211 – 4. 157. Stratakis CA, et al. Ovarian cysts in patients with Carney complex: clinical and genetic studies and evidence for predisposition to cancer. J Clin Endocrinol Metab 2000; 85: 4359 – 66.
158. Carney JA. Psammomatous melanotic schwannoma. A distinctive, heritable tumor with special associations, including cardiac myxoma and the Cushing syndrome. Am J Surg Pathol 1990; 14: 206 – 22. 159. Carney JA, Toorkey BC. Ductal adenoma of the breast with tubular futures. A probable component of the complex of myxomas, spotty pigmentation, endocrine overactivity, and schwannomas. Am J Surg Pathol 1991; 15: 722 – 31. 160. Carney JA, Stratakis CA. Ductal adenoma of the breast [letter]. Am J Surg Pathol 1996; 20: 1154 – 5. 161. Stratakis CA, et al. Thyroid gland abnormalities in patients with the “syndrome of spotty skin pigmentation, myxomas, and endocrine overactivity” (Carney complex). J Clin Endocrinol Metab 1997; 82: 2037 – 43. 162. Stratakis CA, et al. Carney complex, a familial multiple neoplasia and lentiginosis syndrome. Analysis of 11 kindreds and linkage to the short arm of chromosome 2. J Clin Invest 1996; 97: 699 – 705. 163. Casey M, et al. Identification of a novel genetic locus for familial cardiac myxomas and Carney complex. Circulation 1998; 98: 2560 – 6. 164. Stratakis CA, et al. Carney complex, a multiple endocrine neoplasia and familial lentiginosis syndrome: clinical analysis and linkage to the D2S123 locus (chromosome 2p16). Am J Hum Genet 1995; 57: A54. 165. Dewald GW, et al. Chromosomally abnormal clones and nonrandom telomeric translocations in cardiac myxomas. Mayo Clin Proc 1987; 62: 558 – 67. 166. Dijkhuizen T, et al. Cytogenetics of a case of cardiac myxoma. Cancer Genet Cytogenet 1992; 63: 73 – 5. 167. Richkind KE, Wason D, Vidaillet HJ. Cardiac myxoma characterized by clonal telomeric association. Genes Chromosomes Cancer 1994; 9: 68 – 71. 168. Stratakis CA, et al. Cytogenetic and microsatellite alterations in tumors from patients with the syndrome of myxomas, spotty skin pigmentation, and endocrine overactivity (Carney complex). J Clin Endocrinol Metab 1996; 81: 3607 – 14. 169. Takai S, Iwama T, Tonomura A. Chromosome instability in cultured skin fibroblasts from patients with familial polyposis coli and PeutzJeghers syndrome. Jpn J Cancer Res 1986; 77: 759 – 66. 170. Griffin CA, et al. Cytogenetic analysis of intestinal polyps in polyposis syndromes: comparison with sporadic colorectal adenomas. Cancer Genet Cytogenet 1993; 67: 14 – 20. 171. Richard F, Muleris M, Dutrillaux B. Chromosome instability in lymphocytes from patients affected by or genetically predisposed to colorectal cancer. Cancer Genet Cytogenet 1994; 73: 23 – 32. 172. DeMarco L, et al. Sporadic cardiac myxomas and tumors from patients with Carney complex are not associated with activating mutations of the Gsα gene. Hum Genet 1996; 98: 185 – 8. 173. Basson CT, et al. Genetic heterogeneity of familial atrial myxoma syndromes (Carney complex). Am J Cardiol 1997; 79: 994 – 5. 174. Taymans SE, et al. A refined genetic, radiation hybrid, and physical map of the Carney complex (CNC) locus on chromosome 2p16; evidence for genetic heterogeneity in the syndrome. Am J Hum Genet 1997; 61: A84. 175. Scott JD. Cyclic nucleotide-dependent protein kinases. Pharmacol Ther 1991; 50: 123 – 45. 176. Kirschner LS, et al. Genetic heterogeneity and spectrum of mutations of the PRKAR1A gene in patients with the Carney complex. Hum Mol Genet 2000; 9(20): 3037 – 46. 177. Taymans SE, et al. YAC-BAC contig of the Carney complex (CNC) critical region on 2p16 and copy number gain of 2p16 in CNC tumors: evidence for a novel oncogene? Am J Hum Genet 1999; 65: A326. 178. Stratakis CA, et al. Carney Complex, Peutz-Jeghers Syndrome, Cowden Disease, and Bannayan-Zonana Syndrome share cutaneous and endocrine manifestations, but not genetic loci. J Clin Endocrinol Metab 1998; 83: 2972 – 6. 179. Peutz JLA. Very remarkable case of familial polyposis of mucous membrane of intestinal tract and nasopharynx accompanied by peculiar pigmentation of skin and mucous membrane. (Dutch). Nederl Maandschr Geneesk 1921; 10: 134 – 46. 180. Jeghers H, McKusick VA, Katz KH. Generalised intestinal polyposis and melanin spots of the oral mucosa, lips and digits. N Engl J Med 1949; 241: 993 – 1005 and 1031 – 6. 181. Hemminki A. The molecular basis and clinical aspects of PeutzJeghers syndrome. Cell Mol Life Sci 1999; 55: 735 – 50.
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS 182. Hemminki A, et al. Localisation of a susceptibility locus for PeutzJeghers syndrome to 19p using comparative genomic hybridisation and targeted linkage analysis. Nat Genet 1997; 15: 87 – 90. 183. Boardman LA, et al. Increased risk for cancer in patients with the Peutz-Jeghers syndrome. Ann Intern Med 1998; 128: 896 – 9. 184. Westerman AM, Wilson JHP. Peutz-Jeghers syndrome: risks of a hereditary condition. A clinical review. Scand J Gastroenterol 1999; 230: 64 – 70. 185. Giardello FM, et al. Increased risk of cancer in the Peutz-Jeghers syndrome. N Engl J Med 1987; 316: 1511 – 4. 186. Hizawa K, et al. Cancer in Peutz-Jeghers syndrome. Cancer 1993; 72: 2777 – 81. 187. Foley TR, McGarrity TJ, Abt AB. Peutz-Jeghers syndrome: a clinicopathologic survey of the “Harrisburg family” with a 49-year follow up. Gastroenterology 1988; 95: 1535 – 40. 188. Rustgi AK. Hereditary gastrointestinal polyposis and non-polyposis syndromes. N Engl J Med 1994; 331: 1694 – 702. 189. Spigelman AD, Murday V, Phillips RKS. Cancer and Peutz-Jeghers syndrome. Gut 1989; 30: 1588 – 90. 190. Pathomvanich A, Koch C, Stratakis CA. Thyroid abnormalities in Peutz-Jeghers syndrome: report of new observation and review of the literature. J Endocr Genet 1999; 1: 47 – 9. 191. Young RH, Scully RE. Mucinous ovarian tumors associated with mucinous adenocarcinomas of the cervix: clinicopathologic analysis of 16 cases. Int J Gynecol Pathol 1988; 7: 99 – 111. 192. Young RH, et al. Ovarian sex stromal tumor with annular tubules; review of 74 cases including 27 with Peutz-Jeghers syndrome and 4 with adenoma malignum of the cervix. Cancer 1982; 50: 1384 – 402. 193. Young S, et al. Feminizing Sertoli cell tumors in boys with PeutzJeghers syndrome. Am J Surg Pathol 1995; 19: 50 – 8. 194. Markie D, et al. A pericentric inversion of chromosome 6 in a patient with Peutz-Jeghers syndrome and the use of FISH to localize the breakpoints on a genetic map. Hum Genet 1996; 98: 125 – 8. 195. Bali D, et al. Peutz-Jeghers syndrome maps to chromosome 1p. Am J Hum Genet 1995; 57: A186. 196. Tomlinson IPM, et al. Testing candidate loci on chromosomes 1 and 6 for genetic linkage to Peutz-Jegher’s disease. Ann Hum Genet 1996; 60: 377 – 84. 197. Olschwang S, et al. Peutz-Jeghers disease: most, but not all, families are compatible with linkage to 19p13.3. J Med Genet 1998; 35: 42 – 4. 198. Mehenni H, et al. Peutz-Jeghers Syndrome: Confirmation of linkage to chromosome 19p13.3 and identification of a potential second locus, on 19q13.4. Am J Hum Genet 1997; 61: 1327 – 34. 199. Westerman AM, et al. Novel mutations in the LKB1/STK11 gene in Dutch Peutz-Jeghers families. Hum Mutat 1999; 13: 476 – 81. 200. Wang Z.-J, et al. Germline mutations of the LKB1 (STK11) gene in Peutz-Jeghers patients. J Med Genet 1999; 36: 365 – 8. 201. Resta N, et al. STK11 mutations in Peutz-Jeghers syndrome and sporadic colon cancer. Cancer Res 1998; 58: 4799 – 801. 202. Jiang C.-Y, et al. STK11/LKB1 germline mutations are not identified in most Peutz-Jeghers syndrome patients. Cancer Res 1999; 56: 136 – 41. 203. Ylikorkala A, et al. Aaltonen LA: mutations and impaired function of LKB1 in familial and non-familial Peutz-Jeghers syndrome and a sporadic testicular cancer. Hum Mol Genet 1999; 8: 45 – 51. 204. Boardman LA, et al. Genetic heterogeneity in Peutz-Jeghers syndrome. Hum Mutat 2000; 16: 23 – 30. 205. Su J.-Y, Erikson E, Maller JL. Cloning and characterisation of a novel serine/threonine protein kinase expressed in early Xenopus embryos. J Biol Chem 1996; 271: 14430 – 7. 206. Smith DP, et al. The mouse Peutz-Jeghers syndrome gene Lkb1 encodes a nuclear protein kinase. Hum Mol Genet 1999; 8: 1479 – 85. 207. Mehenni H, et al. Loss of LKB1 kinase activity in Peutz-Jeghers syndrome, and evidence for allelic and locus heterogeneity. Am J Hum Genet 1998; 63: 1641 – 50. 208. Nelen MR, et al. Localization of the gene for Cowden disease to chromosome 10q22-23. Nat Genet 1996; 13: 114 – 6. 209. Eng C. Genetics of Cowden syndrome: through the looking glass of oncology. Int J Oncol 1998; 12: 701 – 10. 210. Nelen MR, et al. Germline mutations in the PTEN/MMAC1 gene in patients with Cowden disease. Hum Mol Genet 1997; 6: 1383 – 7.
795
211. Marsh DJ, et al. PTEN mutation spectrum and genotype-phenotype correlations in Bannayan-Riley-Ruvalcaba syndrome suggest a single entity with Cowden syndrome. Hum Mol Genet 1999; 8: 1461 – 72. 212. Longy M, et al. Mutations of PTEN in patients with Bannayan-RileyRuvalcaba phenotype. J Med Genet 1998; 35: 886 – 9. 213. Linehan WM, Lerman MI, Zbar B. Identification of the von HippelLindau (VHL) gene. JAMA 1995; 273: 564 – 70. 214. Richards FM, et al. Molecular genetic analysis of von Hippel-Lindau disease. J Intern Med 1998; 243: 527 – 33. 215. Bar M, et al. Sporadic phaeochromocytomas are rarely associated with germline mutations in the von Hippel-Lindau and RET genes. Clin Endocrinol (Oxf) 1997; 47: 707 – 12. 216. Bernstein I, Gurney JG. Carcinomas and Other Malignant Epithelial Neoplasms. ICCC XI. Pediatric Monograph, NCI SEER. Available at http://Seer.cancer.gov/Publications/childhood/. 217. Wartofsky L. The thyroid nodule. In Wartofsky L (ed) Thyroid Cancer: A Comprehensive Guide to Clinical Management. Totowa, New Jersy: Humana Press, 2000: 3 – 7. 218. Corrias A, et al. Accuracy of fine needle aspiration biopsy of thyroid nodules in detecting malignancy in childhood: comparison with conventional clinical, laboratory, and imaging approaches. J Clin Endocrinol Metab 2001; 86(10): 4644 – 8. 219. Sklar C, et al. Abnormalities of the thyroid in survivors of Hodgkin’s disease: data from The Childhood Cancer Survivor Study. J Clin Endocrinol Metab 2000; 85: 3227 – 32. 220. Tucker MA, et al. Therapeutic radiation at a young age is linked to secondary thyroid cancer. Cancer Res 1991; 51(11): 2885 – 8. 221. Kumar A, Bal CS. Differentiated thyroid cancer. Indian J Pediatr 2003; 70(9): 707 – 13. 222. Giuffrida D, et al. Differentiated thyroid cancer in children and adolescents. J Endocrinol Invest 2002; 25: 18 – 24. 223. Lee YM, et al. Well-differentiated thyroid carcinoma in Hong Kong Chinese patients under 21 years of age: a 35-year experience. J Am Coll Surg 2002; 194: 711 – 6. 224. Brink JS, et al. Papillary thyroid cancer with pulmonary metastases in children: long-term prognosis. Surgery 2000; 128: 881 – 7. 225. Landau D, et al. Thyroid cancer in children: the Royal Marsden Hospital experience. Eur J Cancer 2000; 36: 214 – 20. 226. Chow SM, et al. Differentiated thyroid carcinoma in childhood and adolescence – clinical course and role of radioiodine. Pediatr Blood Cancer 2004; 42: 176 – 83. 227. Powers PA, et al. Tumor size and extent of disease at diagnosis predict the response to initial therapy for papillary thyroid carcinoma in children and adolescents. J Pediatr Endocrinol Metab 2003; 16: 693 – 702. 228. Samuel AM, Sharma SM. Differentiated thyroid carcinomas in children and adolescents. Cancer 1991; 67: 2186 – 90. 229. Ceccarelli C, et al. Thyroid cancer in children and adolescents. Surgery 1988; 104: 1143 – 8. 230. Viswanathan K, Gierlowski TC, Schneider AB. Childhood thyroid cancer: characteristics and long-term outcome in children irradiated for benign conditions of the head and neck. Arch Pediatr Adolesc Med 1994; 148: 260 – 3. 231. Harness JK, et al. Differentiated thyroid carcinoma in children and adolescents. World J Surg 1992; 16: 47 – 54. 232. Schlumberger M, et al. Differentiated thyroid carcinoma in childhood: long term follow-up of 72 patients. J Clin Endocrinol Metab 1987; 65: 1088 – 94. 233. Welch-Dinauer CA, et al. Clinical features associated with metastasis and recurrence of differentiated thyroid cancer in children, adolescents and young adults. Clin Endocrinol 1998; 49: 619 – 28. 234. Lamberg BA, Karkinen-Jaaskelainen M, Franssila KO. Differentiated follicle-derived thyroid carcinoma in children. Acta Paediatr Scand 1989; 78: 419 – 25. 235. Jarzab B, et al. Multivariate analysis of prognostic factors for differentiated thyroid carcinoma in children. Eur J Nucl Med 2000; 27: 833 – 41. 236. Bal CS, et al. Is chest x-ray or high-resolution computed tomography scan of the chest sufficient investigation to detect pulmonary metastasis in pediatric differentiated thyroid cancer ? Thyroid 2004; 14: 217 – 24.
796
PEDIATRIC MALIGNANCIES
237. Haveman JW, et al. Surgical experience in children with differentiated thyroid carcinoma. Ann Surg Oncol 2003; 10(1): 15 – 20. 238. Fenton CL, et al. The ret/PTC mutations are common in sporadic papillary thyroid carcinoma of children and young adults. J Clin Endocrinol Metab 2000; 85: 1170 – 5. 239. Learoyd DL, et al. RET/PTC and RET tyrosine kinase expression in adult papillary thyroid carcinomas [see comments]. J Clin Endocrinol Metab 1998; 83: 3631 – 5. 240. Tuttle RM, et al. Activation of the ret/PTC Oncogene in Papillary Thyroid Cancer from Russian Children Exposed to Radiation Following the Chernobyl Accident. Kyoto, Japan: Twelfth International Thyroid Congress, 2000. 241. Sarasin A, et al. Mechanisms of mutagenesis in mammalian cells. Application to human thyroid tumours. C R Acad Sci III 1999; 322: 143 – 9. 242. Jhiang SM, et al. Thyroid carcinomas in RET/PTC transgenic mice. Recent Results Cancer Res 1998; 154: 265 – 70. 243. Sugg SL, et al. Ret/PTC-1, -2, and -3 oncogene rearrangements in human thyroid carcinomas: implications for metastatic potential? J Clin Endocrinol Metab 1996; 81: 3360 – 5. 244. Kjellman P, et al. Expression of the RET proto-oncogene in papillary thyroid carcinoma and its correlation with clinical outcome. Br J Surg 2001; 88: 557 – 63. 245. Nikiforov YE, et al. Distinct pattern of ret oncogene rearrangements in morphological variants of radiation-induced and sporadic thyroid papillary carcinomas in children. Cancer Res 1997; 57(9): 1690 – 4. 246. Ozaki O, et al. Familial occurrence of differentiated, nonmedullary thyroid carcinoma. World J Surg 1988; 12: 565 – 71. 247. Sturgeon C, Clark OH. Familial nonmedullary thyroid cancer. Thyroid 2005; 15(6): 588 – 93. 248. Hemminki K, Eng C, Chen B. Familial risks for nonmedullary thyroid cancer. J Clin Endocrinol Metab 2005; 90(10): 5747 – 53. 249. Eng C. Role of PTEN, a lipid phosphatase upstream effector of protein kinase B, in epithelial thyroid carcinogenesis. Ann N Y Acad Sci 2002; 968: 213 – 21. 250. Marsh DJ, et al. Mutation spectrum and genotype-phenotype analyses in Cowden disease and Bannayan-Zonana syndrome, two hamartoma syndromes with germline PTEN mutation. Hum Mol Genet 1998; 7(3): 507 – 15. 251. Liaw D, et al. Germline mutations of the PTEN gene in Cowden disease, an inherited breast and thyroid cancer syndrome. Nat Genet 1997; 16(1): 64 – 7. 252. Perrier ND, et al. Thyroid cancer in patients with familial adenomatous polyposis. World J Surg 1998; 22(7): 738 – 42; discussion 743. 253. Bell B, Mazzaferri EL. Familial adenomatous polyposis (Gardner’s syndrome) and thyroid carcinoma. A case report and review of the literature. Dig Dis Sci 1993; 38(1): 185 – 90. 254. Stratakis CA, Kirschner LS, Carney JA. Clinical and molecular features of the Carney complex: diagnostic criteria and recommendations for patient evaluation. J Clin Endocrinol Metab 2001; 86(9): 4041 – 6. 255. Kirschner LS II, et al. A mouse model for the Carney complex tumor syndrome develops neoplasia in cyclic AMP-responsive tissues. Cancer Res 2005; 65(11): 4506 – 14. 256. Sandrini F, et al. Regulatory subunit type I-alpha of protein kinase A (PRKAR1A): a tumor-suppressor gene for sporadic thyroid cancer. Genes Chromosomes Cancer 2002; 35(2): 182 – 92. 257. Degnan BM, McClellan DR, Francis GL. An analysis of fine-needle aspiration biopsy of the thyroid in children and adolescents. J Pediatr Surg 1996; 31: 903 – 7. 258. Amrikachi M, et al. Thyroid fine-needle aspiration biopsy in children and adolescents: experience with 218 aspirates. Diagn Cytopathol 2005; 32(4): 189 – 92. 259. Hopwood NJ, Kelch RP. Thyroid masses: approach to diagnosis and management in childhood and adolescence. Pediatr Rev 1993; 14: 481 – 7. 260. Hung W, et al. Solitary thyroid nodules in 71 children and adolescents. J Pediatr Surg 1992; 27: 1407 – 9. 261. Lugo-Vicente H, Ortiz VN. Pediatric thyroid nodules: insights in management. Bol Asoc Med P R 1998; 90: 74 – 8.
262. Belfiore A, et al. High frequency of cancer in cold thyroid nodules occurring at young age. Acta Endocrinol (Copenh) 1989; 121: 197 – 202. 263. Fassina AS, et al. Thyroid cancer in children and adolescents. Tumori 1994; 80: 257 – 62. 264. Bal CS, et al. Is chest x-ray or high-resolution computerized tomography of the chest sufficient investigation to detect pulmonary metastasis in pediatric differentiated thyroid cancer. Thyroid 2004; 14: 217 – 25. 265. Kowalski LP, et al. Long-term survival rates in young patients with thyroid carcinoma. Arch Otolaryngol Head Neck Surg 2003; 129: 746 – 9. 266. Farahati J, et al. Characteristics of differentiated thyroid carcinoma in children and adolescents with respect to age, gender, and histology. Cancer 1997; 80(ll): 2156 – 62. 267. Stael AP, et al. Total thyroidectomy in the treatment of thyroid carcinoma in childhood. Br J Surg 1995; 82: 1083 – 5. 268. Patwardhan N, Cataldo T, Braverman LE. Surgical management of the patient with papillary cancer. Surg Clin North Am 1995; 75: 449 – 64. 269. Shindo ML. Considerations in surgery of the thyroid gland. Otolaryngol Clin North Am 1996; 29: 629 – 35. 270. Vassilopoulou-Sellin R, et al. Differentiated thyroid cancer in children and adolescents: clinical outcome and mortality after long-term follow-up. Head & Neck 1998; 20(6): 549 – 55. 271. Frankenthaler RA, et al. Lymph node metastasis from papillaryfollicular thyroid carcinoma in young patients. Am J Surg 1990; 160: 341 – 3. 272. Massimino M, et al. Primary thyroid carcinoma in children: a retrospective study of 20 patients. Med Pediatr Oncol 1995; 24: 13 – 7. 273. Robie DK, et al. The impact of initial surgical management on outcome in young patients with differentiated thyroid cancer. J Pediatr Surg 1999; 33: 1134 – 40. 274. Welch-Dinauer CA, et al. Extensive surgery improves recurrence-free survival for children and young patients with class I papillary thyroid carcinoma. J Pediatr Surg 1999; 34: 1799 – 804. 275. Borson-Chazot F, et al. Predictive factors for recurrence from a series of 74 children and adolescents with differentiated thyroid cancer. World J Surg 2004; 28: 1088 – 92. 276. Thompson GB, Hay ID. Current strategies for surgical management and adjuvant treatment of childhood papillary thyroid carcinoma. World J Surg 2004; 28: 1187 – 98. 277. Hung W, Sarlis NJ. Current controversies in the management of pediatric patients with well-differentiated nonmedullary thyroid cancer: a review. Thyroid 2002; 12(8): 683 – 702. 278. Wartofsky L (ed) Thyroid Cancer: A Comprehensive Guide to Clinical Management, 2nd ed. Totowa, New Jersey: Humana Press, 2005. 279. Feinmesser R, et al. Carcinoma of the thyroid in children – a review. J Pediatr Endocrinol Metab 1997; 10(6): 561 – 8. 280. Zimmerman D, et al. Papillary thyroid carcinoma in children and adults: long-term follow-up of 1039 patients conservatively treated at one institution during three decades. Surgery 1988; 104: 1157 – 66. 281. Samaan NA, et al. The results of various modalities of treatment of well differentiated thyroid carcinoma: a retrospective review of 1599 patients. J Clin Endocrinol Metab 1992; 75: 714 – 20. 282. Travagli JP, et al. Differentiated thyroid carcinoma in childhood. J Endocrinol Invest 1995; 18: 161 – 4. 283. Merrick Y, Hansen HS. Thyroid cancer in children and adolescents in Denmark. Eur J Surg Oncol 1989; 15: 49 – 53. 284. Eisenhofer G, Lenders JW, Pacak K. Biochemical diagnosis of pheochromocytoma. Front Horm Res 2004; 31: 76 – 106. 285. Lenders JW, et al. Biochemical diagnosis of pheochromocytoma: which test is best? JAMA 2002; 287: 1427 – 34. 286. Brouwers FM, et al. Pheochromocytoma as an endocrine emergency. Rev Endocr Metab Disord 2003; 4: 121 – 8. 287. Lenders JW, et al. Pheochromocytoma. Lancet 2005; 366: 665 – 675. 288. Kuchel O. Pseudopheochromocytoma. Hypertension 1985; 7: 151 – 8. 289. Kuchel O. New insights into pseudopheochromocytoma and emotionally provoked hypertension. In Mansoor GA (ed) Secondary Hypertension. Totowa, New Jersey: Humana Press, 2004. 290. Bryant J, et al. Pheochromocytoma: the expanding genetic differential diagnosis. J Natl Cancer Inst 2003; 95: 1196 – 204.
UNCOMMON ENDOCRINE TUMORS IN CHILDREN AND ADOLESCENTS 291. Neumann HP, et al. Germ-line mutations in nonsyndromic pheochromocytoma. N Engl J Med 2002; 346: 1459 – 66. 292. Gimenez-Roqueplo AP, et al. Mutations in the SDHB gene are associated with extra-adrenal and/or malignant phaeochromocytomas. Cancer Res 2003; 63: 5615 – 21. 293. Eisenhofer G, et al. Pheochromocytomas in von Hippel-Lindau syndrome and multiple endocrine neoplasia type 2 display distinct biochemical and clinical phenotypes. J Clin Endocrinol Metab 2001; 86: 1999 – 2008. 294. Pacak K, et al. Biochemical diagnosis, localization and management of pheochromocytoma: focus on multiple endocrine neoplasia type 2 in relation to other hereditary syndromes and sporadic forms of the tumour. J Intern Med 2005; 257: 60 – 8. 295. Gertner ME, Kebebew E. Multiple endocrine neoplasia type 2. Curr Treat Options Oncol 2004; 5: 315 – 25. 296. Maher ER. Von Hippel-Lindau disease. Curr Mol Med 2004; 4: 833 – 42. 297. Hes FJ, Hoppener JW, Lips CJ. Clinical review 155: Pheochromocytoma in Von Hippel-Lindau disease. J Clin Endocrinol Metab 2003; 88: 969 – 74. 298. Gimenez-Roqueplo AP, et al. Functional consequences of a SDHB gene mutation in an apparently sporadic pheochromocytoma. J Clin Endocrinol Metab 2002; 87: 4771 – 4. 299. Astrom K, et al. Altitude is a phenotypic modifier in hereditary paraganglioma type 1: evidence for an oxygen-sensing defect. Hum Genet 2003; 113: 228 – 37. 300. Baysal BE. Genomic imprinting and environment in hereditary paraganglioma. Am J Med Genet C Semin Med Genet 2004; 129: 85 – 90. 301. Astuti D, et al. Gene mutations in the succinate dehydrogenase subunit SDHB cause susceptibility to familial pheochromocytoma and to familial paraganglioma. Am J Hum Genet 2001; 69: 49 – 54. 302. Gimm O, et al. Somatic and occult germ-line mutations in SDHD, a mitochondrial complex II gene, in nonfamilial pheochromocytoma. Cancer Res 2000; 60: 6822 – 5. 303. Benn DE, et al. Novel succinate dehydrogenase subunit B (SDHB) mutations in familial phaeochromocytomas and paragangliomas,
304.
305. 306. 307.
308.
309. 310.
311.
312.
313.
314.
797
but an absence of somatic SDHB mutations in sporadic phaeochromocytomas. Oncogene 2003; 22: 1358 – 64. Amar L, et al. Related articles, links abstract genetic testing in pheochromocytoma or functional paraganglioma. J Clin Oncol 2005; 23: 8812 – 8. O’Riordain DS Jr, et al. Clinical spectrum and outcome of functional extraadrenal paraganglioma. World J Surg 1996; 20: 916 – 21. John H, et al. Pheochromocytomas: can malignant potential be predicted? Urology 1999; 53: 679 – 83. Sawka AM Jr, et al. A comparison of biochemical tests for pheochromocytoma: measurement of fractionated plasma metanephrines compared with the combination of 24-hour urinary metanephrines and catecholamines. J Clin Endocrinol Metab 2003; 88: 553 – 8. Eisenhofer G, et al. Pheochromocytoma catecholamine phenotypes and prediction of tumor size and location by use of plasma free metanephrines. Clin Chem 2005; 51: 735 – 44. Eisenhofer G, et al. Plasma metanephrines in renal failure. Kidney Int 2005; 67: 668 – 77. Eisenhofer G, et al. Biochemical and clinical manifestations of dopamine-producing paragangliomas: utility of plasma methoxytyramine. J Clin Endocrinol Metab 2005; 90: 2086 – 75. Ilias I, Pacak K. Current approaches and recommended algorithm for the diagnostic localization of pheochromocytoma. J Clin Endocrinol Metab 2004; 89: 479 – 91. Pacak K, et al. 6-[18F]fluorodopamine positron emission tomographic (PET) scanning for diagnostic localization of pheochromocytoma. Hypertension 2001; 38: 6 – 8. Ilias I, et al. Superiority of 6-[18F]-fluorodopamine positron emission tomography versus [131I]-metaiodobenzylguanidine scintigraphy in the localization of metastatic pheochromocytoma. J Clin Endocrinol Metab 2003; 88: 4083 – 7. Mamede M, et al. Discordant localization of 2-[18F]-fluoro-2deoxy-D-deoxyglucose in 6-[18F]-fluorodopamine – and [123I]metaiodobenzylguanidine-negative metastatic pheochromocytoma. Nucl Med Commun 2006; 27: 31 – 36.
Section 11 : Pediatric Malignancies
71
Uncommon Pediatric Brain Tumors Sharon H. Smith
INTRODUCTION Brain tumors are the most common solid tumors of childhood. In most series, they account for close to 25% of childhood cancer diagnoses.1 The histology of brain tumors in patients 14 years of age and younger reported by the National Cancer Database include astrocytomas (53%), medulloblastomas (24%), and ependymomas (9%) (see Figure 1).1 Older reports of Surveillance and Epidemiology and End Results (SEER) data have documented a similar breakdown of histologies with all types of astrocytomas accounting for 48% of diagnoses. Medulloblastomas accounted for 23%. Brain stem gliomas were diagnosed in 9% of the patients and ependymomas in 8%. Unclassified tumors made up 7% of the total.2 The subject of this chapter are those tumors noted in most registries as “other” or “unclassified” which when combined still account for only a small percentage of pediatric brain tumor diagnoses (see Table 1).
DESMOPLASTIC NEUROEPITHELIAL TUMORS Desmoplastic Cerebral Astrocytoma of Infancy/Desmoplastic Infantile Ganglioglioma The first report of a series of patients with tumors consisting of a mix of astrocytes and fibroblasts was by Taratuto et al.,3 who described six infants with tumors designated superficial cerebral astrocytoma attached to the dura. Others have called these neoplasms desmoplastic cerebral astrocytomas of infancy (DCAI).4 Three years later, Vandenberg et al. described 11 patients having similar clinical features with what they had named desmoplastic infantile gangliogliomas (DIGs).5 Pathology
The gross appearance of both tumors includes multiple cysts and a distinctive firmness. The histology of DIGs is noted for prominent desmoplasia, astrocytic and ganglionic differentiation, glial fibrillary acidic protein (GFAP) positive astrocytes, and neoplastic neuronal elements.5 The superficial cerebral astrocytoma is very similar to the DIG, lacking only neurons, and it has been suggested by some to be
the same entity since neurons may be rare and difficult to identify.6 Others have felt that this explanation is unlikely as extensive sampling by light, electron microscopic, and immunohistochemical evaluation have shown no trace of neuronal differentiation in some tumors.3 Because of their high mitotic activity and focally high cellularity, both may be misdiagnosed as highly malignant lesions, most often as cerebral neuroblastomas5 or malignant astrocytomas.6,7 Clinical Features
Both DCAIs and DIGs are described as massive cystic lesions typically occurring in the cerebral hemispheres. They are most often diagnosed in very young infants. More than half are diagnosed at less than 6 months and almost all by less than 18 months of age. However, one case has been reported in a teenage boy and another in a young adult.8 It has been speculated that these are congenital tumors. One infant, however, had an ultrasonography at birth because of prematurity, which was normal, but subsequently was diagnosed with an extremely large cystic lesion, easily visible on imaging at 5 months of age.6 The tumors are usually located in the frontal or parietal lobes, although occipital and temporal lesions have been described. Most infants present with a short duration of symptoms. Seizures may be noted but are uncommon. Usually, a bulging fontanelle, rapid head growth, sunset sign, and vomiting are described.9 – 11 Computed tomography (CT) scans reveal extremely large tumors with prominent cyst formation and intense contrast enhancement (see Figure 2). Magnetic resonance imaging (MRI) shows hypointense cystic areas with isodense peripheral tumor. The cystic areas have increased T2weighted signal while the solid tumor tends to enhance with gadolinium.6,12 Treatment
All patients with DIGs reported in the literature have had attempted surgical resection. The extent of resection is not always noted. The postoperative morbidity and mortality are high, probably because of the age of the patients and the size of the tumors. Seizures, disseminated intravascular
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
UNCOMMON PEDIATRIC BRAIN TUMORS
799
Craniopharyngioma Low grade Other Supratentorial astrocytomas High grade Brain stem glioma
Pineal tumors 0.5–2%
Ependymoma Medulloblastoma Cerebellar astrocytoma Figure 1 Uncommon pediatric brain tumors.
Table 1 Pediatric brain tumors.
Desmoplastic Tumors Desmoplastic infantile ganglioglioma (DIG)/desmoplastic cerebral astrocytoma of infancy (DCAI) Pleomorphic xanthoastrocytoma (PXA) Gliofibroma Choroid plexus carcinoma (CPC) Dysembryoplastic neuroepithelial tumor (DNT) Pineoblastoma Trilateral retinoblastoma Ependymoblastoma Malignant rhabdoid tumor (MRT) Primary melanocytic tumors
coagulation, intracranial hemorrhage, hemiparesis, and sepsis have all been reported in the postoperative period.6 Adjuvant therapy is not recommended when a gross total resection can be achieved. Many patients are now documented without progression years later in this setting.13 In the setting of gross residual disease, the recommendation is less clear. Radiation would be of a large volume because of the typical size of these tumors. As the children are extremely young, effects on growth and development would likely be devastating and thus this mode of therapy is definitively not recommended. Several patients have been treated with cisplatin, etoposide, vincristine, and cyclophosphamide according to the Pediatric Oncology Group (POG) protocol for infant brain tumors. None have progressed.6 An additional patient is reported to have responded to vincristine and carboplatinum after incomplete resection but she ultimately died of disease progression.14 Some authors have recommended chemotherapy for those patients with gross residual disease,6 while more recently others have recommended observation until progression.7,14 – 16
Pleomorphic Xanthoastrocytomas Pathology
Pleomorphic xanthoastrocytomas (PXA) are thought by some to be related to the superficial cerebral astrocytoma or
Figure 2 Desmoplastic infantile ganglioglioma: contrast enhanced CT scan showing a large frontal lesion resulting in a cranial bulge overlying the tumor.
DCAI because of the superficial cerebral location with leptomeningeal involvement.17 Pleomorphic cells, some spindlelike, are seen on light microscopy (see Figure 3). The tumors are also desmoplastic and cystic but the astrocytes are xanthomatous. In some cases, lymphocytes and plasma cells are prominent. GFAP stains are positive. Lipid droplets are confirmed by electron microscopy.18 Low Ki67 indices are reported in most cases.19 Generally, mitoses are rare and necrosis is not present.20 In lesions with significant mitoses and/or areas of necrosis, the term PXA with anaplastic features is recommended. Molecular genetic evaluations have failed to identify consistent abnormalities but unlike typical cerebral astrocytomas they lack TP53 gene mutations.21 Clinical Features
PXAs were first reported in a group of 12 patients with an average age of 12 years. They were described as superficial tumors with leptomeningeal involvement.15 Vomiting and headaches are common, but seizures are the most common symptoms leading to diagnosis.20,22 MRI scans show lesions with borders distinct from the surrounding brain (see Figure 4). Cysts are often noted and contrast enhancement is prominent. All the original case reports and most of the subsequent ones describe tumors with a supratentorial location.15 Most were in the temporal lobe, but thalamic and suprasellar tumors have also been described.22
800
PEDIATRIC MALIGNANCIES
Figure 4 Pleomorphic xanthoastrocytoma: MRI showing a right temporal lobe lesion with associated cystic structures. Figure 3 Pleomorphic xanthoastrocytoma: closely packed pleomorphic cells.
Positron emission tomography (PET) scanning has shown a wide variation in uptake among patients, but one study of three patients suggests that increased uptake may be a marker for more aggressive tumors.23 Treatment
PXAs have been treated with surgery alone in most cases with good outcomes. In most patients, seizures reduce significantly in frequency or resolve completely after resection.19 Radiation therapy has been used in patients for whom surgical resection was not possible,15 but its use remains controversial.24 Short-course chemotherapy with vincristine and carboplatinum was used successfully to reduce vascularity of a PXA with anaplastic features in a patient whose surgical resection was aborted because of excessive blood loss.25 Otherwise, little role for chemotherapy exists. There does appear to be a subset of patients (20%) who are prone to recurrence or malignant progression. Necrosis has been associated with recurrence in some evaluations26 but the mitotic
index seems to be the most consistent predictor of aggressive clinical behavior.23,27
Gliofibroma Pathology
Gliofibroma, or desmoplastic glioma, first reported by Friede in 1978,28 may also be related to the other desmoplastic tumors (see Table 2). Some pathologists feel that they are not a distinct entity but others argue that since they lack dural attachment, do not have a bulky extra-axial component, but do infiltrate along perivascular spaces unlike the typical desmoplastic tumors of infancy, they should be classified separately.29 Microscopically these tumors show intermingling of astroglial cells with collagen fibers. GFAP-positive glial cells are mildly pleomorphic. Since the tumors have been found to be positive for S-100, some have theorized that Schwann cells rather than fibroblasts are responsible for the collagen production.30 Neuronal cells, mitoses, vascular proliferation, and necrosis are not seen.31 Electron microscopy shows direct apposition of collagen fibers with glial processes.28
UNCOMMON PEDIATRIC BRAIN TUMORS
801
Table 2 Features of desmoplastic tumors.
Histopathology Tumor type
Age
Size
DCAI
<18 months
Huge
DIG
<18 months Most <6 months
Huge
PXA
Children
Variable
Gliofibroma
<20 years
Variable
Spindle cells
Neurons
Mitoses
Necrosis
Xanthoma
GFAP + S-100 +
Good
+
−
±
−
Rare
+
Good
+
+
+
Rare
−
+
Usually good
+
−
±
−
+
+
Usually good
+
−
−
−
−
+
Location
Prognosis
Frontal Parietal Frontal Parietal Temporal Temporal Frontal Parietal Thalamus Frontal Parietal Spinal
Clinical Features
The original report of a gliofibroma was in a 3-year-old girl with a tumor in the medulla, who presented with apnea and seizures.28 Some infants are reported, but children and young adults have also often been diagnosed with these neoplasms. The lesions may occur throughout the central nervous system (CNS), including the spinal cord. Symptoms at diagnosis reflect these locations, but infants with brain lesions typically present with an enlarging head circumference and other signs of hydrocephalus.29 CT scans show enhancing tumors. MRIs generally show low or isointensity on T1-weighted imaging with enhancement. The intensity on T2 weighting is quite variable and may reflect the mixed components of the tumor. Occasionally, gliofibromas behave in a malignant fashion but, like the desmoplastic tumors, they are usually associated with a good prognosis.
have areas with loss of differentiated papillary architecture, malignant cellular features such as mitoses, and nuclear pleomorphism.33 Immunohistochemical stains with epithelial markers, such as cytokeratin, are positive.34 Clinical Features
Most patients have received only surgical treatment and have done well.29,31 However, only 37% of pediatric patients are reported in one recent review to have had a complete resection. The role of radiation is limited because of the young age of many patients and several reports of poor response. In one review, four of five patients treated with radiation alone died of progressive disease. Chemotherapy with vincristine and carboplatinum resulted in a 50% decrease in the size of a suprasellar gliofibroma diagnosed in a 4-monthold infant. The patient is reported without recurrent disease more than 2 years after treatment.30 Another patient with a lower medulla lesion received both radiation and chemotherapy without any surgical resection and died 3 months after presentation.31
The largest series, reported by Ellenbogen et al., is of 40 patients diagnosed over a 46-year period in Boston.31 The patients’ ages ranged from 6 days to 16 years, with a median of 10 months. The lateral ventricles were involved 75% of the time; 10% had tumors in the third ventricle, and 15% in the fourth ventricle and cerebellopontine angle.31 Other series present similar sites of disease with occasional reports of thalamic or posterior fossa primaries.30 Presenting symptoms include signs of increased intracranial pressure, lethargy, vomiting, headache, and irritability.30,35 Infants may present with macrocephaly.36 Symptoms are, of course, site specific, and occasionally focal deficits such as cranial nerve palsies and hemiparesis are noted.30 Unfortunately, CPCs are often not diagnosed until they are extremely large.37 Radiographically, these tumors are large intraventricular masses, which frequently are noted to invade the adjacent cerebral hemisphere.30 Hydrocephalus is very common. Noncontrast CT shows an isodense-to-hyperdense heterogeneous mass. Intense enhancement is present on administration of contrast. Necrosis, intratumoral cysts, and calcifications may be noted.35,38 MRI images with T1 are hypointense (see Figure 5). Often, areas of hemorrhage are noted.34,39 Intracranial40 and spinal41 metastases are reported commonly. Cerebrospinal fluid (CSF) positivity has been reported despite negative radiological evaluation of the craniospinal axis for metastatic disease. In one report, a patient was noted to have pulmonary metastases at recurrence.34
Choroid Plexus Carcinomas
Treatment
Treatment
Pathology
Primary choroid plexus tumors account for up to 2% of brain tumors in the pediatric age-group. Choroid plexus papillomas outnumber choroid plexus carcinomas (CPCs) four to one.32 Tumors are classified as CPCs if they
Gross total resection is the recommended treatment for patients with CPCs. Unfortunately, this is rarely achieved with a single surgery because of brain parenchymal extension and, most commonly, extreme vascularity. The average blood loss in one series was 138% of blood volume.34 Often, shunting is required for hydrocephalus. Several patients with
802
PEDIATRIC MALIGNANCIES
Figure 5 Choroid plexus carcinoma: extremely large (10 cm × 8 cm × 6.7 cm) lesion occupying the right lateral ventricle and extending through the brain parenchyma.
gross resection (3 out of 5 in a total series of 11) reported by Packer et al. are alive without further therapy at more than 1.5, 3.5, and 9.5 years, respectively.30 Survival has been found to correlate with gross total resection, by French investigators also. Berger et al. reported on 22 patients with an overall survival rate of 26%. However, those with a gross total resection had an 86% survival rate.42 Adjuvant therapy has been varied. Most patients have been treated with chemotherapy and/or craniospinal radiation. Treatments with “8 in 1” chemotherapy (one patient),30 ifosfamide/carboplatin/etoposide (four patients),34 cisplatin/vinblastine/bleomycin (one patient),43 vincristine/ lomustine (CCNU) (one patient),44 and cyclophosphamide/etoposide/vincristine/a platinum-containing agent (four patients)45 have shown some efficacy. Two of three children reported, who received cisplatin and etoposide after subtotal resection, had complete responses but eventually died from recurrence.46 Eleven patients were treated according to the infant POG protocol, which included cisplatin, etoposide, vincristine, and cyclophosphamide.38 Four patients had residual disease following surgery. Two had a partial response but eventually
progressed on chemotherapy. One is alive more than 5 years after subsequently receiving craniospinal radiation. Overall, almost 60% of children with gross total resection, with or without adjuvant therapy, are alive without progression. Only 3 of 11 children without gross total resection, who were treated with neuroaxis radiation, have remained disease-free. The largest series of treated patients, reported by Berger and colleagues,40 involved 22 patients diagnosed over an 11-year period. Seven patients had gross total resection followed by varying regimens of chemotherapy, including CCNU (one patient), etoposide and carboplatin (one patient), and a six-drug regimen (carboplatin, procarbazine, etoposide, cisplatin, vincristine, and cyclophosphamide) for slightly more than a year. One child was treated with craniospinal radiation. One patient died of local recurrence at 5 months. The remaining patients remain free of disease 5 to 85 months from completion of treatment. Eleven patients had only partial resection. One was treated with craniospinal radiotherapy and died from metastatic disease. Six were treated with the six-drug combination mentioned above, two with etoposide and carboplatin, one with etoposide, ifosfamide, and carboplatin, and one with carboplatin and ifosfamide, all with poor results. Two patients received, in addition, high-dose therapy with etoposide and thiotepa with peripheral blood stem cell rescue and achieved complete remission, but ultimately died of disease. Only one patient in this group remains free of disease. Responses to radiation therapy have been reported, but treatment for the entire neuroaxis is necessary, and the longterm effects on growth and development are substantial.31 In a review of the literature from 1966 to 1998, Wolff et al. found 207 patients reported with CPCs. Those who received radiation did better than those without radiation after complete and incomplete resections.47 Overall, the results for chemotherapy and/or radiation have been poor if surgical resection was not eventually achieved. Whether the improved survival with extensive surgery is related to the tumors being less invasive, and thus, by definition, being more surgically accessible, is theorized, but not known.38 Chemotherapy has been reported in some cases to reduce the vascularity even when stable disease is noted on neuroimaging, and may thus make gross total resection possible.34
Dysembryoplastic Neuroepithelial Tumors Pathology
Grossly, these tumors are often visible at the cortical surface. Microscopically, multiple cell types are seen, leading to a high degree of cellular polymorphism (see Figure 6). Astrocytes, oligodendrocytes, and neurons are present. A specific glioneuronal element is evident. These are bundles of axons lined by oligodendrocytes and neurons that are noted within an interstitial fluid. These structures are oriented perpendicular to the cortical surface in a columnar fashion.48 The tumor described originally by Daumas-Duport and colleagues49 would now be subtyped as the complex form of dysembryoplastic neuroepithelial tumor (DNT). The simple form is a tumor in which only the glioneuronal element is described.
UNCOMMON PEDIATRIC BRAIN TUMORS
803
images.51,54 Gadolinium enhancement occurs in only onethird of DNTs.57 On fluorodeoxyglucose (FDG) PET, DNTs are seen as hypometabolic lesions.58,59 DNTs were originally thought to occur only supratentorially. Two-thirds of the tumors are present in the temporal lobe, with one-third in the frontal lobe. A recent report has described four patients with caudate nucleus lesions.60 Single and multiple lesions in the cerebellum have also been reported.61,62 Treatment
Surgical resection is the only treatment necessary for DNTs. In many cases, only partial resection is needed to improve seizures. Many patients are seizure-free after surgery, and most have a significant reduction in seizure frequency. However, a recent report detailing the follow-up on 26 children identified with DNT at Toronto’s Hospital for Sick Children found that while seizure outcome was excellent in 22 children (85%) at 12 months, in longer follow-up only 16 (62%) remained seizure-free. Older age, longer duration of epilepsy, and residual tumor were all risk factors for poor seizure outcome.63 There is still no need for chemotherapy or radiation and reresection is recommended in the rare cases of recurrent tumor. Thus, identifications of DNTs by clinical features and pathologic evaluation is paramount to avoid the toxicities of further therapy.
Pineoblastomas Pathology
Pineoblastomas are considered a subset of cerebral primitive neuroectodermal tumors. Macroscopically, they are soft and friable, and infiltrate surrounding structures. They consist of sheets of densely packed small cells with a high nuclearto-cytoplasmic ratio. They are diffusely arranged, often with Homer –Wright rosettes. Mitoses and necrosis are common.64 Figure 6 Dysembryoplastic neuroepithelial tumor: higher power view of a cortical nodule showing microcysts.
Clinical Features
The original series of 39 patients reported by Daumas-Duport and colleagues49 remains the largest to date, although many more patients have now been described.50 – 56 All patients presented with partial complex seizures. A small percentage of the patients had, in addition, generalized seizures. Seizures occurred at least daily in most patients and were often drug resistant. Headache and papilledema were noted in some patients. Symptoms began at a mean age of 9 years, with a range of 1 to 19 years.49 The duration of symptoms is often quite long, in some cases decades, with the diagnosis being made after referral to an epilepsy surgery center.53 CT scans may be normal. Most show hypodense lesions after contrast, some with calcification, or cysts. Edema is not prominent. Often, the skull overlying the tumor is deformed. MRIs show lesions that are of low intensity on T1-weighted images and of high intensity on T2-weighted
Clinical Features
Young children are most often diagnosed with pineoblastomas. Patients present with a short history of symptoms related to increased intracranial pressure. Increasing head circumference and a bulging fontanelle are common in infants. Leptomeningeal dissemination is frequently present at diagnosis.65 Peritoneal metastases have been reported in two patients with ventriculo-peritoneal shunts.66 CT scans show large, homogeneous masses. They are usually hyperdense after contrast enhancement. These tumors are hypointense-to-isointense on T1-weighted MRI scans (see Figure 7). They generally show an increased signal on T2-weighted images and demonstrate homogeneous enhancement. There are no unique features to differentiate pineoblastomas from other tumors located in this area of the brain, by either CT or MRI.67 – 69 Treatment
Pineoblastoma is a highly malignant tumor. Complete surgical resection is unusual because of the location and the
804
PEDIATRIC MALIGNANCIES
autologous stem cell rescue. Nine patients, including the two infants and three who presented with metastatic disease, were alive at the time of the report at a median of 62 months from diagnosis.79
Trilateral Retinoblastoma Pathology
Trilateral retinoblastoma is a term first used by Bader and colleagues80 to describe a pineoblastoma diagnosed in a patient with bilateral retinoblastoma. The pathology is the same as that noted in classic pineoblastoma. Clinical Features
Figure 7 Pineoblastoma: MRI showing a mass centered on the pineal gland with subsequent obstruction of the third ventricle and hydrocephalus.
high degree of vascularity. CSF seeding is common and thus, in older children, craniospinal radiation is recommended.70 Unfortunately, this is a disease of young children in whom the side effects related to growth and development can be devastating. Thus, attempts to delay radiotherapy with aggressive chemotherapeutic protocols have been reported. Eleven patients were treated on the baby POG protocol with cisplatin, etoposide, cyclophosphamide, and vincristine, with very disappointing results. Only one patient had a partial response; while all patients were documented to have progressive disease 2–11 months from initiation of therapy. Failures were both at the local site and from leptomeningeal dissemination. All patients died, despite attempts to salvage them with other chemotherapy regimens or radiation, within 34 months, and 10 of 11 died less than 13 months from diagnosis. In older children, however, etoposide, vincristine, and cisplatin regimen followed by craniospinal radiation led to two long-term survivors in a group of seven patients.71 High-dose cyclophosphamide has been tried by several investigators,72,73 with benefit in some patients. Partial responses to melphalan have also been reported.74 Eight infants treated with “8 in 1” chemotherapy rapidly developed progressive disease, while 17 older patients randomized to CCNU, prednisone, and vincristine or “8 in 1” chemotherapy, in addition to radiation therapy, had a progression-free survival of 61% at 3 years.75 Radiation therapy, in other reports, has been beneficial in some patients, but the majority of patients still die from their disease.75,76 High-dose chemotherapy with autologous bone marrow transplant has been tried in seven patients77 in one study and in eight patients with recurrence in another study78 without any long-term benefit. However, this approach was successful in nine patients treated at the Duke University. Twelve patients were treated with surgery and induction chemotherapy. This was followed by craniospinal radiation with a boost to the pineal region in all but two patients, who were younger than 14 months at the time of diagnosis. Subsequently 11 patients received cytoxan and melphalan and one received busulfan and melphalan followed by
Trilateral retinoblastoma is found in approximately 3% of children with retinoblastoma.81 Holladay et al. reviewed 35 patients in the literature in 1991.82 Almost 70% of patients diagnosed with trilateral retinoblastoma have a family history of retinoblastoma. Other smaller series have reported a lower incidence of a known family history.83 Patients with trilateral retinoblastoma are generally diagnosed at an earlier age (7 months) than those with only bilateral retinoblastoma (15 months). The average time from the diagnosis of ocular tumors to the diagnosis of the intracranial malignancy was 32.6 months. Symptoms are those of elevated intracranial pressure.82 Occasional patients are reported in whom the intracranial tumors are not in the pineal gland but in sellar locations. Several patients with unilateral retinoblastoma have also been reported.81,84 In the last 10 years, more children are receiving chemotherapy for retinoblastoma in an attempt to avoid enucleation or external beam radiation. Of 240 children consecutively treated for newly diagnosed retinoblastoma at a major ocular oncology center, 142 (66%) received chemotherapy with carboplatin, etoposide, and vincristine while 72 (34%) received nonchemoreduction therapy. On the basis of metaanalysis of the prevalence of trilateral retinoblastoma, 5–15 patients with hereditary retinoblastoma were expected to develop an intracranial tumor. In the chemoreduction group, none of the 99 patients who were at risk developed an intracranial malignancy while in the nonchemoreduction group one patient did, a finding consistent with the expected frequency. This has suggested that this treatment may have a protective effect against this highly fatal malignancy.85 Treatment
The prognosis of patients who develop trilateral retinoblastoma is dismal. Patients without adjuvant treatment die within 2 months of diagnosis, usually of widely disseminated CNS metastases. These tumors are sensitive to radiation, and both the symptoms and the neuroimaging improve with treatment. However, the response in almost all cases is short-lived.83,86 There is one report87 of successful treatment of trilateral retinoblastoma with chemotherapy. Three patients were treated with vincristine, cyclophosphamide, and intrathecal triple therapy (methotrexate, hydrocortisone, and cytosine arabinoside), with one complete and two partial responses. All three patients were treated when their tumors were
UNCOMMON PEDIATRIC BRAIN TUMORS
asymptomatic and diagnosed on routine screening. None had metastatic disease. The patients with partial responses also received craniospinal radiotherapy, and all were doing well at the time of publication, 8 years, 33 months, and 12 months after diagnosis. At the Children’s Hospital of Los Angeles, one patient diagnosed with routine screening had gross total resection of the pineal mass followed by conventional chemotherapy with cisplatinum, etoposide, cyclophosphamide, vincristine, carboplatin, and thiotepa. Local therapy consisted of laser photocoagulation. During complete remission, the patient received high-dose cyclophosphamide, etoposide, and thiotepa followed by autologous bone marrow transplant and remains disease-free more than 45 months from diagnosis without radiation.88 There is clear evidence in larger series that patients have longer survival times if they are treated prior to CNS dissemination. Thus, routine screening is recommended for those patients at risk for the development of trilateral retinoblastoma until the age of 4 years.81,83
Ependymoblastoma Pathology
805
CT scans show large, well-circumscribed masses with heterogeneous enhancement, prominent mass effect, necrosis, and often, calcification without hemorrhage. MRI descriptions have been only rarely reported, but large, clearly demarcated masses with heterogeneous enhancement on T1 images are described.91 Treatment
Surgical resection has been attempted in a majority of patients. Often, the extent of resection is not described,89 but 45% (10 of 22) of cases reported by Cervoni et al.93 were able to achieve a gross total resection. Because the tumors are soft and friable, resection was difficult. Postoperative craniospinal radiotherapy was performed in 17 of the 22 cases. The largest report of 12 patients89 notes that three patients received chemotherapy but gives no further details. Chemotherapy using vincristine and nimustine (ACNU) has been tried.93 A case report of an HIV-positive 13-yearold girl treated with resection of nonmetastatic parietooccipital mass followed by craniospinal radiation and “8 in 1” chemotherapy reports no recurrence at 6 years from diagnosis.95 Unfortunately, only one other patient reported in the literature had a long course, with three surgeries over a 9-year period. Tumor recurrence is the cause of death in all other reported cases. Both local recurrence and CSF dissemination are seen. One case was reported to have lung metastases at the time of local recurrence. Survival was prolonged with adjuvant therapy.93 Patients generally receive treatment similar to that of other high-risk primitive neuroectodermal tumors with maximal resection, craniospinal radiation, and aggressive chemotherapy.
Grossly, these tumors are usually extremely large but well circumscribed. Cystic changes are occasionally noted. These tumors are extremely cellular, with poorly differentiated cells containing round or oval nuclei. The most characteristic features are ependymal rosettes and tubules lined by columnar cells. Mitoses are very common.89,90 The ependymoblastoma is felt to be a type of primitive neuroectodermal tumor that has some ependymal differentiation, and is not equivalent to an ependymoma that has undergone anaplastic change.91 Anaplastic ependymomas are noted for increased cellular density, pleomorphism, giant cells, and multinucleated cell forms.89 Despite their differences, some anaplastic ependymomas have been misdiagnosed as ependymoblastomas by less-experienced pathologists.92 Immunohistologic study of ependymoblastomas shows positivity to vimentin and S-100. Electron microscopy demonstrates cells with large nuclei and little cytoplasm. Junctional complexes and sparse rosettes are noted.90
Rhabdoid tumors of the kidney were first reported as a variant of Wilms’ tumor with a poor prognosis.96 Subsequently, a number of cases of primary tumors occurring outside the kidney were reported, with a CNS location being a frequent site. The prevalence of this tumor remains to be defined but may actually be increasing as pediatric neuropathologists are becoming more adept at distinguishing it from typical primitive neuroectodermal tumors.
Clinical Features
Pathology
Ependymoblastoma is most often found in young children, with more than 80% of reported cases presenting by age 5 years. Occasional congenital cases have been reported. On the other end of the spectrum, several adult cases have been reported. The median age at presentation is 3 years.93 Most patients present with signs of elevated intracranial pressure. Infants have macrocephaly. Older children may present with gait or visual disturbance. Vomiting is common. A majority of patients present with less than 1 month of symptoms prior to diagnosis.89,93 Most tumors involve the ventricular cavities and have a supratentorial location, although infratentorial locations and disease in both compartments are described. One case of primary leptomeningeal ependymoblastoma diagnosed at autopsy in a 17-year-old has been reported.94
These tumors have distinctive features, which include polygonal cells with oval nuclei containing prominent centrally located nucleoli. Nest of rhabdoid cells are often surrounded by typical regions of primitive neuroectodermal tumor. One unique feature is the occurrence of an epithelial and/or mesenchymal component in addition to the rhabdoid component.97 Necrosis is prominent and the mitotic rate is quite high. Scattered cells contain a cytoplasmic hyaline globular inclusion adjacent to the nucleus, which stains positively for vimentin.98 Electron microscopy shows intracytoplasmic whorling filaments,98 – 100 which correspond to the inclusions noted by light microscopy. Initially, cytogenetic evaluation revealed abnormalities in chromosome 22.101,102 Further evaluation narrowed the abnormality to the
Atypical Teratoid/Rhabdoid Tumors (AT/RT)
806
PEDIATRIC MALIGNANCIES
22q11 region and subsequently identified a tumor suppressor gene INI1.103,104 Both germ line and acquired mutations have been identified at such a high rate that in a workshop on these tumors sponsored by the National Cancer Institute and the Pediatric Brain Tumor Foundation, the decision was made to define any tumor with features suggestive of medulloblastoma or primitive neuroectodermal tumor with an INI1 mutation documented as an atypical teratoid/rhabdoid tumor (AT/RT).105 Immunohistochemical staining with an INI1 antibody has also been shown to be consistent with the molecular phenotype. Most routine primitive neuroectodermal tumors show expression while AT/RTs do not.106 Clinical Features
AT/RT is generally a disease of infants and young children, with an age range at diagnosis reported from 3 months to 21 years. The median age in the report of 42 patients enrolled in a registry is 24 months.107 Most children present with lethargy and vomiting. Focal neurologic signs, such as hemiparesis, and cranial nerve palsies are also common.98,99,108 – 110 Approximately 50% arise in the posterior fossa, although other sites include the suprasellar region, pineal region, spinal cord, and supratentorial sites. CT scans demonstrate large tumors that are often hyperdense prior to contrast administration. Contrast enhancement is usually intense. On T1-weighted MRIs, the lesion is isointense with hyperintense foci as a result of hemorrhage. The T2 appearance is heterogeneous secondary to the mixed tumor cellularity, hemorrhage, necrosis, and cysts.98 These tumors have a tendency to disseminate widely. Spine MRIs should be done, and lumbar punctures are recommended if the condition of the patient is stable.101 – 103 Treatment
Total resection of these tumors is usually difficult because of their size and the young age of the patients. In addition, a significant percentage are metastatic at diagnosis. Most reports are of subtotal resection or biopsy only. However, the long-term survivors who are reported have generally undergone gross total resection.107 Responses to aggressive chemotherapy regimens are reported, but the length of response is generally short. Treatment regimens include cisplatin and adriamycin,109 intrathecal methotrexate,111 cisplatin/etoposide/vincristine/cytoxan along the lines of the POG infant protocol,99,109 “8 in 1” chemotherapy,109 and Intergroup Rhabdomyosarcoma Study-III Regimen 36 with intrathecal chemotherapy.108 One patient who had a nonmetastatic lesion was treated with gross total excision, local radiotherapy, and chemotherapy with vincristine, doxorubicin, actinomycin D, and triple intrathecal therapy, and is well more than 5 years from diagnosis.112 Most of the reported patients who survived more than 2 months from diagnosis were treated with radiotherapy, despite the fact that some were very young. Doses ranged from 4140 cGy to 6255 cGy.108 A recent review of 31 patients from the St. Jude Children’s Research Hospital diagnosed over a 19-year period confirmed the poor prognosis in 22 patients
under 3 years of age treated with aggressive alkylator therapy and no radiation. Those 22 patients had an event-free survival of 11% at 2 years. In contrast, of the nine patients older than 3 years (median age 3.9 years) who were treated with both aggressive alkylator therapy and in all but two cases of radiation therapy, five remained free of disease at a median of 2.2 years from diagnosis. Three of the four older patients who relapsed were salvaged with ifosfamide, carboplatin, etoposide chemotherapy, and radiation. In a separate analysis of the two groups, there was a suggestion of an association between survival and the extent of surgical resection.113 Thirty-one percent of the patients enrolled on the registry received stem cell rescue as part of their therapy. Even with these aggressive interventions, the prognosis of AT/RT is dismal, with a median event-free survival of 10 months.107
Primary Melanocytic Lesions of the Central Nervous System Pathology
Melanocytic tumors of the CNS occurring in children may be classified as diffuse with meningeal melanocytosis or mass lesions of primary melanoma. Diffuse melanocytic lesions are seen as a black replacement of the subarachnoid area or just a dusky clouding of the meninges. Single melanomas are black, brown, or blue lesions that may be encapsulated. Occasionally, they are not pigmented. Microscopic evaluation of melanocytosis shows uniform nevoid polygonal cells arranged in sheets. Primary melanoma cells usually show mitoses, giant nuclei with prominent nucleoli, necrosis, and foci of hemorrhage, with a background of large bizarre cells.114 CSF showing round cells with abundant cytoplasm and ovoid nuclei with characteristic irregular projections from the cell body may be all that is needed to diagnose melanocytosis. Electron microscopy shows cells without junctions containing melanosomes at varying stages of development.115 These cells are thought to originate from neural crest elements.116 Clinical Features
The largest series of children with leptomeningeal melanocytosis is reported by Allcutt et al.117 Eight patients were diagnosed over a 26-year period, with a mean age of 4.9 years (range 1.3 to 13). Five of the patients had tumor obliteration of the basal cistern, resulting in hydrocephalus and elevated intracranial pressure. Their symptoms ranged from developmental regression to seizures, cranial nerve palsies, vomiting and, in two cases, coma.117 Giant congenital nevomelanocytic nevi, usually in the midline on the head or neck, are associated in 25% of patients with melanocytic CNS lesions. This association has been called neurocutaneous melanocytosis (NM). The criteria for diagnosis include: (i) a giant pigmented nevus of at least 6 cm in neonates or 20 cm in adults, or at least three separate nevi, (ii) no evidence of cutaneous melanoma, since the CNS is a common site for metastasis of malignant skin lesions, and (iii) absence of malignant melanoma in organs other than
UNCOMMON PEDIATRIC BRAIN TUMORS
Figure 8 Melanocytic lesions: MRI with multiple widely disseminated lesions that are bright on T1 imaging.
the meninges.118 NM has been reported in association with the Dandy-Walker complex.119 CT scan may show thickening and enhancement of the meninges when diffuse disease is present or it may be normal. In contrast to most other cerebral tumors, melanocytic lesions are usually bright on T1-weighted MRI scans and dark on T2 images (see Figure 8). This is variable and may be related to the amount of hemorrhage within the tumor.114,117 Treatment
Solitary melanomas may be treated with surgical removal, but the lesions may be locally infiltrative and hemorrhage is common. One long-term survivor is reported with resection followed by Dacarbazine and radiotherapy.120 Surgery is not a viable option for patients with diffuse leptomeningeal disease, except for shunt placement. Although the shunts are necessary for relief of severe symptoms due to hydrocephalus, peritoneal metastases have been reported. Often, patients die of fulminant disease prior to the initiation of any therapy. Two patients treated with cisplatin, DTIC, and craniospinal radiation had survivals of 2 and 3 years. One patient treated with intrathecal methotrexate survived only for 5 months. Overall, survival is dismal, especially with diffuse involvement.117
REFERENCES 1. Grovas A, et al. The National Cancer Data Base report on patterns of childhood cancers in the United States. Cancer 1997; 80(12): 2321 – 32. 2. Duffner PK, et al. Survival of children with brain tumors: SEER program 1973 – 1980. Neurology 1986; 36: 597 – 601. 3. Taratuto A, et al. Superficial cerebral astrocytoma attached to dura. Report of six cases in infants. Cancer 1984; 54: 2505 – 12. 4. de Chadarevian JP, Pattisapu JV, Faerber EN. Desmoplastic cerebral astrocytoma of infancy. Cancer 1990; 66: 173 – 9. 5. Vandenberg SR, et al. Desmoplastic supratentorial neuroepithelial tumors of infancy with divergent differentiation potential (‘desmoplastic infantile gangliogliomas’). Report on 11 cases of a distinctive embryonal tumor with favorable prognosis. J Neurosurg 1987; 66: 58 – 71.
807
6. Duffner PK, et al. Desmoplastic infantile gangliogliomas: an approach to therapy. Neurosurgery 1994; 34(4): 583 – 9. 7. Tamburrini G, et al. Desmoplastic infantile ganglioglioma. Childs Nerv Syst 2003; 19: 292 – 7. 8. Kuchelmeister K, et al. Desmoplastic ganglioglioma: report of two non-infantile cases. Acta Neuropathol 1993; 85: 199 – 204. 9. Rothman S, et al. Desmoplastic infantile ganglioma. Acta Oncol 1997; 36(6): 655 – 7. 10. Paulus W, et al. Desmoplastic supratentorial neuroepithelial tumors of infancy. Histopathology 1992; 21: 43 – 9. 11. Louis DN, et al. Desmoplastic cerebral astrocytomas of infancy: a histopathologic, immunohistochemical, ultrastructural, and molecular genetic study. Hum Pathol 1992; 23: 1402 – 9. 12. Tenreiro-Picon OR, et al. Desmoplastic infantile ganglioglioma: CT and MRI features. Pediatr Radiol 1995; 25: 540 – 3. 13. Sugimaya K, et al. Good clinical course in infants with desmoplastic cerebral neuroepithelial tumor treated by surgery alone. J Neurooncol 2002; 59: 63 – 9. 14. DeMunnynck K, et al. Desmoplastic infantile ganglioglioma: a potentially malignant tumor? Am J Surg Pathol 2002; 26(11): 1515 – 22. 15. Bachli H, et al. Therapeutic strategies and management of desmoplastic infantile ganglioglioma: two case reports and literature overview. Childs Nerv Syst 2003; 19(5 – 6): 359 – 66. 16. Abdollahzadeh M, et al. Benign cerebellar astrocytoma in childhood: experience at the Hospital for Sick Children 1980 – 1992. Childs Nerv Syst 1994; 10(6): 380 – 3. 17. Rushing EJ, Rorke LB, Sutton L. Problems in the nosology of desmoplastic tumors of childhood. Pediatr Neurosurg 1993; 19: 57 – 62. 18. Kepes JJ, Rubinstein LJ, Eng LF. Pleomorphic xanthocytoma: a distinctive meningocerebral glioma of young subjects with relatively favourable prognosis. A study of 12 cases. Cancer 1979; 44: 1839 – 52. 19. Im S, et al. Pleomorphic xanthoastrocytoma: a developmental glioneuronal tumor with prominent glioproliferative changes. J Neurooncol 2004; 66: 17 – 27. 20. Kros JM, Vecht CJ, Stefanko SZ. The pleomorphic xanthoastrocytoma and its differential diagnosis: a study of five cases. Hum Pathol 1991; 22: 1128 – 35. 21. Kaulich K, et al. Genetic alterations commonly found in diffusely infiltrating cerebral gliomas are rare or absent in pleomorphic xanthoastrocytomas. J Neuropathol Exp Neurol 2002; 61(12): 1092 – 9. 22. Ng THK, Fung CF, Ma LT. The pathological spectrum of desmoplastic infantile gangliogliomas. Histopathology 1990; 16(3): 235 – 41. 23. Etzl MM Jr, et al. Positron emission tomography in three children with pleomorphic xanthoastrocytoma. J Child Neurol 2002; 17(7): 522 – 7. 24. Fouladi M, et al. Pleomorphic xanthoastrocytoma: favorable outcome after complete surgical resection. Neuro-oncology 2001; 3(3): 184 – 92. 25. Cartmill M, et al. The use of chemotherapy to facilitate surgical resection in pleomorphic xanthoastrocytoma: experience in a single case. Childs Nerv Syst 2001; 17: 563 – 6. 26. Papahill PA, Ramsay DA, Del Maestro RF. Pleomorphic xanthoastrocytoma: case report and analysis of the literature concerning the efficacy of resection and the significance of necrosis. Neurosurgery 1996; 38: 822 – 8. 27. Giannini C, et al. Pleomorphic xanthoastrocytoma: what do we really know about it? Cancer 1999; 85: 2033 – 45. 28. Friede RL. Gliofibroma. A peculiar neoplasia of collagen forming glia-like cells. J Neuropathol Exp Neurol 1978; 37: 300 – 13. 29. Matsumura A, et al. Cervical intramedullary gliofibroma in a child. Pediatr Neurosurg 2002; 36: 105 – 10. 30. Suarez CR, et al. Carboplatinum and vincristine chemotherapy for central nervous system gliofibroma: case report and review of the literature. J Pediatr Hematol Oncol 2004; 26(11): 756 – 60. 31. Prayson RA. Gliofibroma: a distinct entity or a subtype of desmoplastic astrocytoma? Hum Pathol 1996; 27: 610 – 13. 32. Packer RJ, et al. Choroid plexus carcinoma of childhood. Cancer 1992; 69: 580 – 5. 33. Ellenbogen RG, Winston KR, Kupsky WJ. Experimental and clinical studies: tumors of the choroid plexus in children. Neurosurgery 1989; 25: 327 – 35.
808
PEDIATRIC MALIGNANCIES
34. Gianella-Borradori A, et al. Choroid plexus tumors in childhood: response to chemotherapy, and immunophenotypic profile using a panel of monoclonal antibodies. Cancer 1992; 69: 809 – 16. 35. Boyd MC, Steinbok P. Choroid plexus tumors: problems in diagnosis and management. J Neurosurg 1987; 66: 800 – 7. 36. St Clair SK, et al. Current management of choroid plexus carcinoma in children. Pediatr Neurosurg 1991 – 1992; 17: 225 – 33. 37. Geerts BY, et al. Choroid plexus carcinoma: a report of two cases and review of the literature. Neuropediatrics 1996; 27: 143 – 8. 38. Pascual-Castroviejo I, et al. Childhood choroid plexus neoplasms: a study of 14 cases less than 2 years old. Eur J Pediatr 1983; 140: 51 – 6. 39. Vazquez E, et al. Magnetic resonance imaging of fourth ventricular choroid plexus neoplasms in childhood. Pediatr Neurosurg 1991 – 1992; 17: 48 – 52. 40. Duffner PK, et al. Postoperative chemotherapy and delayed radiation in infants and very young children with choroid plexus carcinomas. Pediatr Neurosurg 1995; 22: 189 – 96. 41. Bennedbaek O, Hamilton Therkildsen M. Choroid plexus carcinoma: report of a case with metastases within the central nervous system. Acta Oncol 1990; 29(2): 241 – 3. 42. Berger C, et al. Choroid plexus carcinomas in childhood: clinical features and prognostic factors. Neurosurgery 1998; 42(3): 470 – 5. 43. Maria BL, et al. Response of a recurrent choroid plexus tumor to combination therapy. J Neurooncol 1985; 3: 259 – 62. 44. Pierga JY, et al. Carcinoma of the choroid plexus: a pediatric experience. Med Pediatr Oncol 1993; 21: 480 – 7. 45. Chow E, et al. Pediatric choroids plexus neoplasms. Int J Radiat Oncol Biol Phys 1999; 44(2): 249 – 54. 46. Allen J, et al. Choroid plexus carcinoma: responses to chemotherapy alone in newly diagnosed young children. J Neurooncol 1992; 12: 69 – 74. 47. Wolff JEA, et al. Choroids plexus tumours. Br J Cancer 2002; 87: 1086 – 91. 48. Daumas-Duport C, Lantos P. Dysembryoplastic neuroepithelial tumours. In Kleihues CW (ed) Pathology and Genetics of Tumours of the Nervous System. Lyon, France: International Agency for Research on Cancer, 1997: 73 – 4. 49. Daumas-Duport C, et al. Dysembryoplastic neuroepithelial tumor: a surgically curable tumor of young patients with intractable partial seizures: report of thirty-nine cases. Neurosurgery 1988; 23(5): 545 – 56. 50. Zentner J, et al. Surgical treatment of neoplasms associated with medically intractable epilepsy. Neurosurgery 1997; 41: 378 – 87. 51. Raymond AA, et al. Dysembryoplastic neuroepithelial tumour features in 16 patients. Brain 1994; 117: 461 – 75. 52. Hirose T, et al. Dysembryoplastic neuroepithelial tumor (DNT): an immunohistochemical and ultrastructural study. J Neuropathol Exp Neurol 1994; 53(2): 184 – 95. 53. Prayson RA, Estes ML, Morris HH. Coexistence of neoplasia and cortical dysplasia in patients presenting with seizures. Epilepsia 1993; 34(4): 609 – 15. 54. Vali AM, Clarke MA, Kelsey A. Dysembryoplastic neuroepithelial tumour as a potentially treatable cause of intractable epilepsy in children. Clin Radiol 1993; 47: 255 – 8. 55. Morris HH, et al. Chronic intractable epilepsy as the only symptom of primary brain tumor. Epilepsia 1993; 34(6): 1038 – 43. 56. Prayson RA, Estes ML. Dysembryoplastic neuroepithelial tumor. Am J Clin Pathol 1992; 97: 398 – 401. 57. Lanzieri CF, et al. Neuroradiology case of the day: case I: dysembryoplastic neuroepithelial tumor. Am J Radiol 1997; 169: 293 – 8. 58. Lee DY, et al. Dysembryoplastic neuroepithelial tumor: radiologic findings (including PET, SPECT, and MRS) and surgical strategy. J Neurooncol 2000; 47: 167 – 74. 59. Kaplan AM, et al. Positron emission tomography using [18 F] fluorodeoxyglucose and [11 C] L-methionine to metabolically characterize dysembryoplastic neuroepithelial tumors. J Child Neurol 1999; 14(10): 673 – 7. 60. Cervera-Pierot P, et al. Dysembryoplastic neuroepithelial tumors located in the caudate nucleus area: report of four cases. Neurosurgery 1997; 40(5): 1065 – 70.
61. Kuchelmeister K, et al. Dysembryoplastic neuroepithelial tumor of the cerebellum. Acta Neuropathol 1995; 89: 385 – 90. 62. Leung SY, et al. Dysembryoplastic neuroepithelial tumor, a tumor with small neuronal cells resembling oligodendro-glioma. Am J Surg Pathol 1994; 18: 604 – 14. 63. Nolan MA, et al. Dysembryoplastic neuroepithelial tumors in childhood: long term outcome and prognostic features. Neurology 2004; 62(12): 2270 – 6. 64. Mena H, Scheithauer BW, Nakazato Y. Pineoblastoma. Pathology and Genetics of Tumours of the Nervous System. Lyon, France: International Agency for Research on Cancer, 1997: 84 – 5. 65. Duffner PK, et al. Lack of efficacy of postoperative chemotherapy and delayed radiation in very young children with pineoblastoma. Med Pediatr Oncol 1995; 25: 38 – 44. 66. Gururangan S, et al. Peritoneal metastases in two patients with pineoblastoma and ventriculo-peritoneal shunts. Med Pediatr Oncol 1994; 22(6): 417 – 20. 67. Chang T, et al. CT of pineal tumors and intracranial germ-cell tumors. AJNR Am J Neuroradiol 1989; 10(5): 1039 – 44. 68. Tien RD, Barkovick AJ, Edwards MS. MR imaging of pineal tumors. AJR Am J Roentgenol 1990; 155(1): 143 – 51. 69. Satoh H, et al. MRI of pineal region tumors: relationship between tumours and adjacent structures. Neuroradiology 1995; 37(8): 624 – 30. 70. Duffner PK, Cohen ME. Primitive neuroectodermal tumors. In Vinken BG (ed) Handbook of Clinical Neurology: Neuro-Oncology, Part II: Gliomas and Other Primary Tumors of the Brain and Spinal Cord. Amsterdam, New York: Elsevier Science, 1997: 211 – 28. 71. Ghim TT, et al. Response to neoadjuvant chemotherapy in children with pineoblastoma. Cancer 1993; 72(5): 1795 – 800. 72. Ashley DM, et al. Treatment of patients with pineoblastoma with high dose cyclophosphamide. Med Pediatr Oncol 1996; 26(6): 387 – 92. 73. Allen JC, Helson L, Jereb B. Preradiation chemotherapy for newly diagnosed childhood brain tumors. Cancer 1983; 52: 2001 – 6. 74. Friedman HS, et al. Phase II treatment of medulloblastoma and pineoblastoma with melphalan: clinical therapy based on experimental models of human medulloblastoma. J Clin Oncol 1989; 7(7): 904 – 11. 75. Jakacki RI, et al. Survival and prognostic factors following radiation and/or chemotherapy for primitive neuroectodermal tumors of the pineal region in infants and children; a report of the Children’s Cancer Group. J Clin Oncol 1995; 13(6): 1377 – 83. 76. Schild SE, et al. Pineal parenchymal tumors. Clinical, pathologic, and therapeutic aspects. Cancer 1993; 72(3): 870 – 80. 77. Graham ML, et al. High-dose chemotherapy with autologous stem cell rescue in patients with recurrent and high-risk pediatric brain tumors. J Clin Oncol 1997; 15(5): 1814 – 23. 78. Broniscer A, et al. High-dose chemotherapy with autologous stemcell rescue in the treatment of patients with recurrent non-cerebellar primitive neuroectodermal tumors. Pediatr Blood Cancer 2004; 42(3): 261 – 7. 79. Gururangan S, et al. High-dose chemotherapy with autologous stem-cell rescue in children and adults with newly diagnosed pineoblastomas. J Clin Oncol 2003; 21(11): 2187 – 91. 80. Bader JL, et al. Trilateral retinoblastoma. Lancet 1980; 2: 582 – 3. 81. DePotter P, Shields CL, Shields JA. Clinical variations of trilateral retinoblastoma. J Pediatr Ophthalmol Strabismus 1994; 31: 26 – 31. 82. Holladay DA, et al. Clinical presentation, treatment, and outcome of trilateral retinoblastoma. Cancer 1991; 67: 710 – 15. 83. Amoaku WM, et al. Trilateral retinoblastoma: a report of five patients. Cancer 1996; 78: 858 – 63. 84. Bejjani GK, et al. Association of a suprasellar mass and intraocular retinoblastoma: a variant of pineal trilateral retinoblastoma? Pediatr Neurosurg 1996; 25: 269 – 75. 85. Shields CL, et al. Chemoreduction for retinoblastoma may prevent intracranial neuroblastic malignancy (trilateral retinoblastoma). Arch Ophthalmol 2001; 119(9): 1269 – 72. 86. Johnson DL, et al. Trilateral retinoblastoma: ocular and pineal retinoblastomas. J Neurosurg 1985; 63: 367 – 70. 87. Nelson SC, et al. Successful therapy for trilateral retinoblastoma. Am J Ophthalmol 1992; 114(1): 23 – 9.
UNCOMMON PEDIATRIC BRAIN TUMORS 88. Jubran RF, et al. Approaches to treatment for extraocular retinoblastoma: Children’s Hospital Los Angeles experience. J Pediatr Hematol Oncol 2004; 26(1): 31 – 4. 89. Mork SJ, Rubinstein LJ. Ependymoblastoma. A reappraisal of a rare embryonal tumor. Cancer 1985; 55(7): 1536 – 42. 90. Cruz-Sanchez FF, et al. Differentiation in embryonal neuroepithelial tumors of the central nervous system. Cancer 1991; 67(4): 965 – 76. 91. Dorsay TA, et al. Ependymoblastoma: MR presentation: a case report and review of the literature. Pediatr Radiol 1995; 25(6): 433 – 5. 92. Robertson PL, et al. Survival and prognostic factors following radiation therapy and chemotherapy for ependymomas in children: a report of the Children’s Cancer Group. J Neurosurg 1998; 88(4): 695 – 703. 93. Cervoni L, et al. Ependymoblastoma: a clinical review. Neurosurg Rev 1995; 18(3): 189 – 92. 94. Wada C, et al. Primary leptomeningeal ependymoblastoma. Case report. J Neurosurg 1986; 64(6): 968 – 73. 95. Mandel M, et al. Ependymoblastoma in an HIV-positive hemophilic girl. Med Pediatr Oncol 1994; 23(5): 441 – 3. 96. Haas JE, et al. Ultrastructure of malignant rhabdoid tumor of the kidney. Hum Pathol 1981; 12: 646 – 57. 97. Packer RJ, et al. Atypical teratoid/rhabdoid tumor of the central nervous system: report on workshop. J Pediatr Hematol Oncol 2002; 24(5): 337 – 42. 98. Biggs PJ, et al. Perspectives in pathology: malignant rhabdoid tumor of the central nervous system. Hum Pathol 1987; 18: 332 – 7. 99. Caldemeyer KS, et al. Primary central nervous system malignant rhabdoid tumor: CT and MR appearance simulates a primitive neuroectodermal tumor. Pediatr Neurosurg 1994; 21: 232 – 6. 100. Hasserjian RP, et al. Clinicopathologic and cytogenetic analysis of malignant rhabdoid tumor of the central nervous system. J Neurooncol 1995; 25: 193 – 203. 101. Sawyer JR, et al. A new reciprocal trans-location (12,22)(q24.3; q11.2 – 12) in a malignant rhabdoid tumor of the brain. Cancer Genet Cytogenet 1998; 101(1): 62 – 7. 102. Beigel JA, et al. Monosomy 22 in rhabdoid or atypical tumors of the brain. J Neurosurg 1990; 73: 710 – 14. 103. Biegel JA, et al. Germ-line and acquired mutations of INI1 in atypical teratoid and rhabdoid tumors. Cancer Res 1999; 59: 74 – 9. 104. Biegel JA, et al. Alterations of the hSNF5/INI1 gene in central nervous system atypical teratoid/rhabdoid tumors and renal and extrarenal rhabdoid tumors. Clin Cancer Res 2002; 8: 3461 – 7.
809
105. Biegel JA, et al. The role of INI1 and the SWI/SNF complex in the development of rhabdoid tumors: meeting summary from the workshop on childhood atypical teratoid/rhabdoid tumors. Cancer Res 2002; 62: 323 – 8. 106. Judkins AR, et al. Immunohistochemical analysis of hSNF5/INI1 in pediatric CNS neoplasms. Am J Surg Pathol 2004; 28(5): 644 – 50. 107. Hilden JM, et al. Central nervous system atypical teratoid/rhabdoid tumor: results of therapy in children enrolled in a registry. J Clin Oncol 2004; 22(14): 2877 – 84. 108. Olson TA, et al. Successful treatment of disseminated central nervous system malignant rhabdoid tumor. Am J Pediatr Hematol Oncol 1995; 17(1): 71 – 5. 109. Agranovich AL, et al. Malignant rhabdoid tumor of the central nervous system with subarachnoid dissemination. Surg Neurol 1992; 37: 410 – 14. 110. Perilongo G, et al. Rhabdoid tumor of the central nervous system. Med Pediatr Oncol 1991; 19: 310 – 17. 111. Satoh H, et al. Primary malignant rhabdoid tumor of the central nervous system: case report and review of the literature. Surg Neurol 1993; 40: 429 – 34. 112. Weinblatt M, Kochen J. Letter to the editor: rhabdoid tumor of the central nervous system. Med Pediatr Oncol 1992; 20: 258. 113. Tekautz TM, et al. Atypical teratoid/rhabdoid tumors (ATRT): improved survival in children 3 years of age and older with radiation therapy and high-dose alkylator-based chemotherapy. J Clin Oncol 2005; 23(7): 1491 – 9. 114. Reyes-Mugica M, et al. Nevomelanocytic proliferations in the central nervous system of children. Cancer 1993; 72: 2277 – 85. 115. Jellinger KA, et al. Melanocytic lesions. Pathology and Genetics of Tumours of the Central Nervous System. Lyon, France: International Agency for Research on Cancer, 1997: 149 – 50. 116. Alwatban J, et al. MRI of leptomeningeal melanocytosis in a patient with neurofibromatosis. J Comput Assist Tomogr 1997; 21(1): 38 – 40. 117. Allcutt D, et al. Primary leptomeningeal melanoma: an unusually aggressive tumor in childhood. Neurosurgery 1993; 32(5): 721 – 9. 118. Kadonaga JN, Frieden IJ. Neurocutaneous melanosis: definition and review of the literature. J Am Acad Dermatol 1991; 24: 747 – 55. 119. Kadonaga JN, et al. Neurocutaneous melanosis in association with the Dandy – Walker complex. Pediatr Dermatol 1992; 9: 137 – 43. 120. Nagawawa H, et al. Long term survival after removal of primary intracranial malignant melanoma. Acta Neurochir 1989; 101: 84 – 8.
Section 11 : Pediatric Malignancies
72
Malignant Tumors of the Skin and Subcutaneous Tissue in Children Ilene L. Rothman, Joyce B. Farah and Thomas N. Helm
INTRODUCTION Malignant tumors of the skin and subcutaneous tissue are rare in childhood. In a study evaluating 775 superficial lumps in children, only 1.4% were malignant.1 Since the great majority of pediatric skin tumors are harmless, the danger is that a low index of suspicion will delay diagnosis and thereby worsen prognosis. This chapter reviews malignant tumors of the skin in children, including primary skin malignancies, tumors that usually arise in other organs but that can present primarily in skin and subcutaneous tissues, and those that can metastasize to the skin. Predisposing factors and syndromes, important in this population, are discussed.
MELANOMA Melanoma is one of the least common malignancies in childhood, accounting for only 1–3% of all pediatric malignancies. Despite being the most common primary skin cancer in childhood,2 melanomas in this population comprise only 1–4% of all melanomas and have an estimated incidence of only one per million per year.3 Melanoma is exceedingly rare in the first decade of life. Over 80% of childhood melanomas occur during the second decade.4 While male to female incidence is equal in adults, some studies have shown a female preponderance in childhood.5,6 There are rare reports of congenital melanoma related to transplacental transmission of maternal melanoma. Melanoma is actually the most common transplacentally transmitted neoplasm.4 Melanoma present at birth in a congenital nevus has also been reported.4 Prevalence of melanoma in the first decade has remained constant. However, prevalence in the teens has been increasing steadily over the past 20 years.1,5,7 In this latter agegroup, melanomas usually arise de novo, whereas most melanomas under age 12 arise in precursor lesions or in the setting of another predisposing factor. Precursor lesions are mainly giant congenital nevi (defined as >20 cm or expected
to reach that size by adulthood). Approximately 2–7% of giant congenital nevi eventually develop melanoma, most prior to age 12.8 (see Figure 1). Blue nevi as precursors to melanoma are extremely rare, particularly in children.9 Predisposing factors for melanoma include xeroderma pigmentosum (XP), prolonged immunodeficiency, and family history of melanoma, especially in the setting of familial atypical (or dysplastic) mole syndrome. XP is a rare inherited disorder of DNA repair. Manifestations include extreme photosensitivity, resulting in a greater than 1000-fold increased risk of skin cancers in young patients. Fifty-seven percent develop squamous cell and basal cell carcinomas (BCCs) and 22% develop melanoma.4 Children with primary immunodeficiencies have a three to sixfold increased risk of melanoma, and those with Hodgkin’s have an eightfold increased risk. Melanoma has also been reported after organ transplant.4 A strong family history of melanoma, especially when involving two or more relatives, increases the risk of childhood melanoma. The entity of familial atypical (or dysplastic) mole syndrome is controversial. There is widespread disagreement over its definition and even its existence. There is agreement, however, that melanoma risk is increased in patients with a high number of ordinary nevi (over 50 or 100) as well as in the setting of clinically atypical nevi. The histologic definition of atypical or dysplastic nevi has been more difficult to delineate.10 A rare but important predisposing condition associated with some giant congenital nevi is neurocutaneous melanosis (leptomeningeal melanosis, neurocutaneous melanocytosis) characterized by melanocytic infiltration of the leptomeninges. The skin lesions are usually giant nevi in the midline area of the lower back or posterior scalp or neck. Neurocutaneous melanosis can be asymptomatic or can cause symptoms related to increased intracranial pressure. Magnetic resonance (MR) findings of T1 shortening can be indicative of this diagnosis.11 Most cases of neurocutaneous melanosis develop within the first two years of
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
MALIGNANT TUMORS OF THE SKIN AND SUBCUTANEOUS TISSUE IN CHILDREN
Figure 1 Melanoma arising in a giant congenital nevus. This patient had associated leptomeningeal involvement. (Reprinted from Cohen, B, Pediatric Dermatology 3rd ed., 2005, with permission from Elsevier).
life. Symptomatic patients have a higher incidence of leptomeningeal melanoma than do asymptomatic patients. The actual incidence in the asymptomatic population is unclear but is thought to be very low.11,12 Typical presentation of melanoma is a pigmented lesion, usually asymptomatic, often with history of recent change and showing asymmetry and irregularities in pigmentation. The majority of pediatric melanomas are incorrectly diagnosed as benign on initial biopsy. Melanomas in children may histologically resemble a benign Spitz nevus, especially when only a portion of the lesion is available for pathologic examination. The presence of mitoses at the base of the neoplasm, absence of symmetry and circumscription, and marked variation in nuclear size and shape all support a diagnosis of melanoma. Cytogenetic studies can now be used to help differentiate Spitz nevi from melanomas. Treatment of melanoma always involves excision of the lesion. Margins depend on the level of invasion. In recent years, the trend has been toward narrower margins with current margin recommendations of the American Academy of Dermatology Task Force as follows: 0.5 cm for melanoma in situ, 1.0 cm for melanoma thickness under 2 mm, and 2.0 cm for melanoma thickness of 2 mm or greater.1 Sentinel lymph
811
node biopsy has proven to be a minimally invasive and accurate predictor of lymph node involvement. The procedure has been shown to be safe and useful in the pediatric population.13 With giant congenital nevi, attempts should be made to remove as much of the lesion as possible. This is often done in two or more stages. An exception is in the presence of neurocutaneous melanosis. Since these melanocytic lesions involve the central nervous system (CNS), it is impossible to remove all melanocytic tissue, so treating the skin aggressively may not be warranted. Patients with high-risk primary cutaneous melanoma, especially with regional lymph node involvement, are often treated with adjuvant interferon α-2b and various melanoma vaccines. Biologic response modifiers such as granulocytemacrophage colony-stimulating factor (GM-CSF), interleukins (IL-2, IL-12), and interferon γ are often integrated into vaccine strategies. Metastatic melanoma is treated with chemotherapeutic agents such as dacarbazine (DTIC) or combined biologic and chemotherapy (biochemotherapy) or radiation or both.14 As with adult onset melanoma, prognosis is most reliably predicted by tumor thickness at diagnosis. Ulceration, lymph node involvement, and distant metastases are additional prognostic indicators. In one study, overall survival in pediatric melanoma was 89% at 5 years and 79% at 20 years.5 Prognosis in transplacentally transmitted melanomas is poor.4 Metastatic spread to the brain also has a dismal prognosis, with an estimated five-year survival rate of just 3%.8,10 The prognosis in patients with symptomatic neurocutaneous melanosis is also very grave.8 Approximately 40–50% of these develop CNS melanoma.15 Because it is so uncommon, melanoma in children may be overlooked, delaying diagnosis and treatment and resulting in a poorer prognosis.
BASAL CELL CARCINOMA BCC, the most common form of skin cancer in adults, is a rare malignancy in children. In one large sample, BCC accounted for 13% of all cutaneous malignancies in children.16 In recent years, it has been seen with increasing frequency in adolescents, especially those with fair skin and particularly in sunny climates.17 This is presumably actinically induced as in adults. However, BCC is so rare in the first decade that an underlying cause must be looked for when it occurs in this age-group. Predisposing factors include defined syndromes having BCC as a feature, immunodeficiency or immunosuppression, ionizing radiation and arsenic exposure.16 (see Table 1) Defined syndromes where BCC is a prominent feature include XP in which BCC and squamous cell carcinoma (SCC) are the most frequent neoplasms (see previous section), albinism, and the most common predisposing factor, Nevoid Basal Cell Carcinoma Syndrome (NBCCS, Gorlin’s Syndrome). This autosomal-dominant disorder is characterized by numerous BCCs starting in the first decade. Other features include odontogenic keratocysts of the jaw, epidermal cysts, rib and craniofacial anomalies, and palmar
812
PEDIATRIC MALIGNANCIES
Table 1 Predisposing factors for basal and squamous cell carcinomas. (Reprinted from Pediatric Dermatology, 3rd ed., Schachner L, p 879, 2004, with permission from Elsevier).
Condition Sunlight Nevoid basal cell carcinoma syndrome Bazex – Dupre – Christol syndrome Xeroderma pigmentosum Rothmund – Thomsen syndrome Epidermolysis bullosa dystrophica Hidrotic ectodermal dysplasia Albinism Fanconi anemia Dyskeratosis congenita Erythropoietic porphyria Congenital/acquired immunodeficiency Immunosuppression Epidermodysplasia verruciformis Other human papilloma virus infections Burn scars Tuberculosis sinus tracts Osteomyelitis sinus tracts Chemotherapy Ionizing radiation Arsenic exposure Nevus sebaceous
BCC
SCC
X X X X X
X
X
X X
X X X X
X X X X X X X X X X X X X X X X X
pits. These patients also have an increased incidence of other tumors including fibromas of the ovaries and heart as well as medulloblastomas. (Up to 10% of patients with medulloblastoma have NBCCS.)16,18 Some patients who develop BCC after radiation therapy for another childhood malignancy are found to have NBCCS. Until very recently, BCC was thought to commonly arise in congenital sebaceous nevi (SN) with reported incidence as high as 50%.19 However, current consensus is that almost all of these tumors previously diagnosed as BCC were actually the benign basaloid tumor, trichoblastoma,17,19 – 23 and that BCC actually arises very rarely from SN, reported now in most studies as less than 1%.17,19,22 Clinical presentation of BCC is a pearly or translucent papule with overlying telangiectasia, sometimes with a central dell. Crusting and hyperpigmentation can be features. In NBCCS, however, the lesions present as multiple, sometimes numerous (50–100) small papules, often resembling skin tags. (see Figure 2) On biopsy, basaloid tumor extends from the epidermis into the dermis with retraction artifact. There is a surrounding mucinous stroma. Peripheral palisading of nuclei is encountered. The most common treatment for BCC is excision. Other modalities frequently employed include electrodessication with curettage, imiquimod cream for superficial lesions, and Mohs micrographic surgery (MMS) for poorly defined lesions. Radiation therapy is to be avoided, especially in patients with NBCCS where hundreds of new tumors can result.16 A recent report of three children with very extensive areas of diffuse BCCs associated with NBCCS showed treatment with 5-aminolevulinic acid photodynamic therapy to be well-tolerated and highly effective.24 (see Figure 3)
Figure 2 Multiple BCCs in child with NBCCS.
The prognosis for BCC is excellent in that the tumor virtually never metastasizes. However, some variants can be locally destructive if left untreated.
SQUAMOUS CELL CARCINOMA SCC is extremely rare in children, accounting for less than 0.1% of all childhood malignancies and about 5% of childhood cutaneous malignancies.16 When seen in adults, SCC is usually actinically induced or associated with tobacco and alcohol use, whereas in children, SCC is almost always associated with an underlying predisposing factor. Incidence without predisposing factors amounts to a collection of isolated reports.25 – 27 Predisposing factors for SCC include XP, immunodeficiency or immunosuppression including that after organ transplantation,28 human papillomavirus infections (especially when in the setting of immunosuppression), ionizing radiation, burn scars, and chronic trauma, particularly that associated with dystrophic epidermolysis bullosa (EB) congenita.16 (see Table 1) SCC has only rarely been noted as a second malignancy following chemotherapy.29 SCC or BCC in immunosuppressed transplant patients is rare in childhood, as the long latency period usually leads to development of these skin cancers in adulthood.28 A similar
MALIGNANT TUMORS OF THE SKIN AND SUBCUTANEOUS TISSUE IN CHILDREN (a)
(b)
813
photodynamic therapy. Radiation is avoided if possible in childhood SCC because of the risk of inducing secondary cancers.27 Prognosis depends on size and depth of the tumor. Again because of the rarity of pediatric SCC, specific data as to prognosis in this population are not readily available. Prognosis seems to mirror that in adults, with early lesions readily curable and more advanced lesions being less responsive to treatment.
MALIGNANT HISTIOCYTOSIS
(c)
(d)
Malignant histiocytosis is a rare systemic disease involving a rapid proliferation of atypical histiocytes, usually with prominent erythrophagocytosis. Skin involvement occurs in 10–15% of cases.30 Skin lesions present as reddish-brown papules, plaques, nodules (sometimes ulcerated), purpura, and occasionally hypertrophic gingiva. Lesions are 1 mm to 5 cm in diameter and have a predilection for the upper body and mucosa. Cutaneous lesions can be the initial manifestation or appear later in the disease course. Systemic manifestations result from multisystem invasion by malignant histiocytes and include fever, lymphadenopathy, hepatosplenomegaly, jaundice, wasting, and pancytopenia.31 Treatment of malignant histiocytosis involves the use of intensive, multiagent chemotherapy regimens, similar to those used for malignant lymphoma. In general, prognosis is poor, with survival time in months after initial diagnosis.31
LYMPHOMA
Figure 3 Child with NBCCS post radiation therapy for medulloblastoma: (a) Before treatment with 5-aminolevulinic acid photodynamic therapy (ALA-PDT); note hundreds of BCCs in radiation field. (b) Response to initial 2 cm-diameter test spots. (c) Six months after the seventh ALA-PDT treatment, field is more than 98% clear. (d) Four years later, essentially no new BCCs have arisen in the ALA-PDT treated areas. (Reprinted from Archives of Dermatology, 141; 60 – 67. 2005 American Medical Association. All rights reserved).
situation exists for SCC in dystrophic EB patients where the tumor can present in childhood but is more common starting around age 40.16 SCC presents clinically as a firm erythematous or skincolored papule or nodule. The surface can be smooth or scaly and sometimes verrucous or ulcerated. On histology, keratinocytes with enlarged and pleomorphic nuclei extending into the dermis are seen. Many keratinocytes may be surrounded by an infiltrate of lymphocytes, histiocytes, and eosinophils. Necrotic keratinocytes and abnormal mitotic figures are usually present. Because of the rarity of SCC in the pediatric population, specific treatment guidelines do not exist. As with adults, location and size of the tumor impact treatment selection. The lesions are most frequently surgically excised. Other treatments include radiation, chemotherapy, MMS, and
Lymphomas are the third most common malignancy in childhood, after leukemia and CNS tumors, with about 13 cases per million children per year. Hodgkin’s lymphoma (HL), while very rare before age 5, accounts for about 40% of all childhood lymphomas, occurring mainly in adolescents and young adults.16 Cutaneous lymphomas are rare in children, with specific skin lesions of malignant cells being especially rare in HL.16 The majority of cutaneous lymphomas are either mycosis fungoides (MF) (a T-helper cell lymphoma and major subset of cutaneous T cell lymphoma) or anaplastic large cell lymphoma (ALCL, CD30+ lymphoma), with MF being about twice as common as ALCL.32 MF is primarily an indolent disease of skin. When spread of MF to lymph nodes, bone marrow, and viscera does occur, it is often many years after presentation. MF is most common in older adults but is being recognized in children with increasing frequency. The incidence in children is not well established and may be underestimated because of a lack of clinical awareness. It has been reported as 0.5–5% of all MF cases, but patients diagnosed as young adults often report that symptoms began during their teens.33 Skin involvement in other non-Hodgkin’s Lymphomas is extremely rare. When cutaneous lesions do occur, they tend to be seen late in the disease course. There have been isolated reports of primary skin presentation in Precursor B cell Lymphoblastic Lymphoma,34 Marginal Zone Lymphoma,
814
PEDIATRIC MALIGNANCIES
Figure 4 Hypopigmented patches of MF in a child. (Reprinted from Pediatric Dermatology, 3rd ed., Schachner L, p 887, 2004, with permission from Elsevier).
human T-lymphotrophic virus type 1 (HTLV-1) associated Lymphoma, and Cytotoxic natural killer/T cell lymphoma. Panniculitic Lymphoma, very rare in children, can present with facial swelling and acneiform lesions. While skin involvement is common in Burkitt’s Lymphoma, it is almost always secondary to direct extension.16 Clinical presentations of MF vary and include erythematous patches and scaly plaques as well as nodules, sometimes crusted. Children most commonly present with patch-stage disease,33 with 20–88% of MF in children presenting with hypopigmented lesions,1,35,36 sometimes with only a single lesion.1 This hypopigmented presentation is mainly seen in children.37 (see Figure 4) Some studies report lesions in children most commonly occur on the buttocks.33,36 Since MF is generally considered a disease of middle or advanced age and clinical manifestations can mimic more common skin disorders such as eczema, recognition in children requires a high index of suspicion. The diagnosis should be considered in the setting of persistent eczematoid eruptions, especially when resistant to treatment. Skin manifestations of systemic ALCL are usually one or more ulcerating nodules. When ALCL presents in skin in children, the chances are great that the systemic disease is present, as the primary cutaneous form of ALCL is extremely rare in the pediatric population.16 Lymphomas in skin are associated with a varied histologic appearance. Hypopigmented MF is characterized by infiltrative atypical T cells within the epidermis. Immunohistochemical marker studies may show deletion of CD7. The hypopigmented variant is usually CD8 (suppressor cell) dominant. Gene rearrangement studies are used to confirm the diagnosis. ALCL is characterized by expression of CD30 antigen. Primary systemic forms usually express the anaplastic lymphoma kinase (ALK) fusion protein. There are no established treatment protocols specially designed for the treatment of MF in children. Many cases, especially when limited to the early patch stage, are treated with high-potency topical corticosteroids. One study showed a 94% response rate to this treatment.33 Other treatments include oral or topical Psoralen followed by ultraviolet
A light (PUVA) and topical nitrogen mustard.16 For later stage disease, systemic chemotherapy with radiation is used. Electron beam, frequently employed in adults, is rarely needed in children.1 Treatment of ALCL depends on the extent of the disease. Localized skin lesions can be excised or treated with radiotherapy, whereas systemic disease requires multiagent chemotherapy.38 The most important factors affecting prognosis in MF are type and extent of skin involvement and whether the lymph nodes or viscera are affected.1 Cures are considered rare in MF as the disease is more often controlled with therapy. However, in early stage disease, young people have an excellent prognosis,1 and complete remissions have been reported in children.16 One study showed that survival rate in hypopigmented MF was comparable with that of classic MF. There was no progression to systemic or lymph node involvement.37 In another study of 34 patients with juvenileonset MF, survival at 5 and 10 years was 95 and 93% respectively.39 Prognosis in systemic ALCL with skin involvement at presentation depends on expression of the ALK protein. Patients with positive ALK have a five-year survival rate of 70–80% while in those with negative ALK, survival is only 15–30%.38
LEUKEMIA Leukemia is the most common malignancy of childhood at 40 cases per million, accounting for 25–40% of all childhood malignancies.16 Leukemic skin infiltrates are metastatic lesions and as such are the most common cause of cutaneous metastases in children.1,16 Skin lesions preceding all other clinical and laboratory evidence of leukemia, so-called “aleukemic leukemia” are extremely rare. In these cases, other signs follow within weeks or months.40 While only about 1% of children with leukemia have skin infiltrates (“leukemia cutis”),16 the frequency varies widely depending on the type of leukemia. Acute lymphoblastic leukemia (ALL) is the most common form of childhood leukemia, accounting for about 80% of all cases.16 Leukemic infiltrates are rare in ALL, with one study showing cutaneous findings at presentation in only 1.2% of cases. When cutaneous involvement occurs, it can be an early manifestation.34 Conversely, in acute myeloid leukemia (AML) which accounts for about 20% of childhood leukemias and for most congenital leukemias,16 cutaneous infiltrates are common, occurring in 30–50% of cases40,41 with congenital leukemia presenting with leukemia cutis in about 50% of cases.16 Leukemic infiltrates most commonly present clinically as multiple lesions. These can be papules or nodules, with color ranging from violaceous to red or brown.16,41 Less common presentations include bullae and ulcerations.16 Lesions in dark-skinned patients can be easily missed. Although reports of leukemia cutis in this population are rare, one study found skin findings, when looked for, were as common as in Caucasians.42 In congenital leukemia, purpuric nodules can appear identical to so-called “blueberry
MALIGNANT TUMORS OF THE SKIN AND SUBCUTANEOUS TISSUE IN CHILDREN
muffin” lesions of dermal erythropoiesis seen in intrauterine infections or to lesions of congenital rhabdomyosarcoma (RMS) or neuroblastoma. These lesions are most often multiple and can be widespread, although presentation as a single lesion has been reported.12 Leukemic infiltrates of the skin typically extend around blood vessels and cutaneous appendages. The epidermis is often spared. Mature as well as immature cells may be encountered. Immunohistochemistry can be an important diagnostic aid. Treatment of leukemia depends on type but generally includes aggressive chemotherapy, and, in some cases, radiation, or bone marrow transplantation or both. Prognosis in ALL is favorable with a five-year survival rate of 75–80% depending on subtype.16 Five-year survival in AML is closer to 40%,16 and prognosis is especially poor when presentation is in the first few months of life.40 Although leukemia cutis has been reported as an adverse prognostic indicator, its predictive value actually varies depending on several factors, the most important of which is when in the disease course it occurs. A large prospective study points out that leukemia cutis lesions present at initial diagnosis have no predictive value as to eventual outcome.34 However, cutaneous involvement is felt to be a poor prognostic sign when it occurs later in the disease, often signifying late extramedullary relapse following chemotherapy.16 A small study suggests that dense leukemic infiltrates may correlate with a poor prognosis.41
RHABDOMYOSARCOMA RMS, the most common childhood sarcoma, refers to a heterogeneous group of tumors of mesenchymal stem cell origin. Fewer than one percent involve the skin, either primarily or as metastases.16 Patients with NBCCS are predisposed to developing these tumors.43 When arising primarily in the skin, RMS presents as erythematous to violaceous papules, nodules and plaques. Unusual presentations include RMS presenting in the neonatal period as a “blueberry muffin” rash with an associated soft tissue mass44,45 and RMS presenting as a large ulcerated nodule in a congenital melanocytic nevus.46 Congenital RMS is rare and uniformly fatal, with more than half presenting with multiple cutaneous metastases.47 Alveolar and embryonal are the histological variants of RMS that can present in the skin. The alveolar subtype which is the most likely to present with metastasis to the skin16 is composed of large ill-defined aggregates of poorly differentiated, round or oval tumor cells. A loss of cellular cohesion is noted and accounts for the “alveolar-like” spaces. The embryonal subtype presents with areas of hyper and hypocellularity. Poorly oriented undifferentiated spindle type cells are evident, and there is a prominent myxoid stroma with little collagen. Current treatment regimens are based on stage and other prognostic factors and generally includes surgery, multiagent chemotherapy, and radiotherapy for residual disease. The prognosis for patients with metastatic disease remains unsatisfactory,48 and more effective treatments are needed.
815
Survival data obtained from the Surveillance, Epidemiology and End Results (SEER) program from 1973–2000 revealed that children who survived the first five-years after diagnosis had an excellent prognosis49 with an overall survival of 90%.50 Age and histological subtype play an important role in prognosis. Children between 1 and 9 years of age have a better outcome than older children. Recurrences of RMS usually occur within two years, with reported survival approximately 17%.50
FIBROSARCOMA Fibrosarcoma is the second most common childhood sarcoma involving the skin, with almost all pediatric cases occurring before 5 years of age.16 These tumors are generally locally invasive but rarely metastatic.51,52 Fibrosarcoma is a tumor of deep soft tissue with secondary involvement of the skin. It can present at birth or during the first year of life, most frequently on the trunk and extremities.51 Clinical presentation is of an asymptomatic swelling or mass that is rapidly growing with a violaceous to bluish hue on the overlying skin.53,54 Fibrosarcomas are typically comprised of spindle cells arranged in short fascicles with a “herring bone-like” appearance. Mitoses are more common than in dermatofibrosarcoma protuberans (DFSP). The congenital infantile variant of fibrosarcoma is cellular and has prominent mitotic activity. Wide local excision is the treatment of choice. One series retrospectively analyzed the treatment and outcome of 11 infants with infantile fibrosarcoma, and found initial chemotherapy with surgery to be successful in most cases.52 Chemotherapy can be used prior to surgery to reduce the size of the tumor allowing for adequate resection.51,53 Prognosis in childhood fibrosarcoma is markedly better than in the adult type (Fibrosarcoma in children over the age of 10 is viewed as adult type). Fifty percent of fibrosarcomas in adults metastasize versus 7% in the childhood variant.16,54
DERMATOFIBROSARCOMA PROTUBERANS DFSP typically occurs on the trunk of young to middle-aged adults. It is locally invasive with a high rate of recurrence and a very low metastatic potential. Pediatric DFSP is rare but has been increasingly recognized in recent years55 comprising 5–6% of all soft tissue sarcomas in children.56 In one case series, 5.9% of DFSP presented before 13 years of age.16 Clinical presentation is of a nonspecific slowly growing firm plaque or nodule most commonly on the trunk, developing nodular foci with time. (see Figure 5) In infants, it can present as erythematous-blue plaques resembling a vascular malformation. Congenital DFSP can present as “blueberry muffin baby”.47 Bednar tumors are currently considered to be a variant of DFSP.57 An unusual presentation of DFSP was reported on a child’s foot.58 Because of its rarity, there is usually a delay in diagnosis of DFSP in childhood, sometimes for as long as five years.56 Biopsy of DFSP reveals spindle cells with hyperchromatic nuclei associated with minimal inflammatory response. Spindle cells in a storiform arrangement extend into the panniculus and may be deceptively bland. The tumor often involves
816
PEDIATRIC MALIGNANCIES
Histology reveals small cells with hyperchromatic nuclei that are often arranged in a rosette pattern. Neuronspecific enolase stains are positive. Rosettes and lobules of small rounded cells with round or oval nuclei are evident, and the cytoplasm is indistinct. A central core neurofibrillar material may be seen within the rosette. Treatment depends on the extent of disease and can include surgery, chemotherapy, radiation and/or bone marrow transplantation. Prognosis varies markedly depending on several factors. Patients with localized surgically resectable disease have a 95% cure rate. Outcome is also excellent in children presenting under age one regardless of whether their disease is localized or metastatic. Two-year survival in this under age one group is 90%.16 In contrast, older patients, especially those with metastatic disease, have a much poorer prognosis.
LEIOMYOSARCOMA
Figure 5 DFSP on a shoulder. Note the multinodular appearance. (Reprinted from Schachner L, Pediatric Dermatology 3rd ed., 2004, with permission from Elsevier).
both the dermis and subcutis. Spindle cells in DFSP are CD34 positive, whereas the spindle cells in a dermatofibroma are typically factor XIIIa positive. When melanin-containing dendritic cells are present, the tumor is known as a Bednar tumor.59 Traditional treatment has been surgical excision with 3cm margins. However, this treatment is associated with recurrence rates of up to 50% within 5 years. In recent years, MMS is gaining acceptance as an alternative to traditional excision. With this technique, the recurrence rate has decreased to approximately 2%.47 Thornton et al. used magnetic resonance imaging (MRI) to outline the tumor extent and plan surgery. They found that in 10 cases evaluated preoperatively by MRI, there were no recurrences with follow-up in some cases up to 17 years.56
NEUROBLASTOMA Neuroblastoma is the third most common childhood neoplasm accounting for 8–10% of childhood cancers. It is the most common congenital cancer.60 Boys are more affected than girls.54 The tumor arises from the adrenal medulla or along the sympathetic ganglia. Thus, the most common presentation is of an enlarging mass in the abdomen or retroperitoneum. While cutaneous metastasis occurs in only 2% of all children with neuroblastoma,16 30% of congenital metastatic neuroblastoma involves the skin.16,61 Cutaneous metastases manifest as multiple blue-to-purple dermal papules or nodules, often with the appearance of “blueberry muffin baby”. Periorbital ecchymosis is another dermatologic manifestation of metastatic disease.16,61 Of diagnostic value is the fact that the nodules may blanch on stroking for approximately 30–60 seconds and remain refractory for 1–2 hours secondary to catecholamine release.
Leiomyosarcoma (LMS) usually presents in adults but can rarely occur in childhood. It is the second most common malignancy in children with human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) where it is thought to be secondary to smooth muscle cell infection by Epstein Barr Virus (EBV).16 LMS is classified as superficial or deep, the superficial type arising from the arrector pili smooth muscle and the deep type arising from the smooth muscles of arteries and veins.62 Skin tumors are firm smooth skin-colored nodules that can be pedunculated, umbilicated or painful ulcerated nodules.62 – 65 At diagnosis, the majority of tumors are invasive.66 The mainstay of therapy for dermal tumors is surgical excision, while deeper tumors also require adjuvant chemotherapy.65 – 67 There are several reports in adults of successful treatment with MMS.65 Because of the rarity of the tumor, there are no widely accepted treatment or follow-up guidelines. For prognostic and treatment purposes, distinguishing between superficial and deep LMS is important. Tumors arising from the arrector pili muscle have a better prognosis than the deeper tumors.66 Metastases are rare in superficial cutaneous tumors but occur in 30–60% of deep LMS.63,64
LIPOSARCOMA Liposarcoma is a very rare pediatric tumor. Incidence of childhood liposarcoma is bimodal with peaks in infancy and the second decade.68 It can often be confused with the more common benign lipoblastoma.54,68,69 The most common location of this tumor in children is the lower extremities, especially the thigh.54,68 The tumor presents as an enlarging subcutaneous mass. Treatment is by surgical resection. Radiation or chemotherapy or both are added for residual tumor.68,70,71 Because of the rarity of liposarcoma, no large-scale controlled studies on treatment have been performed. Prognosis is excellent with complete surgical resection with negative margins. Children with LMS have a more favorable outcome than do adults.68
MALIGNANT TUMORS OF THE SKIN AND SUBCUTANEOUS TISSUE IN CHILDREN
SYNOVIAL SARCOMA Synovial sarcoma is a rare tumor, accounting for approximately 2–10% of all soft tissue sarcomas in childhood. The tumor is usually found on the lower extremities and presents as a painless soft tissue mass in close proximity to articular surfaces.54 Currently treatment involves resection, but there is no consensus on the role of adjuvant chemotherapy. Local recurrence as well as metastases may develop after periods as long as 20 years.54
EWING’S SARCOMA Ewing’s sarcoma, a primary bone tumor of children and adolescents, can involve the skin by direct extension from bone or soft tissue. Clinical presentation is of a solitary dermal/subcutaneous mass associated with edema and localized pain. The mean age of occurrence is 13 with a male predominance.72 Surgical resection and chemotherapy are the mainstay of treatment. Despite aggressive treatment, sometimes also including radiation, mortality ranges from 20–40% for patients with localized disease and almost 80% for patients with metastatic disease.73
MALIGNANT PERIPHERAL NERVE SHEATH TUMORS (NEUROFIBROSARCOMA, MALIGNANT SCHWANNOMA) Malignant peripheral nerve sheath tumor (MPNST) is the least common sarcoma in children.54 The cells of origin are usually perineural or endoneural fibroblasts rather than Schwann cells.74 They virtually always occur in the setting of neurofibromatosis 1 (von Recklinghausen disease, NF1) with 4% of patients with NF1 developing MPNST.75 Clinically, these lesions present as painful nodules ranging in size from 2–33 cm often with cysts, hemorrhage, and necrosis.54 The mainstay of treatment is wide local resection. Adjuvant radiation and chemotherapy are often added because of the aggressive nature of the tumor.74,76 Sacrificing of a main nerve is often unavoidable. MMS has been reported as successfully treating one case of MPNST in an adult.66 MPNST is generally associated with a poor prognosis.77
OTHER TUMORS WITH RARE SKIN METASTASES Several other tumors can rarely present in the skin. Wilm’s tumor of the kidney and some CNS tumors can have cutaneous metastases.16 Metastases to the scalp have been reported with ependymomas and germinomas.78,79 Additionally, choriocarcinomas in childhood can metastasize to the skin.16
CONCLUSION Diagnosis of malignant tumors of the skin and subcutaneous tissue in children requires a high index of suspicion. Prompt recognition can lead to improved cure rates. In the case of
817
primary skin tumors, their presentation can signify a predisposing condition or syndrome. In the case of metastases, skin lesions can be the first sign of the underlying malignancy, providing an opportunity to diagnose and start treatment.
REFERENCES 1. Hoang M, Friedlander S. Rare cutaneous malignancies of childhood. Current Opin Pediatr 1999; 11(5): 464 – 75. 2. Levine N. Neoplastic disorders of melanocytes. In Schachner L, Hansen R (eds) Pediatric Dermatology: Mosby, 2003: 516 – 524. 3. Scalzo D, et al. Childhood melanoma: a clinicopathological study of 22 cases. Melanoma Res 1997; 7(1): 63 – 8. 4. Pratt C, Pappo A. Management of infrequent cancers of childhood. In Pizzo P, Poplack D (eds) Principles and Practice of Pediatric Oncology: Lippincott Williams & Wilkins, 2002: 1165 – 1168. 5. Hamre M, et al. Cutaneous melanoma in childhood and adolescence. Pediatr Hematol Oncol 2002; 19(5): 309 – 17. 6. Berg P, Lindelof B. Differences in malignant melanoma between children and adolescents. A 35 year epidemiological study. Arch Dermatol 1997; 133(11): 1459 – 60. 7. Silverberg NB. Update on malignant melanoma in children. Cutis 2001; 67(5): 393 – 6. 8. Gibbs P, et al. Pediatric Melanoma: are recent advances in the management of adult melanoma relevant to the pediatric population. J Pediatr Hematol Oncol 2000; 22(5): 428 – 32. 9. Popovic M, et al. Childhood malignant blue nevus of the ear associated with two intracranial melanocytic tumors – metastases or neurocutaneous melanosis? Hum Pathol 2004; 35(10): 1292 – 6. 10. Barnhill R, Llewellyn K. Benign melanocytic neoplasms. In Bolognia J, et al.. (eds) Dermatology: Mosby, 2003: 1757 – 1787. 11. Foster R, et al. Giant congenital melanocytic nevi: The significance of neurocutaneous melanosis in neurologically asymptomatic children. Plast Reconstr Surg 2001; 107: 933. 12. Eichenfield L, Gibbs N. Hyperpigmentation disorders. In Eichenfield L, Frieden I, Esterly N (eds) Textbook of Neonatal Dermatology: W.B. Saunders, 2001: 370 – 394. 13. Toro J, et al. Sentinel lymph node biopsy in children and adolescents with malignant melanoma. J Pediatr Surg 2003; 38(7): 1063 – 5. 14. Swetter S. Malignant melanoma. In Bury G, Albertini J (eds) Dermatology eMedicine: eMedicine.com, Inc, 2002. 15. Bittencourt F, et al. Large congenital melanocytic nevi and the risk for development of malignant melanoma and neurocutaneous melanocytosis. Pediatrics 2000; 106(4): 736 – 41. 16. Burgdorf W, Ruiz-Maldonado R. Benign and malignant tumors. In Schachner L, Hansen R(eds) Pediatric Dermatology: Mosby, 2003: 863 – 899. 17. Cohen B. Pediatric Dermatology: Elsevier Inc, 2005. 18. Amlashi S, et al. Nevoid basal cell carcinoma syndrome: relation with desmoplastic medulloblastoma in infancy. A population-based study and review of the literature. Cancer 2003; 98(3): 618 – 24. 19. Santibanez-Gallerani A, et al. Should nevus sebaceous of Jadassohn in children be excised? A study of 757 cases, and literature review. J Craniofac Surg 2003; 14(5): 658 – 60. 20. Munoz-Perez M, et al. Sebaceous naevi: a clinicopathologic study. J Eur Acad Dermatol Venereol 2002; 16(4): 319 – 24. 21. Turner C, Shea C, Rosoff P. Basal cell carcinoma originating from a nevus sebaceus on the scalp of a 7-year old boy. J Pediatr Hematol Oncol 2001; 23(4): 247 – 9. 22. Cribier B, Scrivener Y, Grosshans E. Tumors arising in nevus sebaceus: a study of 596 cases. J Am Acad Dermatol 2000; 42: 263 – 8. 23. Dunkin C, Abouzeid M, Sarangapani K. Malignant transformation in congenital sebaceus naevi in childhood. J R Coll Surg Edinb 2001; 46(5): 303 – 6. 24. Oseroff A, et al. Treatment of diffuse basal cell carcinomas and basaloid follicular hamartomas in nevoid basal cell carcinoma syndrome by wide-area 5-aminolevulinic acid photodynamic therapy. Arch Dermatol 2005; 141(1): 60 – 7. 25. Travelute C, et al. Squamous cell carcinoma of the lip in a 19-monthold child: a case report. J Craniofac Surg 1996; 7(1): 60 – 3.
818
PEDIATRIC MALIGNANCIES
26. Soni S, et al. Stage 4 squamous cell carcinoma of the tongue in a child: complete response to chemoradiotherapy. J Pediatr Hematol Oncol 2001; 23(9): 612 – 15. 27. De Carvalho M, et al. Head and neck squamous cell carcinoma in childhood. Med Pediatr Oncol 1998; 31(2): 96 – 9. 28. Euvrard S, et al. Skin cancers following pediatric organ transplantation. Dermatol Surg 2004; 30(4 Pt2,): 616 – 21. 29. Hirota T, et al. Squamous cell carcinoma of the tongue as a second malignancy in a patient previously treated for osteosarcoma. Pediatr Hematol Oncol 2000; 17(5): 421 – 4. 30. Morgan NE, et al. Clinical and pathologic cutaneous manifestations of malignant histiocytosis. Arch Dermatol 1983; 119(5): 367 – 72. 31. Ringer E, Moschella S. Primary histiocytic dermatoses. Arch Dermatol 1985; 121(12): 1531 – 41. 32. Fink-Puches R, Chott A, Ardigo M. The spectrum of cutaneous lymphomas in patients less than 20 years of age. Pediatr Dermatol 2004; 21(5): 525 – 33. 33. Pabsch H, et al. Treatment of childhood mycosis fungoides with topical PUVA. J Am Acad Dermatol 2002; 47(4Pt 1): 557 – 61. 34. Millot F, et al. Cutaneous involvement in children with acute lymphoblastic leukemia or lymphoblastic lymphoma. Pediatrics 1997; 100: 60 – 4. 35. Ben-Amitai D, et al. Juvenile mycosis fungoides diagnosed before 18 years of age. Acta Derm Venereol 2003; 83(6): 451 – 6. 36. Tan E, Tay YK, Giam YC. Profile and outcome of childhood mycosis fungoides in Singapore. Pediatr Dermatol 2000; 17(5): 352 – 6. 37. El-Shabrawi-Caelen L, et al. Hypopigmented mycosis fungoides: frequent expression of a CD8+ T-cell phenotype. Am J Surg Pathol 2002; 26(4): 450 – 7. 38. Zic J. Lymphomatoid papulosis. In Raugi G, Mowad C (eds) Dermatology eMedicine: eMedicine.com, Inc, 2002. 39. Wain E, et al. Outcome in 34 patients with juvenile-onset mycosis fungoides: a clinical immunophenotypic and molecular study. Cancer 2003; 98(10): 2282 – 90. 40. Canioni D, et al. Skin lesions revealing neonatal acute leukemias with monocytic differentiation. A report of 3 cases. J Cutan Pathol 23(3): 254 – 8. 41. Koga M, Furukawa S. Leukemia cutis in three children : clinical and immunohistochemical studies. Pediatr Dermatol 1996; 13(3): 200 – 6. 42. Riyat M. Mucocutaneous manifestations of lymphomas and leukemias in black Kenyan children. Int J Dermatol 1995; 34(4): 249 – 55. 43. Dans M, Fakharzadeh SS. Genetic basis of skin cancer. In Rigel DS (ed) Cancer of the Skin: Elsevier Inc, 2005: 15. 44. Godambe SV, Rawal J. Blueberry muffin rash as a presentation of alveolar cell rhabdomyosarcoma in a neonate. Acta Paediatr 2000; 89(1): 115 – 7. 45. Schmidt D, Fletcher CD, Harms D. Rhabdomyosarcomas with primary presentation in the skin. Pathol Res Pract 1993; 189(4): 422 – 7. 46. Hoang MP, Sinkre P, Albores-Saavedra J. Rhabdomyosarcoma arising in a congenital melanocytic nevus. Am J Dermatopathol 2002; 24(1): 26 – 9. 47. Guillen DR, Cockerell CJ. Sarcomas of the skin. In Rigel DS (ed) Cancer of the Skin: Elsevier Inc, 2005: 15. 48. Ferrari A, Casanova M. Current chemotherapeutic strategies for rhabdomyosarcoma. Expert Rev Anticancer Ther 2005; 5(2): 283 – 94. 49. Punyki JA, et al. Long-term survival probabilities for childhood rhabdomyosarcoma: a populations-based evaluation. Cancer 2005; 103(7): 1475 – 83. 50. Stuart A, Radhakrishnan J. Rhabdomyosarcoma. Indian J Pediatr 2004; 71(4): 331 – 7. 51. Hicks J, Mierau G. Pediatric fibroblastic and myofibroblastic tumors: what’s new and what’s not. J Surg Oncol 2001; 78(4): 225 – 38. 52. Loh ML, et al. Treatment of infantile fibrosarcoma with chemotherapy and surgery: results from the Dana-Farber Cancer Institute and Children’s Hospital, Boston. J Pediatr Hematol Oncol 2002; 24(9): 722 – 6. 53. Dixon NE, et al. Congenital fibrosarcoma: report of one case treated with pre-surgical chemotherapy. Int Pediatr 2003; 18(2): 87 – 91. 54. Variend S. Paediatric neoplasia. In Gresham A (ed) Current Histopathology: Kluwer Academic Publishers. Hingham, MA, 1993: Vol 22.
55. Strauss RM, et al. A case of childhood dermatofibrosarcoma protuberans without detected cytogenetic abnormality. Br J Dermatol 2003; 148: 1051 – 5. 56. Thornton S, et al. Childhood dermatofibrosarcoma protuberans: role of preoperative imaging. J Am Acad Dermatol 2005; 53(1): 76 – 83. 57. Reis-Filho JS, et al. Pediatric pigmented dermatofibrosarcoma protuberans (Bednar tumor): case report and review of the literature with emphasis on the differential diagnosis. Pathol Res Pract 2002; 198(9): 621 – 6. 58. Cione JA, Lynn B, Boylan J. Dermatofibrosarcoma protuberans. A rare case involving the pediatric foot. J Am Podiatr Med Assoc 1999; 89(8): 419 – 23. 59. Vidimous A, Helm T, Papay F. In Miller S, Malloney, M (eds) Dermatofibrosarcoma Protuberans in Cutaneous Oncology: Blackwell Science. Cambridge, MA. 1998: 822 – 31. 60. McGrory JE, et al. Nonrhabdomyosaroma soft tissue sarcomas in children: the Mayo Clinic experience. Clin Orthop Rel Res 2000; 374: 2478 – 258. 61. Schwartz A. Metastatic tumors of the skin. In Burg G (ed) Atlas of Cancer of the Skin: Churchill Livingston, 2000: 209 – 218. 62. Porter CJ, Januszkiewicz JS. Cutaneous leiomyosarcoma. Plast Reconst Surg 2002; 109(3): 964 – 7. 63. Kuflik JH, Schwartz RA, Rothenberg J. Dermal leiomyosarcoma. J Am Acad Dermatol 2003; 48(5 Suppl): S51 – 3. 64. Wrone DA, Sober AJ. Uncommon cutaneous malignancies. In Sober AJ, Halusk FG (eds) Skin Cancer: BC Decker Inc, 2001: 118 – 126. 65. Kohler S. Muscle, adipose, and cartilage neoplasm. In Bolognia JL, Jorizzo JL, Rapini RP. (eds) Skin Cancer: Mosby, 2003: . 1883 – 1898. 66. Bricca GM, Brodland D. Mohs surgery: the full spectrum of application. In Sober AJ, Halusk FG (eds) Skin Cancer: BC Decker Inc, 2001: 573 – 548. 67. Tsutusmida A, et al. Management of superficial leiomyosarcoma: a retrospective study of 10 cases. Plast Reconstr Surg 2005; 116(1): 8 – 12. 68. Ferrari A, et al. Childhood liposarcoma: a single-institutional twenty – year experience. Pediatr Hematol Oncol 1999; 16: 415 – 21. 69. Vocks E, Worret WI, Burgdorf WH. Myxoid liposarcoma in a 12-yearold girl. Pediatr Dermatol 2001; 18(1): 85 – 6. 70. El-Baradie MM, et al. Prognostic factors for retroperitoneal soft-tissue sarcoma. J Egypt Natl Canc Inst 2003; 15(4): 265 – 74. 71. Marcus KC, et al. Childhood soft tissue sarcoma: a 20-year experience. J Pediatr 1997; 31(4): 603 – 7. 72. Chow EC, et al. Cutaneous and subcutaneous Ewing’s sarcoma: an indolent disease. Int J Radiat Oncol Biol Phys 2000; 46(2): 433 – 8. 73. Thacker MM, Temple HT, Scully SP. Current treatment for Ewing’s sarcoma. expert rev. Anticancer Ther 2005; 5(2): 319 – 31. 74. Argeyi ZB. Neural and neuroendocrine neoplasms (other than neurofibromatosis). In Bolognia JL, Jorizzo JL, Rapini RP. (eds) Dermatology: Mosby, 2003: 1843 – 1861. 75. Baehring JM, Betensky RA, Batchelor TT. Malignant peripheral nerve sheath tumor: the clinical spectrum and outcome of treatment. Neurology 2003; 61(5): 696 – 8. 76. DeCou JM, et al. Malignant peripheral nerve sheath tumors: the St. Jude children’s research hospital experience. Ann Surg Oncol 1995; 2(6): 524 – 9. 77. Ueda R, et al. Malignant peripheral nerve sheath tumor in the anterior skull base associated with neurofibromatosis type 1. Neurol Med Chir 2004; 44: 38 – 42. 78. Orozco-Covarrubias M, Lourdes T.-S, Suran-McKinster C. Malignant cutaneous tumors in children: twenty years experience at a large pediatric hospital. JAAD 1994; 30: 243 – 9. 79. Pesce K et al.. Metastatic lesions to the skin in children and adolescents. In James WD (ed) Advances in Dermatology: St. Louis, MO: Mosby Year Book, 1997: Vol 12: 237 – 75.
FURTHER READING Guidelines of Care for Primary Cutaneous Melanoma. American Academy of Dermatology Task Force and Guidelines/Outcomes Committee, 2001. Jubran RF, et al. Predictors of outcome in children with langerhans cell histiocytosis. Pediatr Blood Cancer 2005; 45: 37 – 42.
Acknowledgments Chapter 1 Based in part on the chapters ‘Uncommon tumors of renal parenchyma’ by Robert P. Huben and Satish R.C. Velagapudi, and ‘Adult Wilms’ tumor’ by Mary Jane Petruzzi and Daniel M. Green, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 3 Based in part on the chapter ‘Urethral cancer’ by Oscar E. Streeter, Jocelyn Speight and Garth Green, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 4 Based in part on the chapters ‘Small cell undifferentiated (neuroendocrine) carcinoma of the prostate’ by Derek Raghavan and Peter Russell, ‘Sarcomas of the adult prostate gland’ by Ellis G. Levine, John F. Gaeta and Robert F. Huben, and ‘Transitional cell carcinoma of the prostate’ by David M. Kraklau and James E. Montie, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 5 Based in part on the chapter ‘Rare tumors of the testis and paratesticular tissues’ by Alan Horwich, Charles R. Hamilton and Cyril Fisher, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 6 Based in part on the chapter ‘Uncommon epithelial tumors of the oral cavity’ by K. Thomas Robbins and Yao S. Fu, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 8 Based in part on the chapter ‘Nasopharyngeal Carcinoma in Caucasians’ by Lester J. Peters, John G. Batsakis, Helmuth Goepfert and Waun K. Hong, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 11 Based in part on the chapter ‘Uncommon cancers of the thyroid’ by Jia Bi, Andy E. Sherrod and Derek Raghavan, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 13 Based in part on the chapter ‘Metaplastic breast carcinoma’ by Brenda P. Nicholson, Alexander D. Borowsky and
David H. Johnson, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 14 Based in part on the chapter ‘Adenoid cystic carcinoma of the breast’ by Brenda P. Nicholson, Masako Kasami, David Page and David H. Johnson, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 21 Based in part on the chapter ‘Primary lymphomas of the lung’ by Mark S. Allen and Daniel L. Miller, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 22 Based in part on the chapter ‘Primary sarcomas of the lung’ by Guru Sonpavde, Thomas M. Ulbright and Alan Sandler, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 25 Based in part on the chapter ‘Large cell neuroendocrine tumors of the lung’ by Henry Wagner, Jr and Andras Khoor, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 30 Based in part on the chapter ‘Uncommon cancers of the esophagus’ by Jonathan D. Cheng, Carolyn C. Compton and Neal J. Meropol, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 31 Based in part on the chapter ‘Uncommon cancers of the stomach’ by Jonathan D. Cheng, Carolyn C. Compton and Neal J. Meropol, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 32 Based in part on the chapter ‘Unusual pancreatic tumors’ by John S. Macdonald and Monty S. Metcalfe, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 34 Based in part on the chapter ‘Cancers of the small bowel’ by David J. Vaughn, E. Emma Furth and Stephen E. Rubesin, which appeared in Textbook of Uncommon Cancer, Second Edition.
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
820
ACKNOWLEDGMENTS
Chapter 35 Based in part on the chapter ‘Unusual tumors of the colon, rectum and anus’ by Paul Cheng and Leonard Saltz, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 36 Based in part on the chapter ‘Cancer of the appendix’ by Matthew H. Kulke, Robert T. Osteen and Charles S. Fuchs, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 38 Based in part on the chapter ‘Small cell carcinomas of the gastrointestinal tract’ by Paulo M. Hoff and Richard Pazdur, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 41 Based in part on the chapter ‘Stromal tumors of the ovary’ by Agustin A. Garcia and C. Paul Morrow, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 43 Based in part on the chapter ‘Fallopian tube cancer’ by Peter G. Rose, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 46 Based in part on the chapter ‘Tumors of the vulva and vagina’ by Agustin A. Garcia and J. Tate Thigpen, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 51 Based in part on the chapter ‘Unusual cutaneous malignancies’ by Nathalie C. Zeitouni, Richard T. Cheney and Allan R. Oseroff, which appeared in Textbook of Uncommon Cancer, Second Edition.
Chapter 54 Based in part on the chapter ‘Primary leptomeningeal melanoma’ by Mark R. Gilbert and David H. Lawson, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 55 Based in part on the chapter ‘Langerhans’ cell histiocytosis of the central nervous system’ by Vanita Kaitithan and Jonathan Finlay, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 59 Based in part on the chapter ‘Primary intracranial germ cell tumors’ by Mark T. Jennings, Elizabeth L. King and Paul L. Moots, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 60 Based in part on the chapter ‘Primary central nervous system lymphoma’ by Jeffrey J. Raizer and Lisa M. DeAngelis, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 67 Based in part on the chapter ‘Uncommon pediatric tumors of the thorax’ by Joanne M. Hilden, Jan Watterson and Louis P. Dehner, which appeared in Textbook of Uncommon Cancer, Second Edition. Chapter 70 Based in part on the chapter ‘Adrenocortical tumors in children’ by Raul C. Ribeiro and Edson L. Michalkiewicz, which appeared in Textbook of Uncommon Cancer, Second Edition.
Subject Index Notes: Page numbers in italics and bold refer to figures and tables respectively
Abdomen, acute, appendiceal carcinoid 751 Abdominal carcinomatosis, primary adenocarcinoma of fallopian tube 479 Abdominal distention, mucinous ovarian carcinoma 452 Abdominal mass adrenal gland disorders 143 borderline ovarian tumors 449 clear cell ovarian carcinoma 452 investigation in stromal tumors of the ovary 456 – 7 mucinous ovarian carcinoma 452 neuroblastoma 816 Abdominal pain adult granulosa cell tumors 459 appendiceal carcinoid 751 borderline ovarian tumors 449 endometrioid ovarian carcinoma 452 GIST in children 753 pancreatic cystadenocarcinoma 368 sex cord tumor with annular tubules 462 Abdominal radical trachelectomy, cervical tumors 515 Abortion (spontaneous), hydatidiform moles and 533 Abrikossoff’s tumor 585 Acid ingestion, verrucous carcinoma 338 Acid suppression therapy, Zollinger – Ellison syndrome 374 – 5, 376 Acinar cell carcinoma 370 pediatric malignancy 754, 755 Acinar cell carcinoma syndrome 370 Acinic cell carcinoma, salivary glands 92 Ackerman’s tumor 103 – 4 Acquired immunodeficiency syndrome (AIDS) see HIV infection/AIDS Acromegaly, Carney complex and 784 Acrospiroma, malignant 578 Actinomycin D (dactinomycin) gestational trophoblastic neoplasia 539 – 40 hydatidiform mole treatment 536 paratesticular rhabdomyosarcoma 80 Acute lymphocytic leukemia (ALL) leukemia cutis in children 814, 815 ovarian involvement, children 770 prognosis/treatment 815 Acute myeloid leukemia (AML) granulocytic sarcoma 512 late sequelae of chemotherapy 473 leukemia cutis in children 814, 815 prognosis/treatment 815 prostatic involvement 57 Adamantinomatous craniopharyngiomas 706 – 7, 707, 708 Adenocanthoma see Adenosquamous carcinoma Adenocarcinoid tumors, appendix 413 Adenocarcinoma aggressive digital papillary 578
bladder 21 – 3 breast tumors pathology 182, 182 – 3 WHO classification 181 cervical 504 clear cell variant 23, 507 colloid 506 – 7 endometrial serous adenocarcinoma 494, 494 – 6 endometrioid 505 fallopian tube see under Fallopian tube tumors gastrointestinal appendix 412 – 3 bile duct 385, 386 bowel 401 colorectal 749 esophagus 337, 340, 341 – 2 gallbladder 383 small bowel 392 – 4 intestinal-type 506 – 7 pancreatic 367 – 8 cystic 371 ductal 367 – 8, 754 pancreatoblastoma 754, 755 pediatric malignancy 754, 755, 755 pleomorphic 370 – 1 pediatric malignancy lung parenchyma 739 pancreatic 754, 755, 755 prostate 764 prostate, pediatric malignancy 764 rete testes (ART) 74 – 5 clinical features 74 differential diagnosis 74 epidemiology 74 investigation/staging 74 – 5 treatment/prognosis 75 salivary glands 92, 93 serous papillary 507 – 8 signet-ring cell 506 – 7 in situ, 479 undifferentiated thymic carcinoma 244 urethra female 29 microscopic appearance 28 – 9 vagina tumors 527 see also specific tumors/sites Adenocystic basal cell cancer see Adenoid cystic carcinoma Adenocystic basaloid carcinoma see Adenoid cystic carcinoma Adenofibroma metanephric tumors 11, 12 nephrogenic in children 763 Adenoid basal carcinoma (ABC), cervical 509, 510 Adenoid cystic carcinoma 578 basaloid squamous cell carcinoma vs. 106, 322
breast see Adenoid cystic carcinoma of the breast cervical 509 – 10 cutaneous 579 esophagus 340 – 1 oral cavity 579 salivary glands see under Salivary gland tumors pediatric malignancy, airways 737 – 8 pulmonary see Adenoid cystic carcinoma of the lung sites 187 Adenoid cystic carcinoma of the breast 187 – 93 biology 187 classification 187 – 8 clinical presentation/diagnosis 187, 190 – 1 definitions/terminology 187 differential diagnosis collagenous spherulosis vs. 189, 190 invasive cribriform carcinoma vs. 187, 189 – 90 epidemiology 187 grade 190 local recurrence 191, 192 males 187 metastatic potential 191, 192 pathology 187 – 90 atypical 190 – 1 histopathology 188, 188 – 9, 189, 190 hormone receptor expression 188, 191 microglandular adenosis and 187, 189, 190 in situ pattern 188 – 9 solid tumors 189, 190 prognosis 190 – 1 treatment 191 – 2 primary tumors 191 – 2 recurrence/metastases 192 Adenoid cystic carcinoma of the lung 321 – 8 age-at-onset 322 clinical characteristics 322 – 3 diagnosis 323 – 4 differential diagnosis/misdiagnosis 323 basaloid squamous carcinoma vs. 322 mucoepidermoid carcinoma vs. 330 gene expression 322 growth patterns 322, 322 histogenesis/pathology 321 – 2 natural history 322 pediatric malignancy 737 – 8 presentation 323 prognosis 324 – 5 factors affecting 325 – 6 growth patterns and 322, 322 histology and 326 tumor grade and 325 – 6, 326 recurrence/metastasis 324, 326 risk factors 323
Textbook of Uncommon Cancer Third Edition. Edited by D. Raghavan, M.L. Brecher, D.H. Johnson, N.J. Meropol, P.L. Moots, P.G. Rose. Associate Editor: I.A. Mayer. 2006 John Wiley & Sons, Ltd. ISBN: 0-470-01202-1
822 Adenoid cystic carcinoma of the lung (cont.) subtypes 322 survival rates 324, 326 adjuvant radiation therapy and 325 symptoms/signs 323 treatment 324 – 5 chemotherapy 325 interventional bronchoscopy 325 radiation therapy 325 surgical resection 324 targeted therapy 325 Adenoma adenoid cystic carcinoma as 321 adrenocortical see Adrenocortical adenoma bronchial 329 pediatric malignancy 738 definition 392 insulinoma 756 nephrogenic 28 oncocytic see Oncocytoma (oncocytic adenoma) pituitary, Carney complex and 784 renal metanephric tumors 11, 11, 12 papillary 2 small bowel 392 see also specific types/locations Adenoma – carcinoma sequence, bile duct tumors 385 Adenoma malignum (minimal deviation adenocarcinoma) 506, 506 Adenomatous polyposis coli (APC), PNET and 696 Adenomyoepithelioma see Adenoid cystic carcinoma Adenosarcoma M¨ullerian, children 767 with sarcomatous overgrowth, uterine 489 uterine 489, 489 Adenosquamous carcinoma bile duct 386 breast 181 pathology 182, 182 – 3 prognosis 184 WHO classification 181 cervical 508 – 9 esophageal 341 – 2, 342 gallbladder 384 gastric 354 – 5 pancreatic 369 – 70 salivary glands 92 – 3, 95 Adipose tumors see Liposarcoma Adjuvant therapy adenoid cystic carcinoma of the lung 325 ameloblastic carcinoma 727 anaplastic thyroid carcinoma 168 bile duct tumors 386 borderline ovarian tumors 451 – 2 cervical melanoma 513 cervical small cell carcinoma 511 colorectal carcinoma in children 752 – 3 gallbladder tumors 383 – 4, 384 meningioma 643, 643 nasopharyngeal carcinoma 126 prostate sarcomas 53 serous papillary adenocarcinoma 508 thymic carcinoma 249 uterine sarcoma 490 – 2, 491 recurrence/metastases following 491 vulvar squamous cell carcinoma 523 – 4 see also Biologic agents; Chemotherapy; Radiation therapy Adolescents carcinoid tumor of the rectum 751 colorectal carcinoma 751 – 3 embryonal rhabdomyosarcoma 512 endocrine tumors 775 – 97 primary germ cell tumors 650 Adrenalectomy, male breast cancer therapy 206
SUBJECT INDEX Adrenal glands atrophy 144 catecholamine production 143 clinical presentations of disease 143 adrenogenital syndrome see Adrenogenital syndrome Conn’s syndrome see Conn’s syndrome Cushing’s syndrome see Cushing’s syndrome development 775 – 6 fetal cortex 776 hyperfunction (hypercorticalism) 143 hyperplasia 143, 144, 147 congenital 766 metastases 160 neoplasia 143 – 63 adrenocortical tumors see Adrenocortical tumors “incidentalomas,” 143, 160, 160 medullary tumors see Pheochromocytoma nonfunctioning carcinomas 143 nodules 147, 150, 160 PPNAD 783, 784 structure/function 148, 150 Adrenocortical adenoma “black,” 144 carcinoma etiology and 149 Conn’s syndrome and 147, 147 – 8 typical morphology 148 Cushing’s syndrome and 144, 144, 145 diagnosis biochemical 151 – 2 imaging 152 – 3 etiology 149 – 50 pathology adrenogenital syndrome 145 – 6, 146 Conn’s syndrome 147, 147 – 8 Cushing’s syndrome 144, 144 prognosis/treatment 153 – 5 adrenogenital syndrome 146 Conn’s syndrome 148 Cushing’s syndrome 145 Adrenocortical carcinoma adenomas and 149 adrenogenital syndrome and 145, 146, 147, 777, 777 – 8 children/adolescents 775 – 81, 777, 778 Conn’s syndrome and 148, 148 Cushing’s syndrome and 144, 145 diagnosis 150, 150 biochemical 151 – 2 imaging 152, 152 – 3, 153 etiology 149 – 50 metastases 145, 780, 780 mixed type 778 “nonfunctioning,” 148 – 9 biochemical diagnosis 152 pathology adrenogenital syndrome 146, 779 Conn’s syndrome 148, 148 Cushing’s syndrome 144 – 5, 145 prognosis/treatment 153 – 5 adrenogenital syndrome 147 children/adolescents 780 – 1 Conn’s syndrome 148 Cushing’s syndrome 145 “nonfunctioning” carcinomas 149 staging 153, 781, 781 Adrenocortical tumors 143 – 55 ACTH dependence 150 – 1 adenoma see Adrenocortical adenoma adrenogenital syndrome see Adrenogenital syndrome carcinoma see Adrenocortical carcinoma children/adolescents 145, 775 – 87 clinical features 145, 145, 145 – 6, 146 Cushing’s syndrome 144, 144, 775, 778 diagnosis 778 – 80, 779, 780 epidemiology 775, 776, 777 pathogenesis 775 – 7, 776
prognosis 781 treatment 780 Conn’s syndrome see Conn’s syndrome Cushing’s syndrome see Cushing’s syndrome diagnosis/differential diagnosis 151 – 3 adenomas vs. nodules 147 benign vs. malignant 145, 150, 150 biochemical diagnosis 151 – 2, 778 – 9 in children/adolescents 778 – 80 computed tomography 152, 152, 779 – 80, 780 histological 778, 779 MRI 152, 152, 779 – 80 PET 152 – 3, 153, 779, 780 primary investigations 153 selective venous sampling 153 etiology 776, 776 – 7 etiology/growth promotion 149 – 50 familial 149 “nonhormonal,” 148 – 9, 152 oncocytic variants 149 staging/classification 153, 778, 781, 781 steroidogenesis and 150 – 1 treatment 153 – 5 children/adolescents 780 o, p -DDD (Mitotane) 153 – 4, 154, 780 platinum chemotherapy 154, 154, 780 radiation therapy 154 – 5 streptozocin 154 surgical 153, 780 virilization 143, 145, 777, 777 – 8, 778, 779 Adrenocorticotropic hormone (ACTH) adrenocortical tumors 150 – 1, 777 islet cell tumors 756 large cell neuroendocrine carcinoma (LNEC) of the lung 301 small cell carcinoma of the pancreas 372 small cell undifferentiated carcinoma of prostate 38, 41 Adrenogenital syndrome 143, 145 – 7 biochemical diagnosis 151 – 2 children/adolescents 777, 777 – 8 clinical presentation 145 differential diagnosis 145 feminization 143, 145, 146, 146 pathology 145 – 6, 146, 779 prognosis/treatment 146 – 7 steroidogenesis and 151 testicular tumors in children 766 virilization 143, 145, 151 – 2 Adult T cell leukemia/lymphoma (ATLL) 561 – 2 Age/aging lung sarcoma, primary 264 male breast cancer 201 nasopharyngeal carcinoma 114 stromal tumors of the ovary 456 Aggressive angiomyxoma, clitoris/vulva in pediatric malignancy 767 Aggressive digital papillary adenocarcinoma 578 Aicardi’s syndrome, choroid plexus papilloma/carcinoma 668 AIDS see HIV infection/AIDS Air bronchograms, primary pulmonary lymphoma 259 Airway tumors, pediatric 737 – 9 see also specific tumors Albinism, pediatric basal cell carcinoma 811 Alcohol epitheliectomy, squamous cell carcinoma of the conjunctiva 714 Alcohol intake nasopharyngeal carcinoma 115 pancreatic cancer and 371 Alcohol withdrawal, pheochromocytoma diagnosis and 157 Aldosteronism (primary) see Conn’s syndrome
SUBJECT INDEX ALK protein, anaplastic large cell lymphoma 564 Alkylating agents, thymomas/thymic carcinomas 250, 251 Alpha-fetoprotein (AFP) hepatoid adenocarcinoma 353 – 4 ovarian germ cell tumors 471 Sertoli-stromal cell tumors 462 yolk sac tumor 468 Alport syndrome 753 Alveolar carcinoma see Bronchioloalveolar carcinoma Alveolar rhabdomyosarcoma 269 cutaneous 815 Alveolar soft part sarcoma, pediatric malignancy cervix 767 uterus 768 vagina 768, 768 Ameloblastic carcinoma 726, 726 – 7 Ameloblastoma 88, 723 – 6 biology 724 clinical presentation 724 differential diagnosis 725 embryology 723 epidemiology 723 evaluation 724 – 5 malignant see Malignant ameloblastoma management 90 pathology 724 presentation 88 treatment 725 – 6 American Academy of Dermatology Task Force, pediatric melanoma treatment 811 American Joint Committee on Cancer (AJCC) nasopharyngeal carcinoma 120 urethral cancer 30 vulvar squamous cell carcinoma 522 American National Cancer Institute, breast carcinosarcoma epidemiology 218 AMG706 426 Amine precursor uptake and decarboxylation (APUD) 394, 431 AMN107 426 Amosite (brown asbestos), mesothelioma and 279, 280 Amphiboles, mesothelioma and 279 Amphicrine carcinomas, appendix 413 Ampulla of Vater, small cell carcinomas 433 Amsterdam Criteria, HNPCC 752 Amyloid-like material, nasopharyngeal carcinoma 117 Amyloid, soft tissue sarcoma vs. 97 Anal carcinoma 406 – 7 clinical presentation/ diagnosis 406 predisposing factors 406 staging 406 treatment 406 – 7 Anal melanoma 407 – 8 Anaplastic carcinoma of the paranasal sinuses 727 – 8 Anaplastic large cell lymphoma (ALCL) 563, 564 cutaneous lymphoma in children 813, 814 Anaplastic oligodendroglioma 690, 691 Anaplastic thyroid carcinoma 168 – 9 clinical presentation/diagnosis 168 treatment 168 – 9 Androblastomas see Sertoli – stromal cell tumors Androgen receptors, meningiomas 639 Androgens gynandroblastoma 463 see also specific hormones Androstenedione, adrenogenital syndrome 146 Anemia, complete hydatidiform moles 535 Angiocentric lymphoma (AIL), children 765 – 6
Angiodysplasia, metanephric tumors 12 Angio-embolization, renal angiomyolipoma 8–9 Angiogenesis inhibitors, pleural mesothelioma 285 Angiogenesis, mesothelioma 282 Angiography meningeal sarcomas 628 meningioma 642 pancreatic adenocanthoma (adenosquamous carcinoma) 369 Angioimmunoblastic T cell lymphoma (AITL) 563 – 4 Angioleiomyosarcoma, renal, children 762 Angiomatoid malignant fibrous histiocytoma 580 Angiomyolipoma epithelioid 8 renal 7 – 9 background 7 clinical presentation 8 metastases 7, 8 pathology 7, 7 – 8 pregnancy and 9 thrombus formation 9 treatment/prognosis 8 – 9 tuberous sclerosis and 7, 8 Angiomyxoma, clitoris/vulva, pediatric malignancy 767 Angiosarcoma 170 cutaneous 581 epithelioid 98, 581 hepatic, primary 387 lung 270, 270 meningeal 633 oral cavity and adjacent structures 97 Angiotropic lymphoma (intravascular lymphomatosis) 258 Ann Arbor staging system, cervical lymphoma 514 Anthophyllite, mesothelioma and 279, 280 Anthracyclines, Merkel cell carcinoma 598 Antibody dependent cellular cytotoxicity (ADCC), nasopharyngeal carcinoma 118 Anti-CD20 monoclonal antibody see Rituximab (anti-CD20 monoclonal antibody) Antiestrogen agents meningioma 644 tamoxifen see Tamoxifen see also specific drugs Antiprogesterone agents recurrent meningioma 644 see also specific drugs Antiretroviral therapy (HAART), Kaposi’s sarcoma (KS) 275 Antrum, carcinoid tumors 358 Apocrine glands carcinoma 578 mixed eccrine gland tumors 578 – 9 Apoptosis, large cell neuroendocrine cancer (LNEC) of the lung 299 Appendectomy appendiceal adenocarcinoma 412 carcinoid tumors 403, 751 Appendiceal carcinoid 403, 404, 414 – 5, 749, 750, 750 – 1 Appendicitis, carcinoid tumors and 414, 750, 751 Appendix 410 – 7 adenocarcinoid 413 adenocarcinoma 412 – 3 benign mucinous neoplasms 410 carcinoid tumors 403, 404, 414 – 5, 749, 750, 750 – 1 observed survival 411 pseudomyxoma peritonei 410 – 2, 411 AR gene, male breast cancer 202 Armed Forces Institute of Pathology (AFIP), yolk sac tumor 468
823 Aromatase inhibitors male breast cancer therapy 206 metaplastic breast carcinoma 181 Arterial embolization colon/rectum carcinoid tumors 404 liver metastases 415 Arterial thrombosis, osteosclerotic myeloma 572 Arthritis, acinar cell carcinoma syndrome 370 Arytenoid 103 Asbestos, mesothelioma and 279, 280, 280 Ascites, ovarian germ cell tumors 471 Asian patients, nasopharyngeal carcinoma treatment 121 Asthma adenoid cystic carcinoma vs. 323 carcinoid syndrome 750 Astrocytic neoplasms 681 – 9 astrocytoma see Astrocytoma glioblastoma 674, 678, 680, 681 – 3 variants 684 – 9 Astrocytoma 684, 688 imaging 679 low-grade 676 pediatric malignancy 799 desmoplastic 798 – 9 Atrioventricular (AV) node, benign mesothelioma 288 Atypical adenomatous hyperplasia (AAH) 314 Atypical Cushing’s syndrome (ACS) 784 Atypical ductal hyperplasia 230, 232 Atypical fibroxanthoma 581 Atypical proliferative ovarian tumors see Borderline ovarian tumors Atypical teratoid/rhabdoid tumors (AT/RT) 805 – 6 medulloblastoma 698 Auditory canals, pretreatment assessment 128 Autoimmunity hairy cell leukemia 548 Langerhans’ cell histiocytosis 610 thymoma and 246 see also specific conditions Autologous stem cell transplantation mantle cell leukemia 550 Waldenstr¨om macroglobulinemia 571 Autologous tumor vaccines, bronchioloalveolar carcinoma 318 Avastin (bevacizumab), colon/rectum carcinoid tumors 404 Axilla, phyllodes tumors 209 Bacillus Calmette – Guerin (BCG), transitional cell carcinoma of prostate 56, 57 Bannayan – Riley – Ruvalcaba (BRR) syndrome Carney complex and 785 Cowden disease and 786 Barrett’s esophagus, basaloid squamous carcinoma 340 Basal cell carcinoma (BCC) 811 pediatric malignancy 811 – 2 Gorlin’s syndrome and 811 – 2, 812 photodynamic therapy 812, 813 predisposing factors 811 – 2, 812 salivary glands 92 trichilemmal carcinoma vs. 577 vulva tumors 525 Basaloid squamous cell carcinoma (BSCC) 340 adenoid cystic carcinoma vs. 106, 322 laryngeal 106 lung 322 thymic carcinoma 242, 242, 243 Basaloma see Adenoid cystic carcinoma Basic fibroblast growth factor (bFGF), meningiomas 639
824 Basophilic leukemia, acute 547 Basosquamous carcinoma see Basaloid squamous cell carcinoma (BSCC) Bax, large cell neuroendocrine cancer (LNEC) of the lung 299 B-cell lymphoma primary CNS 658 prostatic involvement 58 testicular 67 B-cells (lymphocytes) antigens 548 neoplasms 557 – 60 see also specific types Bcl-2 gene/protein large cell neuroendocrine cancer of the lung 299 Merkel cell carcinoma 599 thymomas/thymic carcinoma 245 BCNU implantable (Gliadel), glioblastoma 683 BCR-ABL fusion oncoprotein 405 Beckwith – Weidman syndrome, adrenocortical tumors 149 Bednar tumors 580, 589, 815, 816 Benign lesions adrenocortical, malignant vs. 145, 150, 150 appendix 410 leiomyoma 494 lung 273 meningioma 638 mesothelioma (localized fibrous) 288 pheochromocytoma 155, 156 phyllodes tumor of the breast 209, 211, 211 – 2 salivary gland tumors 91, 92, 93 thymomas 237, 240, 240, 245, 248 urethral 28 see also specific types/locations Benign prostatic hyperplasia (BPH), transitional cell carcinoma of prostate 55 Bergh staging system, thymomas/thymic carcinomas 247 Bernatz classification, thymomas 241 Bevacizumab (Avastin ), colon/rectum carcinoid tumors 404 Bifrontal craniotomy 140 Bile duct tumors 385 – 7 adjuvant therapy 386 advanced/metastatic disease 386 clinical presentation/diagnosis 385, 385 epidemiology/risk factors 385 pathology 385 supportive care 386 surgical resection 386 Bile production, hepatoid adenocarcinoma 354 Bing – Neel syndrome 570 Biochemical diagnosis see Laboratory evaluation Biologic agents cervical melanoma 513 dermatofibrosarcoma protuberans 591 – 2 Merkel cell carcinoma 599 nasopharyngeal carcinoma, recurrent/metastatic 129 phyllodes tumor of the breast 214 pulmonary carcinoid tumors 310 stromal tumors of the ovary 457 – 8 see also specific drugs/drug types Biologically effective dose (BED), nasopharyngeal carcinoma 128 Biomarkers see Tumor markers Biopsy basal cell carcinoma, pediatric 812 bone tumors, oral cavity 89 dermatofibrosarcoma protuberans 815 fine needle aspiration see Fine needle aspiration (FNA) biopsy/cytology gastric lymphomas 362
SUBJECT INDEX large cell neuroendocrine carcinoma of the lung 302 – 3 lymphoma primary pulmonary 260 – 1 primary thyroid 169 melanoma, oral cavity 91 nasopharyngeal carcinoma 119 neurocutaneous melanosis 606 – 7 ovarian tumors borderline 450 stromal 457 phyllodes tumor of the breast 213, 216 primary intracranial germ cell tumors 653 pulmonary carcinoid tumors 309 salivary glands 91, 93 sebaceous carcinoma of the eyelid 713 sentinel lymph nodes see Sentinel lymph node biopsy/dissection thymomas/thymic carcinomas 247 transitional cell carcinoma of prostate 55 transoral brush 118 Biotherapy see Biologic agents Birbeck granules, Langerhans’ cell histiocytosis (LCH) of CNS 610 Birt – Hogg – Dube syndrome, oncocytoma 6 Bisphosphonates, primary plasma cell leukemia 549 “Black” adenoma 144 Bladder tumors adenocarcinoma 21 – 3 clear cell variant 23 carcinoma in situ (CIS), prostate TCC and 53, 54 lymphoepithelioma 23 – 4 micropapillary cancer 21 papillomas 28 pediatric 763 – 4 sarcoma 25 sarcomatoid carcinoma 18 – 9 small cell carcinoma 19 – 21 squamous carcinoma 23 urachal cancer 24 – 5 Blastic NK cell lymphoma 560 Bleeding gastrointestinal gastric choriocarcinoma 355 – 6 gastric lymphomas 362 rectal, anal carcinoma and 406 imatinib mesylate and 423 vaginal see Vaginal bleeding, abnormal see also Hemorrhage Bleomycin nasopharyngeal carcinoma, recurrent/metastatic 129 pulmonary fibrosis 473 verrucous carcinoma 339 see also specific combinations Bleomycin, etoposide, and cisplatin (BEP), stromal tumors of the ovary 458, 458, 460 Blood-brain barrier neuroepithelial neoplasms 679 radiation disruption 682 Blood lakes, thymomas 240 Blood supply, craniopharyngiomas 705 Blood – thymus barrier 237 Blue asbestos (crocidolite), mesothelioma and 279, 280 “Blueberry muffin” lesions congenital dermatofibrosarcoma protuberans 815 congenital rhabdomyosarcoma 815 leukemia cutis in children 814 – 5 neuroblastoma 816 BMS 354825, 426 Body cavity lymphomas 559 Bone age, adrenocortical tumors and 778 Bone marrow chronic neutrophilic leukemia 543 gastric lymphomas and 362
mast cell leukemia 551 panmyelosis with myelofibrosis, acute 547 T-large granular lymphocyte leukemia 551 transplantation, pineoblastoma 804 Bone metastasizing renal tumor of childhood 760 – 1 Bone scans colon/rectum carcinoid tumors 403 – 4 small cell undifferentiated carcinoma of prostate 44 Bone tumors Ewing’s sarcoma see Ewing’s sarcoma metastases/metastatic 89, 760 – 1 oral cavity and adjacent structures 87 – 90, 88, 89 diagnosis 89 – 90 management 90 presentation 87 – 9 see also specific types/locations Bony lesions, osteosclerotic myeloma 572 ‘Boomerang’ technique 125, 125 Borderline ovarian tumors 447 – 54 Brenner 449 clear cell 449 clinical presentation/diagnosis 449 – 50 endometrioid 449 endometriosis and 449 epidemiology 447 microinvasion 448 mucinous tumors endocervical-like 449 gastrointestinal 448 – 9, 449 pediatric malignancy 769 pathology 447 – 9 gross 447 – 8, 448, 449 histology 448, 448, 448 – 449, 449, 449 prognosis 447, 450 recurrence rates 451 risk factors 447 serous tumors 447 – 8, 448 pediatric malignancy 769 staging 447, 450 treatment adjuvant 451 – 2 advanced/recurrent disease 451 chemotherapy 451 – 2 early stage disease 450 – 1 fertility preservation 450, 452 surgical 450 – 1 Borderline phyllodes tumor of the breast 210, 212, 215 – 6 Boron-neuron capture 682 Bortezomib bronchioloalveolar carcinoma 318 primary plasma cell leukemia 549 Bowel cancer colorectal see Colorectal carcinoma small bowel see Small bowel cancer Brachytherapy chordomas 622 endometrial carcinoma 495 intracavitary, urethral cancer 33 meningioma 644 pleural mesothelioma 285 – 6 uveal melanoma 716 BRAF mutations anaplastic thyroid carcinoma 168 pediatric thyroid cancer 786 Brain stem gliomas 687 – 8, 799 lesions in children 681 Brain tumors cutaneous manifestations, children 817 germ cell tumors 649 – 56 meningeal sarcomas 023: 1 – 12 metastatic 43 pediatric see under Pediatric malignancy retinoblastoma and 718 see also specific types/locations Brain Tumor Study Group (BTSG) 681
SUBJECT INDEX BRCA genes extra-ovarian primary peritoneal carcinomas 438 male breast cancer 202, 206 pancreatic cancer and 367 primary adenocarcinoma of fallopian tube 478 Breast fibroepithelial lesions 209 fibrosclerotic lesions 181 hyperplasia 230, 232 lymphocytic lobulitis 196 Breast cancer 218 adenoid cystic carcinoma see Adenoid cystic carcinoma of the breast classification 187 – 8 WHO 181 Cowden disease 786 differential diagnoses 222 ductal carcinoma in situ, 219, 222, 230, 232 male 203, 205 invasive cribriform carcinomas 187, 189 – 90, 230, 233 invasive ductal carcinomas 222 juvenile granulosa cell tumors and 460 lobular carcinoma in situ, 230 male see Male breast cancer (MBC) metaplastic carcinomas see Metaplastic carcinoma of the breast metastatic/secondary 347, 347 non-Hodgkin’s lymphoma see Non-Hodgkin’s lymphoma (NHL), breast phyllodes tumor see Phyllodes tumor of the breast radiation-associated sarcoma 49 small bowel metastases 398 tubular areas in 232 tubular carcinoma see Tubular carcinoma see also specific types Breast carcinosarcoma 218 – 29 clinical presentation/diagnosis 222 – 3 clinical studies 220 – 1 differential diagnosis 222 epidemiology 218 – 9 etiology 218 – 9 historical background 218 imaging 223 molecular biology and potential targets 226 – 7 pathology 219, 219 – 22 fibroepithelial lesions 219 prognosis/outcomes 223, 224 terminology/definitions 218, 222 treatment 223 – 6 chemotherapy 225 – 6 radiation therapy 226 surgical 225 Breast conserving therapy (BCT) adenoid cystic carcinoma of the breast 191 breast carcinosarcoma 225 male breast cancer 204 metaplastic breast carcinoma 181, 185 phyllodes tumor 213 – 4 Brenner borderline ovarian tumors 449 British Testicular Tumour Panel (BTTP) non-Hodgkin’s lymphoma 66 pathology 67 prognosis/relapse 67 – 8 paratesticular rhabdomyosarcoma 77 Sertoli cell tumors 72 pathology 73 treatment 74 Bromodeoxyuridine (BUDR) labeling index, meningioma 640, 644 Bronchial adenomas 329 pediatric 738 Bronchial carcinoma, pediatric 738, 738 – 9 Bronchial tumors
mucoepidermoid 329 – 30 pediatric 737 – 9 see also specific tumors Bronchioloalveolar carcinoma 313 – 20 biology 313 – 4 classification 315 clinical presentation/diagnosis 315, 315 – 6 epidemiology 313 multifocal 316 pathology 314 – 5 pediatric 739 pneumonic 316 radiology 316 treatment 316 – 8 investigational therapies 318 surgery 316 – 7 systemic therapy 317, 317 – 8 Bronchioloar carcinoma see Bronchioloalveolar carcinoma Bronchoalveolar lavage (BAL), primary pulmonary lymphoma diagnosis 260 Bronchogenic carcinoma differential diagnosis 272 – 3 pediatric, lung parenchyma 739 Bronchorrhea 315 – 6 Bronchoscopy interventional, adenoid cystic carcinoma of the lung 325 mucoepidermoid carcinoma 331 primary pulmonary lymphoma diagnosis 260 pulmonary carcinoid tumors 309 small cell carcinomas of the gastrointestinal tract 431 Bronchus-associated lymphoid tissue (BALT), primary pulmonary lymphoma 258 Brooke – Spiegler syndrome, malignant cylindroma 578 Brown asbestos (amosite), mesothelioma and 279, 280 Brown – Sequard syndrome 688 B symptoms, primary pulmonary lymphoma 259 Burkitt’s lymphoma 559 – 60 breast lymphoma 195, 196, 198 nasopharyngeal carcinoma 115 ovarian involvement, children 770 small bowel 396 Burt – Hogg – Dub´e syndrome, endocrine cancers in children 781 CA125 borderline ovarian tumors 450 extra-ovarian primary peritoneal carcinomas 440 E-Cadherin, breast carcinosarcoma 227 Caf´e au lait spots, Carney complex and 784 Calcification adrenocortical carcinoma 144 ganglion cell neoplasms 686 intracranial germ cell tumors 651 meningeal chondrosarcomas 631 Calcifying epithelial odontogenic tumors (CEOTs) 725 Calcimimetics, parathyroid carcinoma 177 Calcitonin elevation, management 167 hypersecretion in MEN 2 783 medullary thyroid carcinoma and 166 neuroendocrine laryngeal tumors and 110 small cell undifferentiated carcinoma of prostate and 41 Calcium channel blockers, pheochromocytoma surgery and 159 Calcium homeostasis, parathyroid carcinoma 174 Calcium levels, parathyroid carcinoma 176, 176 Calcium-reducing agents, parathyroid carcinoma 177
825 Cancer and Leukemia Group B (CALGB), pleural mesothelioma prognosis 287 Cancer-specific survival (CSS) small cell carcinoma of the bladder 20, 20 small cell undifferentiated carcinoma of prostate 44 – 5 Capillary hemangiomas, urethra 28 Carbohydrate metabolism glucagonoma 375 pheochromocytoma 155 Carboplatin extra-ovarian primary peritoneal carcinomas 443 medulloblastoma 700 see also specific combinations Carboplatin/paclitaxel, mucoepidermoid carcinomas 333 Carcinoembryonic antigen (CEA) acinar cell carcinoma 370 bladder adenocarcinoma 22 – 3 cervical mucinous adenocarcinomas 504 medullary thyroid carcinoma 166 small cell undifferentiated carcinoma of prostate 41 Carcinogenesis, bile duct tumors 385 Carcinogens, male breast cancer 202 Carcinoid syndrome 308, 377, 750 appendiceal carcinoid tumors 415 pancreatic tumors 377 small bowel cancer 391, 392, 394 – 5 Carcinoid tumorlets 308 Carcinoid tumors atypical 109, 302 carcinoid syndrome see Carcinoid syndrome cervical 511 classification 109 differential diagnosis large cell neuroendocrine carcinoma vs. 302 pancreatic microadenocarcinoma vs. 371 ’functioning,’ children 750 gastrointestinal appendiceal 377, 403, 404, 414 – 5, 749, 750, 750 – 1 bile duct 385, 386 – 7 colorectal 749 – 51 esophageal 342 – 3 gallbladder 385 gastric 357 – 60, 358 hepatic 388 pancreatic 377 – 8 pediatric malignancy 749 – 51 rectal 377, 751 small bowel 377, 394 – 5 mediastinal 734 metastases 10, 377, 394 prostate, pediatric malignancy 764 pulmonary see Pulmonary carcinoid tumors renal 9 – 10 background 9 clinical presentation 9 – 10 oncocytic in children 762 pathology 9 treatment/prognosis 10 thymus 237 treatment 377 true 109 Carcinoma definition 392 differential diagnosis 273 see also specific sites/types Carcinoma adenoides cysticum see Adenoid cystic carcinoma Carcinoma ex pleomorphic carcinoma, salivary glands 92 Carcinoma in situ (CIS) bladder, prostate TCC and 53, 54 cervical, pregnancy and 517
826 Carcinosarcoma 104 bile duct tumors 387 breast see Breast carcinosarcoma cervical 511 definition 181, 218, 222 esophagus 339, 339 – 40 gallbladder tumors 385 laryngeal tumors 3 – 4 monoclonicity 226 prostate 48 uterine 488, 488 – 9 chemoradiation therapy 491 – 2 chemotherapy 492, 492 – 3, 493 see also Spindle cell (sarcomatoid) carcinoma Cardiac arrhythmias, pheochromocytoma 155, 787 Cardiac fibroma 736, 737 Cardiac hamartoma (oncocytic (histiocytoid) cardiomyopathy) 736, 737 Cardiac lipoblastoma 736 Cardiac myxoma 736, 783 Cardiac rhabdomyoma 736 Cardiac transplant, rhabdomyoma 736 Cardiac tumors benign mesothelioma 288 hamartoma 736 mesothelioma of the pericardium 287 myxomas in Carney complex 783 primary neoplasms 736 – 7 see also specific types Carney complex (CNC) 781, 783 – 5 clinical presentation 783 – 4, 786 molecular genetics 784 – 5 PPNAD 783, 784 testicular tumors 74, 784 Carney triad, stromal tumors 360 – 1 β-Carotene intake, hydatidiform moles and 533 Castleman’s disease, osteosclerotic myeloma 572 Catecholamine production adrenal disorders 143, 155 pheochromocytoma 143, 155, 157 – 8, 787, 789 – 90 see also specific molecules Caustic ingestion, verrucous carcinoma 338 Cavernous sinus, meningioma 641 Cavernous syndrome, meningioma associated 641 CCNU glioblastoma 683 medulloblastoma 700 CD4 cells, thymus 238 CD5 cell marker, thymic carcinoma 244 CD8+ lymphocytes Langerhans’ cell histiocytosis (LCH) of CNS 611 thymus 238 CD34 cell marker, gastrointestinal stromal cell tumors 419 CD56 cell marker, large cell neuroendocrine cancer (LNEC) of the lung 300 CD117 cell marker dysgerminoma 468 gastric stromal tumors 360 gastrointestinal stromal cell tumors 419 2-CdA, hairy cell leukemia 549 Celiac disease (gluten-sensitive enteropathy) small bowel adenocarcinoma 392 small bowel lymphoma 396 Cell cycle regulation large cell neuroendocrine cancer (LNEC) of the lung 299 parathyroid carcinoma 175 Cellular fibromas/fibrosarcomas, ovarian 461 Central nervous system (CNS) brain tumors see Brain tumors congenital anomalies, neurocutaneous melanosis and 605, 606
SUBJECT INDEX infection, neuroepithelial neoplasms 675 Langerhans’ cell histiocytosis 610 – 3 melanin 605 melanoma 811 metastases extra-ovarian primary peritoneal carcinomas 440 gestational trophoblastic neoplasia 537 pediatric, pleuropulmonary blastoma 742 primary CNS lymphoma 659 retinoblastoma spread 718 trilateral retinoblastoma 804 neoplasms, incidence 675 pediatric malignancy see under Pediatric malignancy primary lymphoma see Primary central nervous system lymphoma (PCNSL) see also specific sites Central node dissection, medullary thyroid carcinoma (MTC) 166 Cerebellopontine angle, meningioma 641 Cerebellum, pediatric astrocytomas 799 Cerebral atrophy intracranial germ cell tumors 651 Langerhans’ cell histiocytosis (LCH) of CNS 612 Cerebral radiation necrosis, esthesioneuroblastoma 137 Cerebrospinal fluid (CSF) choroid plexus papilloma/carcinoma 667, 669, 801 cytology medulloblastoma 701 primary CNS lymphoma 658 melanotic meningeal lesions 605, 606, 806 meningioma dissemination 645 pineoblastoma 804 Cervical conization 514 Cervical intraepithelial neoplasia (CIN), pregnancy and 517 Cervical lymph node, parathyroid carcinoma 176 Cervical punch biopsy, adenoma malignum 506 Cervical stump cancer 515 – 6 Cervical tumors 501 – 20 history 501 immunocompromised host 516 lymphomas 493 pediatric malignancy 767 – 8 pregnancy 19, 516 – 7 management 517f squamous cell carcinomas 502 – 4 suboptimal surgery 515 – 6 uncommon histologies 501 – 14 curiosities 514 epithelial tumors 508 – 11 glandular tumors 503 – 8 lymphoma 513 – 4 melanoma 513 sarcomas 511 – 3 uterine sarcoma following radiation therapy for 485 see also specific types Cetuximab bronchioloalveolar carcinoma 318 nasopharyngeal carcinoma, recurrent/metastatic 129 Charged particle irradiation, chordomas 621 – 2 CHEK2 gene, male breast cancer 202 Chemodectomas see Paraganglioma Chemoradiation therapy anal carcinoma 407 bile duct tumors 386 gallbladder tumors 384 Merkel cell carcinoma (MCC) 598 nasopharyngeal carcinoma 127 primary thyroid lymphoma 169
stomach cancers 352 thymomas/thymic carcinomas 250 – 2 uterine sarcoma 491 – 2 whole-brain radiation therapy (WBRT) and 662 Chemotherapeutic retroconversion ovarian germ cell tumors 472 pediatric, pleuropulmonary blastoma 742 Chemotherapy breast cancers carcinosarcoma 225 – 6 male breast cancer 205, 206 metaplastic breast carcinoma 184 – 5 non-Hodgkin’s lymphoma 197 phyllodes tumor of the breast 214 tubular carcinoma 233 cutaneous malignancy angiosarcoma 581 fibrosarcoma in children 815 Merkel cell carcinoma 584, 598, 598 – 9 endocrine tumors adrenocortical 154, 154 anaplastic thyroid carcinoma 169 malignant hemangioendothelioma 170 medullary thyroid carcinoma 167 parathyroid carcinoma 177 pheochromocytoma 159 gastrointestinal malignancy appendiceal adenocarcinoma 413 appendiceal carcinoid tumors 415 bile duct tumors 386 carcinoid 377 colorectal carcinoma in children 752 – 3 colorectal lymphoma 406 esophageal adenoid cystic carcinoma 341 esophageal choriocarcinoma 345 – 6 esophageal melanoma 346 – 7 esophageal small cell carcinomas 432 extrapulmonary small cell carcinomas of the pancreas 372 gallbladder tumors 384, 384 gastric endocrine cell proliferations 359 gastric lymphomas 363 gastrointestinal stromal cell tumors 345, 422 hepatic 387, 388 hepatoid adenocarcinoma 354 intestinal leiomyosarcoma 405 – 6 islet cell tumors of the pancreas 376 – 7 poorly differentiated esophageal small cell carcinoma 343 pseudomyxoma peritonei 411, 412 small bowel adenocarcinomas 393 – 4 small bowel carcinoid 395 small bowel lymphomas 396 small cell carcinomas of the gastrointestinal tract 432 stomach cancers 352 verrucous carcinoma 339 genitourinary malignancy metastatic disease 34 micropapillary bladder cancer 21 preoperative, prostate sarcomas 52 primary lymphoma of prostate 59 prostate sarcomas 52 small cell bladder carcinoma 20 small cell undifferentiated carcinoma of prostate 44 transitional cell carcinoma of prostate 56 urachal cancer 25 urethral cancer 33 – 4, 34 see also renal tumors; testicular/paratesticular tumors (below) gynecological malignancy borderline ovarian tumors 451 – 2 cervical lymphoma 514 dysgerminoma 474 – 5
SUBJECT INDEX embryonal rhabdomyosarcoma of cervix 512 endometrial carcinomas 495 extra-ovarian primary peritoneal carcinomas 441 – 2, 442 gestational trophoblastic disease 532, 536, 538 – 40, 539 ovarian germ cell tumors 471, 472 – 3 primary adenocarcinoma of fallopian tube 480 – 1 serous papillary adenocarcinoma of cervix 508 stromal tumors of the ovary 457 – 8, 458, 460 uterine sarcoma 490 – 1, 491, 492, 492 – 3 vaginal squamous cell carcinoma 527 vulvar squamous cell carcinoma 523, 524 head and neck cancer esthesioneuroblastoma 138 – 9 esthesioneuroblastoma, recurrence 139 – 40 nasopharyngeal carcinoma 126 paranasal sinuses tumors 728 pediatric extraosseous Ewing’s sarcoma 723 see also laryngeal tumors; oral cavity tumors (below) hematological malignancy mantle cell leukemia 550 nasopharyngeal carcinoma 126 – 8 primary plasma cell leukemia 549 laryngeal tumors adenoid cystic carcinoma 107 – 8 mucoepidermoid carcinoma 108 neuroendocrine tumors 110 spindle cell (sarcomatoid) carcinoma 105 limitations 700 neurological malignancy AIDS-related primary CNS lymphoma 664 anaplastic oligodendroglioma sensitivity 691 atypical teratoid/rhabdoid tumors 806 chordomas 622 choroid plexus carcinoma 671, 802 craniopharyngiomas 709 desmoplastic glioma (gliofibroma) 801 ependymoblastoma 805 ependymoma 691 glioblastoma 682 – 3, 684 Langerhans’ cell histiocytosis of CNS 612 – 3 low-grade oligodendroglioma 690 – 1 medulloblastoma 697, 698, 700 melanocytic lesions 807 meningeal sarcomas 634 – 5 meningioma 643, 643 neurocutaneous melanosis 607 pineoblastoma 804 pleomorphic xanthoastrocytoma 800 primary intracranial germ cell tumors 654 primitive neuroectodermal tumor 697, 702 spinal cord gliomas 689 trilateral retinoblastoma 804 – 5 see also ophthalmic tumors (below) ophthalmic tumors optic pathway glioma 685 retinoblastoma 718 sebaceous carcinoma of the eyelid 713 squamous cell carcinoma of the conjunctiva 714, 715 oral cavity tumors bone tumors 90 hematologic malignancies 96 – 7 salivary gland tumors 93 – 4, 332, 332
soft tissue sarcomas 99 – 100 pediatric malignancy ameloblastoma 725 clear cell renal sarcoma 761 colorectal carcinoma 752 – 3 cutaneous fibrosarcoma 815 extraosseous Ewing’s sarcoma 723 mediastinal germ cell neoplasms 734 – 5 mediastinal tumors 733 – 4 medulloblastoma 698, 700 paranasal sinuses tumors 728 renal cell carcinoma 760 see also specific malignancies renal tumors adult Wilms tumor 13 – 4 clear cell sarcoma in children 761 collecting duct (of Bellini) carcinomas (CDCs) 3 – 4 ossifying renal tumor of infancy 762 RCC in children 760 renal medullary carcinoma 4 sarcomas 11 testicular/paratesticular tumors adenocarcinoma of the rete testes 75 Leydig cell tumors 72 malignant mesothelioma of the tunica vaginalis 76 – 7 non-Hodgkin’s lymphoma 70 rhabdomyosarcoma 80 – 1 Sertoli cell tumors 73 thoracic tumors adenoid cystic carcinoma of the lung 325 bronchioloalveolar carcinoma 317 large cell neuroendocrine carcinoma of the lung 303 mediastinal germ cell neoplasms 734 – 5 mediastinal tumors, pediatric malignancy 733 – 4 mucoepidermoid lung cancer 332 – 3 peritoneal mesothelioma 287 pleural mesothelioma 284 – 5, 285, 286, 287 primary lung sarcomas 275 primary melanoma of lung 295 – 6 primary pulmonary lymphoma 261, 262 pulmonary carcinoid tumors 310 thymomas/thymic carcinomas 249 – 52, 251 toxicity/adverse effects 598 see also specific drugs/drug combinations (regimens) Chest pain, primary pulmonary lymphoma 259 Chest, pleural mesothelioma 282 Chest radiography (CXR) gastrointestinal stromal cell tumors 344 lung mucoepidermoid carcinoma 331 molar pregnancy follow-up 536 pleural mesothelioma 282 primary pulmonary lymphoma 259, 259, 260 pulmonary carcinoid tumors 308 – 9 uterine sarcoma 486 Chiasmal syndrome, meningioma associated 641 Childbearing ovarian cancer risk 447 see also Pregnancy Children see Pediatric malignancy Children’s Cancer Group (CCG), medulloblastoma 699 Children’s Solid Tumor Group Study, paratesticular rhabdomyosarcoma chemotherapy 80 Chloroma 512 – 3 Cholangiocarcinoma 385 hepatocellular carcinoma and 387 Cholelithiasis, somatostatinoma 375 Cholesterol, steroidogenesis and 151
827 Chondroid chordomas 615, 615 Chondroma, chondrosarcoma vs. 89, 108 Chondromata, pulmonary 361 Chondrosarcoma chondroma vs. 89, 108 esophageal 345 laryngeal 108, 108 – 9 dedifferentiated 109 meningeal 630, 630, 630, 630 – 631 prognosis 635 radiation therapy 634 oral cavity and adjacent structures 89, 89 rare causes 272 Chondrosarcoma with additional malignant mesenchymal component (CAMMC) 109 Chordoma 614 – 25 adults 614 chondroid pattern 615, 615 clinical features/presentation 616 – 8, 617 of clivus 614, 617, 617, 620 – 1 cytogenetics 616 differential diagnosis 616, 617 embryonic origins 614 epidemiology 614 – 6 histology 615, 615 – 6 historical background 614 imaging 618, 618 – 20, 619 intracranial cavity 617, 617 with malignant degeneration 615 – 6 molecular biology 616 pediatric 614 prognosis/survival 616, 618, 620, 621, 622 pseudoencapsulated 615 recurrence rates 620, 621 of sacrum 614, 617, 617, 621 of skull base 617, 617, 620 – 1 treatment 620 – 2 chemotherapy 622 radiation therapy 621 – 2 surgical 620 – 1 typical (classical) 615, 615 of vertebrae 617 – 8 Choriocarcinoma 469, 470, 514, 532, 537, 649 clinical presentation 537 epidemiology 533, 537, 650 esophageal 345 – 6 etiology 533 hCG secretion 534 mediastinal tumors 734 metastases 534, 537 pathology 534, 650 tumor markers 652 Chorionic sac, development 533 Choroidal metastases, pediatric 739 Choroidectomy complications 670 melanoma 716 papillomas/carcinomas 669 – 70, 801 – 2 Choroid plexus 715 carcinoma see Choroid plexus carcinoma (CPC) imaging 667 melanoma 715, 716 papilloma see Choroid plexus papilloma (CPP) Choroid plexus carcinoma (CPC) 667 – 73, 802 anatomy 667 biology 667 – 8 clinical presentation 669 hydrocephalus 667 epidemiology 667 – 8, 801 historical background 667 imaging 667, 669, 670, 801, 802 metastases 671 pathology 668, 668 – 9, 669 pediatric 667, 801 – 2 prognosis 671
828 Choroid plexus carcinoma (CPC) (cont.) treatment 667, 669 – 71 chemotherapy 671, 802 radiation therapy 670 – 1, 802 surgical 669 – 70, 801 – 2 WHO classification 668 Choroid plexus papilloma (CPP) 667 – 73 anatomy 667 biology 667 – 8 clinical presentation 669, 801 hydrocephalus 667, 668 epidemiology 667 – 8, 801 historical background 667 imaging 667, 669, 670 pathology 668, 668 – 9 pediatric 667, 801 postoperative management 671 prognosis 671 surgical treatment 667, 669 – 70 WHO classification 668 Chromaffin reaction 156 Chromogranin A (CgA) appendiceal carcinoid tumors 414 esthesioneuroblastoma 135 pheochromocytoma 158 Chromogranin(s), small cell undifferentiated carcinoma of prostate 41 Chromosomal abnormalities acute erythroblastic leukemia 546 acute megakaryoblastic leukemia 547 BCR-ABL fusion 405 see also Cytogenetics; specific chromosomes/abnormalities Chromosome 1 abnormalities, Peutz-Jeghers syndrome 785 Chromosome 2 abnormalities, Carney complex (CNC) 784 Chromosome 6 abnormalities Peutz-Jeghers syndrome 785 thymomas/thymic carcinomas and 238, 239 Chromosome 9 abnormalities, childhood adrenocortical tumors 775 Chromosome 10 abnormalities Cowden disease 786 meningioma 640 Chromosome 11 abnormalities adrenocortical tumors 149 – 50, 777 paragangliomas 157 Chromosome 14 abnormalities, meningioma 640 Chromosome 17 abnormalities adrenocortical tumors 150, 777 Carney complex (CNC) 784 – 5 dermatofibrosarcoma protuberans (DFSP) 589 Chromosome 19 abnormalities, Peutz-Jeghers syndrome 786 Chromosome 22 abnormalities dermatofibrosarcoma protuberans (DFSP) 589 meningeal sarcomas 628 meningioma 640 Chronic eosinophilic leukemia (CEL) 544 – 5 Chronic lymphocytic leukemia (CLL), prostatic involvement 57 Chronic neutrophilic leukemia (CNL) 543 – 4 Chrysotile (white asbestos), mesothelioma and 279, 280 Chung staging systems, vulvar squamous cell carcinoma 524, 524 Ciliary body 715 melanoma 715, 716 Cisplatin/anthracycline, esophageal adenoid cystic carcinoma 341 Cisplatin-based chemotherapy adrenocortical tumors 154, 154 bronchioloalveolar carcinoma 317 cervical adenoid cystic carcinoma 510 cervical small cell carcinoma 511
SUBJECT INDEX dysgerminoma 474 esthesioneuroblastoma 140 extra-ovarian primary peritoneal carcinomas 442 – 3 mediastinal germ cell neoplasms 734 – 5 medulloblastoma 700 Merkel cell carcinoma 598 mucoepidermoid tumors 332 nasopharyngeal carcinoma 126, 127 recurrent/metastatic 129 nephrotoxicity due to 473 pleural mesothelioma 284 – 5, 286, 287 pregnancy 517 – 8 primary adenocarcinoma of fallopian tube 481 squamous cell carcinoma, cervix/vagina 34 thymomas/thymic carcinomas 249, 250, 251 see also specific combinations Cisplatin, bleomycin, and methotrexate (MTX), urethral cancer 33 – 4 Cisplatin, carboplatin, procarbazine and diaziquone (AZQ), mucoepidermoid lung cancer 683 Cisplatin, doxorubicin, and cyclophosphamide (PAC), thymomas/thymic carcinomas 249, 250, 251, 251 Cisplatin/doxorubicin, pleural mesothelioma 285 Cisplatin, doxorubicin, vincristine, and cyclophosphamide (ADOC), thymomas/thymic carcinomas 248, 250, 251 Cisplatin, etoposide and bleomycin (BEP), ovarian germ cell tumors 472, 472 Cisplatin/etoposide, small cell undifferentiated carcinoma of prostate 45 Cisplatin/gemcitabine, collecting duct (of Bellini) carcinomas 3 – 4 c-KIT gene/protein dysgerminoma 468 gastric stromal tumors 360 treatment targets 361 gastrointestinal stromal cell tumors (GISTs) 344 pediatric 753 large cell neuroendocrine cancer of the lung 299 phyllodes tumor of the breast 211, 214 uterine sarcoma 486 Cladribine, mast cell leukemia 552 Clark and Breslow staging systems, vulvar squamous cell carcinoma 524, 524 Classification systems see Tumor classification/staging systems Clear cell adenocarcinoma 507 Clear cell borderline ovarian tumors 449, 452 Clear cell carcinoma cutaneous 585 endometrial 494 – 6 management 495 – 6 pathology 494 – 5, 495 gallbladder tumors 384 ovarian 452 thymus 244, 244 Clear cell chondrosarcoma, laryngeal 109 Clear cell eccrine hidradenocarcinoma 767 Clear cell sarcoma renal in children 760 – 1 small bowel 396 Clitoris, pediatric malignancy 766 – 7 Clival chordomas 614 clinical features 617, 617 imaging 618, 619 surgical resection 620 – 1 Clonorchis sinensis, bile duct tumors 385 Cognitive changes medulloblastoma 702 neuroepithelial neoplasms 677
Colectomy, appendiceal adenocarcinoma 412 Collaborative Ocular Melanoma Study (COMS) trials 715, 716, 717, 717 Collagenous spherulosis, adenoid cystic carcinoma of the breast vs. 189, 190 Collecting duct (of Bellini) carcinomas (CDCs) 2–4 background 2 – 3 clinical presentation 3 mucinous carcinoma vs. 5 pathology 3, 3 renal medullary carcinoma vs. 4 treatment/prognosis 3 – 4 Collin’s law, medulloblastoma 701 “Collision hypothesis” breast carcinosarcoma 218, 219 laryngeal spindle cell (sarcomatoid) carcinoma 105 Colloid adenocarcinoma cervical 506 – 7 pancreatic 371 Colon cancer 401 carcinoid tumors 401 clinical presentation 402 – 3 management 403 metastatic disease 403 – 4 pathology 401 – 2 prognosis 404 malignant melanoma 407 – 8 see also specific types Colonic small cell carcinomas 433 Colonic-type adenocarcinomas, appendix 412 Colorectal cancers 401 see also specific types/locations Colorectal carcinoma epidemiology 749, 751 metastases, small bowel 398 pediatric malignancy 749, 751 – 3 adjuvant therapy 752 – 3 adult tumors vs. 752 associated conditions 751 – 2 desmoid tumors 752 pathology 752, 752 staging 752, 752 treatment 752 – 3 Colorectal sarcoma 404 – 6 Combined modality therapy see Multimodality treatment Comparative genomic hybridization (CGH), esthesioneuroblastoma 135 – 6 Computed tomography (CT) endocrine tumors adrenocortical tumor diagnosis 152, 152, 779 – 80, 780 medullary thyroid carcinoma 166 pheochromocytomas 158, 790, 791 primary thyroid lymphoma 169 extranodal NK/T cell lymphoma, nasal type 561 gastrointestinal tumors bile duct tumors 385 colon/rectum carcinoid tumors 403 – 4 gallbladder tumors 383 gastric lymphomas 362 gastrointestinal stromal cell tumors 344 pancreatic cystadenocarcinoma 369 pancreatic tumors 368 small bowel adenocarcinoma 393, 393 small bowel carcinoid 394, 395 small cell carcinomas of the gastrointestinal tract 431 genitourinary cancer bladder small cell carcinoma 20 congenital mesoblastic nephroma 762 prostate sarcomas 49 renal angiomyolipoma 8 Sertoli cell tumors 73 small cell undifferentiated carcinoma of prostate 44 urethral cancer 30
SUBJECT INDEX gynecological malignancy borderline ovarian tumors 449 extra-ovarian primary peritoneal carcinomas 441 primary adenocarcinoma of fallopian tube 480 head and neck cancer ameloblastoma 724 cancers of oral cavity and adjacent structures 89 – 90, 96, 97 esthesioneuroblastoma 137, 729 extranodal NK/T cell lymphoma, nasal type 561 extraosseous Ewing’s sarcoma 722 laryngeal chondrosarcoma 109 nasopharyngeal carcinoma 119 – 20 paranasal sinuses tumors 728 magnetic resonance imaging vs. 137 Merkel cell carcinoma 597 neurological malignancy atypical teratoid/rhabdoid tumors 806 chordoma diagnosis 618, 618 – 20 choroid plexus papilloma/carcinoma 667, 669, 670, 801 craniopharyngiomas 707 – 8 desmoplastic cerebral astrocytoma of infancy 798 desmoplastic glioma (gliofibroma) 801 desmoplastic infantile ganglioglioma 798, 799 dysembryoplastic neuroepithelial tumors 803 ependymoblastoma 805 intracranial germ cell tumors 651 medulloblastoma 697 melanotic meningeal lesions 807 meningeal sarcomas 628 meningioma 641 – 2 neuroepithelial neoplasms 679 pineoblastoma 803 primary CNS lymphoma 660 primitive neuroectodermal tumor 697 pediatric malignancy ameloblastoma 724 atypical teratoid/rhabdoid tumors 806 congenital mesoblastic nephroma 762 desmoplastic cerebral astrocytoma of infancy 798 desmoplastic glioma (gliofibroma) 801 desmoplastic infantile ganglioglioma 798, 799 dysembryoplastic neuroepithelial tumors 803 ependymoblastoma 805 esthesioneuroblastoma 137, 729 extraosseous Ewing’s sarcoma 722 melanotic meningeal lesions 807 paranasal sinuses tumors 728 pineoblastoma 803 thoracic 732 thoracic tumors bronchioloalveolar carcinoma 316 large cell neuroendocrine carcinoma of the lung 302 lung sarcomas, primary 265 mediastinal tumors 733 – 4 pediatric 732 pleural mesothelioma 282 primary melanoma of lung 295 primary pulmonary lymphoma 259, 260 pulmonary carcinoid tumors 308 – 9 thymomas/thymic carcinomas 247 Condyloma acuminata, cervical verrucous carcinoma 502 – 3 Congenital adrenal hyperplasia (CAH), testicular tumors in children 766 Congenital dermatofibrosarcoma protuberans (DFSP) 815 Congenital hemihypertrophy, adrenocortical tumors 777
Congenital leukemia 814 – 5 Congenital melanoma 810 Congenital mesoblastic nephroma 761 – 2 Congenital nevi neurocutaneous melanosis 605, 606 pediatric basal cell carcinoma 812 pediatric melanoma 810 – 1, 811 Congenital rhabdomyosarcoma 815 Conjunctiva anatomy 713 cancers 713, 713 – 5 see also specific types Carney complex and 784 dysplasia 714 Conjunctival intraepithelial neoplasia (CIN) 714 Conjunctival squamous cell carcinoma (SCC) 713 – 5 biology/epidemiology 713 – 4 clinical presentation 714, 714 differential diagnosis/misdiagnosis 714 historical background 713 pathology 714 prognosis 714 – 5 treatment 714, 715 Conn’s syndrome 143, 147 – 8 biochemical diagnosis 152 clinical features 147 etiology 147 incidence 147, 147 pathology 147, 147 – 8, 148 prognosis/treatment 148 structural/functional heterogeneity of tumors 148 Contrast enhancement, neuroepithelial neoplasms 679 Conus elasticus 102, 103 Cornea, cancers 713, 713 – 5 Cortical thymomas 240, 240 Corticosteroids primary CNS lymphoma 662 thymomas/thymic carcinomas 250, 252 T-large granular lymphocyte leukemia 551 see also individual steroids Corticotrophin releasing hormone (CRH), adrenocortical tumor diagnosis 151 Cortisol adrenocortical tumor diagnosis 151 overproduction 143, 775 see also Cushing’s syndrome Cough adenoid cystic carcinoma of the lung 323 primary pulmonary lymphoma 259 COUP-TP transcription factor, adrenocortical tumors 151 Cowden syndrome 786 endocrine cancers in children 781 thyroid cancer 786 CPT-11 (irinotecan hydrochloride), colorectal carcinoma in children 753 Cranial/craniospinal radiation astrocytoma 684 pineoblastoma 804 primary intracranial germ cell tumors 653 – 4 Cranial nerve involvement chordomas 617 nasopharyngeal carcinoma 119 sinonasal undifferentiated carcinomas 728 Craniofacial resection esthesioneuroblastoma 140 paranasal sinuses tumors 728 Craniopharyngiomas 705 – 10 adamantinomatous 706 – 7, 707, 708 adults 705 age distribution 705, 706 anatomical considerations 705, 706 biology 705 – 6 blood supply 705 clinical features/presentation 707, 708
829 differential diagnosis 708 embryonic origins 705 endocrine evaluation 708 epidemiology 705 – 6 historical background 705 imaging 707 – 8, 708 medical evaluation 708 papillary 707, 708 pathology 706 – 7 pediatric malignancy 705, 799 prognosis 709 recommendations 709 treatment 708 – 9 chemotherapy 709 pretreatment management 708 radiation therapy 709 radiosurgery 709 surgical resection 708 – 9 Cricoid cartilage 102, 103 tumors 109 Cricothyroid membrane 103 Crocidolite (blue asbestos), mesothelioma and 279, 280 Crohn’s disease, small bowel adenocarcinoma 392 Cronkhite – Canada syndrome 751 Crow – Fukase syndrome 572 – 3 Cryoglobulinemia, lymphoplasmacytic lymphoma 557 Cryotherapy retinoblastoma 718 squamous cell carcinoma of the conjunctiva 714 Crypt cell carcinomas, appendix 413 CsA levels, T-large granular lymphocyte leukemia 551 c-sis/PDGF-2 gene, meningiomas 639 Curettage hydatidiform mole treatment 536 odontogenic tumors 90 Cushing’s syndrome adrenal hyperplasia 143, 144 adrenocortical tumors and 143 – 5 adenomas 144, 144 adults vs. children 144, 144 carcinomas 144 – 5, 145 children/adolescents 144, 144, 775, 778 incidence 143, 144 metastases 145 pathology 144, 144 – 5, 145 prognosis/treatment 145 atypical 784 biochemical diagnosis 151 Carney complex and 783, 784 familial 144 iatrogenic 144 oncocytic renal carcinoid tumors 762 periodic 784 presentation 143 – 4 symptomatic treatment 154 Cutaneous lymphoma 562 – 4 pediatric malignancy 813 – 4 Cutaneous malignancies 577 – 88 epidermis/epidermal appendages 577 – 9 pediatric malignancies see under Pediatric malignancy predisposing factors 812 soft tissue tumors 579 – 84 see also Skin lesions; specific malignancies/locations Cutaneous T cell lymphoma (CTCL) 562 ‘Cut-thru’ hysterectomy 515 – 6 CXC chemokines, adrenocortical tumors 149 Cyclectomy, melanoma 716 Cyclic AMP signaling, Carney complex and 785 Cyclin D1, parathyroid carcinoma 175 Cyclo-oxygenase 2 (COX2), breast carcinosarcoma 227 Cyclopamine, medulloblastoma/PNET 696
830 Cyclophosphamide appendiceal carcinoid tumors 415 extra-ovarian primary peritoneal carcinomas 442 – 3 medulloblastoma 700 Merkel cell carcinoma 598, 599 mucoepidermoid tumors 332 paratesticular rhabdomyosarcoma 80 pineoblastoma 804 pulmonary carcinoid tumors 310 see also specific combinations Cyclophosphamide, doxorubicin, and cisplatin (CAP), adenoid cystic carcinoma of the lung 70 Cyclophosphamide, doxorubicin, and vincristine (CAV) meningioma 643, 643 Merkel cell carcinoma (MCC) 599 small cell carcinomas of the gastrointestinal tract 432 Cyclophosphamide, doxorubicin, vincristine and dexamethasone (CHOD), primary CNS lymphoma 662 Cyclophosphamide, doxorubicin, vincristine, prednisone (CHOP) cervical lymphoma 514 non-Hodgkin’s lymphoma (NHL) breast 198 oral cavity and adjacent structures 96 testicular 70 primary CNS lymphoma 662 small bowel lymphomas 396 Cyclophosphamide, hydroxyurea, actinomycin D, methotrexate (with folic acid), vincristine, and doxorubicin (CHAMOCA), gestational trophoblastic neoplasia 539 Cyclophosphamide, methotrexate, and 5-FU (CMF) breast carcinosarcoma 226 male breast cancer treatment 205 Cylindroma see Adenoid cystic carcinoma Cylindromatous carcinoma see Adenoid cystic carcinoma CYP17 gene, male breast cancer 202 Cyst(s) adrenocortical carcinoma 144 ganglion cell neoplasms 686 meningeal sarcomas 628 oral cavity and adjacent structures bone tumors vs. 87 salivary glands 91, 93, 95 ovarian complete hydatidiform moles and 535 dermoid, ovarian germ cell tumors 467 – 8 pancreatic cystadenocarcinoma 368 pancreatic neoplasms 371 – 2 pediatric brain tumors desmoplastic astrocytoma/gangliogliomas 798, 799 ependymoblastoma 805 pleomorphic xanthoastrocytoma (PXA) 799, 800 thymoma/thymic carcinoma 240, 242 see also entries beginning cystic Cystadenocarcinoma hepatic 388 pancreatic 368 – 9 Cystadenomas, hepatic 388 Cystectomy bladder small cell carcinoma 20 micropapillary bladder cancer 21 urachal cancer 25 Cystic adenomatoid malformation (CAM), lung parenchyma 739 Cystic carcinomas pancreatic 371 salivary glands 95 Cystic mesothelioma 288
SUBJECT INDEX Cystic neoplasms ovarian 467 pancreatic 371 – 2 pediatric malignancy 754 – 5, 755, 755, 755 – 756 salivary glands 95 see also specific types Cystic teratomas, ovarian 467 Cystoprostatectomy, prostate sarcomas 53 Cystosarcoma phyllodes see Phyllodes tumor Cytarabine ocular lymphoma 664 primary CNS lymphoma 662 Cytochrome p450, nasopharyngeal carcinoma 115 Cytogenetics alveolar soft part sarcoma 768 breast cancer carcinosarcoma 226 non-Hodgkin’s lymphoma (NHL) 196 phyllodes tumor of the breast 211 cutaneous clear cell carcinoma 585 dermatofibrosarcoma protuberans (DFSP) 579, 589 hydatidiform moles 533, 534 Merkel cell carcinoma 596 mesothelioma 281 – 2 nasopharyngeal carcinoma 116 neurological malignancy chordomas 616 meningioma 640 – 1 neuroepithelial neoplasms 676 peripheral primitive neuroectodermal tumor 585 primary CNS lymphoma 657 primary intracranial germ cell tumors 649 – 50 pheochromocytoma 157 renal cancers 1 mucinous carcinomas 5 papillary adenoma 2 renal cell carcinoma in children 760 small bowel lymphomas 396 stromal tumors of the ovary, granulosa cell tumors 459 synovial sarcoma 269 thymomas/thymic carcinomas and 238, 239 see also specific chromosomes Cytokeratins breast carcinosarcoma 219, 222 gastrointestinal stromal cell tumors 419 – 20 renal cancer markers 1 – 2 Cytology adenoma malignum (minimal deviation adenocarcinoma) 506 fine needle aspiration see Fine needle aspiration (FNA) biopsy/cytology large cell neuroendocrine carcinoma (LNEC) of the lung 301 leiomyosarcoma 266 male breast cancer diagnosis 203 pancreatic tumors 368 phyllodes tumor of the breast 213 primary lung sarcomas 274 transitional cell carcinoma of prostate 55 Cytoreduction anaplastic thyroid carcinoma 168 extra-ovarian primary peritoneal carcinomas 441, 442, 444 glioblastoma 681 intraperitoneal chemotherapy, pseudomyxoma peritonei 411, 412 low-grade oligodendroglioma 690 medulloblastoma in children 699 ovarian germ cell tumors 471 pancreatic carcinoid 377 pseudomyxoma peritonei 411 stromal ovarian tumors 457, 458, 463
uterine sarcoma 490 see also individual tumors; specific techniques Cytotoxic chemotherapy see Chemotherapy Cytotrophoblast 533 choriocarcinoma 469 Dacarbazine (DTIC) appendiceal carcinoid tumors 415 gastric endocrine cell proliferations 359 see also specific combinations Dacarbazine/adriamycin, recurrent meningioma 644 Dactinomycin see Actinomycin D (dactinomycin) Dandy – Walker malformation, neurocutaneous melanosis and 605, 606, 807 Daunorubicin, mucoepidermoid tumors 332 DAX-1 transcription factor, adrenocortical tumors 151 Debulking operations see Cytoreduction Dedifferentiated tumors, laryngeal chondrosarcomas 109 Dehydroepiandrostenedione sulfate (DHEAS) 462 Dehydroisoandrosterone (DHA), adrenogenital syndrome 146 Deleted in pancreatic cancer (DPC) loci 367, 373, 376 Dental extraction, nasopharyngeal carcinoma 128 Dentists/dental examination bone tumors 87, 88, 89 nasopharyngeal carcinoma 128 Dermatitis, glucagonoma 375 Dermatofibroma (DF), dermatofibrosarcoma protuberans vs. 589 Dermatofibrosarcoma protuberans (DFSP) 579 – 80, 589 – 93 Bednar tumors 580, 589, 815, 816 biology 589 clinical presentation/diagnosis 590, 590, 815, 816 differential diagnosis 589 giant cell fibroblastoma and 580 historical background 589 malignant fibrous histiocytoma and 590 myxoid 580 oral cavity and adjacent structures 97 pathology 579 – 80, 589 – 90, 815 – 6 “honeycomb appearance,” 589, 590 pediatric malignancy 579, 590, 815 – 6, 816 sarcomatous transformation 589 – 90 treatment/prognosis 580, 590 – 2, 816 adjuvant radiation therapy 580, 591 Mohs technique 580, 590 – 1, 816 recommendations 592 systemic therapy 580, 591 – 2 wide excision 590, 591 Dermoid cysts, ovarian germ cell tumors (OGCT) 467 – 8 Desmoplastic cerebral astrocytoma of infancy (DCAI) 798 – 9, 801 Desmoplastic glioma (gliofibroma), pediatric 800 – 1, 801 Desmoplastic infantile ganglioglioma (DIG) 798 – 9, 799, 801 Desmoplastic medulloblastoma 699 Desmoplastic melanoma, gingiva 91 Desmoplastic neuroepithelial tumors 798 – 801, 799 features 801 see also specific types Desmoplastic small round cell tumor (DSRCT), pediatric pleural tumors 743, 744 Dexamethasone, primary CNS lymphoma 662 Dexamethasone suppression test, adrenocortical tumor diagnosis 151
SUBJECT INDEX Diabetes insipidus, Langerhans’ cell histiocytosis of CNS 611, 613 Diabetes mellitus pancreatic cancer and 367 somatostatinoma 375 uterine sarcoma etiology 485 Diarrhea, carcinoid syndrome 395, 750 o, p -1,1-Dichloro-diphenyl-dichloroethane (DDD; Mitotane) 153 – 4, 154, 780 Dietary factors, stomach cancers 352 Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia (DIPNECH) 308 Digital papillary adenocarcinoma, aggressive 578 Digital rectal examination, prostate sarcomas 49 DiGuglielmo disease 545 Diplopia chordomas 617 sinonasal undifferentiated carcinomas 728 Disseminated peritoneal leiomyomatosis 494 DNA cervical adenoid basal carcinoma 509 content in thymomas/thymic carcinoma 244 – 5 repair, mucinous carcinomas 372 Dopamine receptors, meningiomas 639 Dopamine secretion, pheochromocytomas 790 Down’s syndrome, acute megakaryoblastic leukemia 547 “Downstaging,” gallbladder tumors 384 Doxorubicin malignant mesothelioma of the tunica vaginalis (MMTV) 77 mucoepidermoid tumors 332 pleural mesothelioma 284 – 5 thymomas/thymic carcinomas 250, 251 see also specific combinations Doxorubicin and ifosfamide (AI), breast carcinosarcoma 225, 226 DPC (deleted in pancreatic cancer) loci 367, 373, 376 Ductal carcinoma in situ (DCIS) 219, 222, 230, 232 male breast cancer 203 treatment 205 Dukes’ colonic cancer staging, fallopian tube cancer 477 Dulguerov staging, esthesioneuroblastoma 134 Duodenum 391 adenocarcinoma 393 Dura mater, meningeal sarcomas 630, 631 Dysembryoplastic neuroepithelial tumors (DNT) 687, 687, 802 – 3, 803 Dysgerminoma 474 – 5 chemotherapy 474 – 5 ovarian 468, 468 radiation 474, 474 stage l tumor observation 474, 474 Dysphagia, gastrointestinal stromal cell tumors (GISTs) 344 Dysplasia, basaloid squamous carcinoma 340 Dyspnea adenoid cystic carcinoma of the lung 323 primary pulmonary lymphoma 259 Eastern Cooperative Oncology Group (ECOG) anal carcinoma 407 carcinoid treatment 377 mucoepidermoid carcinomas 333 pulmonary carcinoid tumors 310 renal carcinoid treatment 10 small bowel adenocarcinoma treatment 394 streptozotocin and 5FU in pancreatic tumors 377
systemic therapy in bronchioloalveolar carcinoma 317 thymomas/thymic carcinoma chemotherapy 250 EBV-specific cytotoxic T lymphocytes, nasopharyngeal carcinoma 129 Ecchymosis, periorbital in neuroblastoma 816 Eccrine epithelioma (syringoid eccrine carcinoma) 579 Eccrine gland tumors 578 mixed apocrine 578 – 9 Ectopic hormonal secretion small cell carcinomas of the gastrointestinal tract 430 small cell undifferentiated carcinoma of prostate 42 EGFR see Epidermal growth factor receptor (EGFR) Electron microscopy acinar cell carcinoma 370 fibrosarcoma, lung 267 Langerhans’ cell histiocytosis of CNS 610 large cell neuroendocrine cancer of the lung 300 – 1 leiomyosarcoma, lung 266, 266 malignant fibrous histiocytoma, lung 269 meningeal chondrosarcomas 631 Merkel cell carcinoma 584, 596 mesothelioma 281 micropapillary bladder cancer 21 pancreatic oncocytic carcinoma 373 primary lung sarcomas 274 synovial sarcoma 269 EMA reactivity, thymic carcinoma 244 Emboli/embolization colon/rectum carcinoid tumors 404 liver metastases 415 lung sarcomas 270 pancreatic carcinoid, therapy 377 pulmonary 274 Embryology/embryogenesis adrenal gland 775 – 6 ameloblastoma 723 chordoma origins 614 craniopharyngioma origins 705 germ cell tumor pathogenesis 649 – 50 meningeal sarcoma 626 meninges 626 partial hydatidiform moles 534 testicular/paratesticular tumors 66, 67 Embryonal carcinoma 469 mediastinal tumors 734 primary intracranial epidemiology 650 tumor markers 652 Embryonal rhabdomyosarcoma 512 cutaneous 815 oral cavity and adjacent structures 98 – 9, 99 Endobronchial laser treatment, pulmonary carcinoid tumors 309 Endocarditis, acinar cell carcinoma 370 Endocervical-like mucinous borderline ovarian tumors 449 Endocervical type adenocarcinoma 504 Endocrine cell neoplasia, gastric cancer 357 Endocrine cell proliferations, stomach 357 – 60 clinical presentation 357 – 9 pathology 357 prognosis 360 treatment 359 Endocrine disturbance adrenal gland disorders 143 adrenocortical tumors and 150 – 1 appendiceal carcinoid tumors 415 craniopharyngiomas 707, 708 ectopic hormonal secretion
831 small cell carcinomas of the gastrointestinal tract 430 small cell undifferentiated carcinoma of prostate 42 esthesioneuroblastoma 133 islet cell tumors 374, 374 – 6, 756, 756 male breast cancer 201 – 2 primary intracranial germ cell tumors 651 small cell carcinoma of the pancreas 372 stromal tumors of the ovary 456, 461, 463 see also specific hormones Endocrine evaluation craniopharyngiomas 708 Leydig cell tumors 72 Sertoli cell tumors 73 Endocrine – paracrine cells, small cell undifferentiated carcinoma of prostate 38 Endocrine tumors adrenal see Adrenal glands children/adolescents 775 – 97, 782 gastric 357 – 60 treatment 359 pancreatic 373 – 8, 754 see also specific types of the skin see Merkel cell carcinoma (MCC) see also Endocrine disturbance; specific types/locations Endodermal sinus tumor 468 – 9, 469, 528 gastric 356 intracranial 649 epidemiology 650 mediastinal 734 primary intracranial pathology 650 tumor markers 652 vaginal, pediatric malignancy 768 Endolymphatic stromal myosis (ESM) 488 Endometrial cancers 494 – 6 see also specific types Endometrial carcinoma 494 – 6 clear cell carcinoma 494 – 6, 495 management 495 – 6 pathology 494, 494 – 5, 495 small cell undifferentiated 495 squamous cell 495 – 6 staging 495 type I vs. type II 494 uterine papillary serous carcinoma 494, 494 – 6 Endometrial stromal sarcoma (ESS) hormonal therapy 492 pathology 487, 487 – 8, 488 Endometrial-type ovarian stromal sarcoma, children 770 Endometrioid adenocarcinoma 505 Endometrioid borderline ovarian tumors 449 Endometrioid ovarian carcinoma 452 Endometriosis, ovarian cancer and 449, 452 Endometrium, stromal tumors of the ovary and 456, 459, 461, 770 Endosalpingiosis 439 – 40 Endoscopic biopsy esophageal 346 gastrointestinal stromal cell tumors (GISTs) 344 – 5 Endoscopic mucosal resection (EMR), gastric endocrine cell proliferations 359 Endoscopic retrograde cholangiopancreatography (ECRP) bile duct tumors 385, 385 pancreatic tumors 368 Endoscopic ultrasound (EUS) gastrointestinal stromal cell tumors (GISTs) 344 pancreatic tumors, pancreatic cystadenocarcinoma 369 Zollinger – Ellison syndrome 374
832 Endoscopy gastric carcinoid tumors, type 1, 358 gastric lymphomas 362 small bowel cancer 392 squamous cell carcinoma 170 see also specific techniques Endothelin receptors, meningiomas 639 Enteroclysis, small bowel cancer 391, 393 Enteropathy-type T cell lymphoma (ETTL) 561 – 2 Environmental factors adrenocortical tumors 777 medulloblastoma/primitive neuroectodermal tumor (PNET) 696 mesothelioma 279 – 81 nasopharyngeal carcinoma 114, 115 neuroepithelial neoplasms 675 stomach cancers 352 uveal melanoma 716 Eosinophils chronic eosinophilic leukemia 545 hypereosinophilic syndrome 545 Ependymoblastoma 805 Ependymoma 688, 691, 691 pediatric malignancy 678, 689, 799 spinal cord 688 Epidermal growth factor receptor (EGFR) adrenocortical tumors 149 bronchioloalveolar carcinoma 314 meningiomas 639 neuroepithelial neoplasms 676 targeted therapy bronchioloalveolar carcinoma 318 nasopharyngeal carcinoma 129 thymomas/thymic carcinoma 252, 646 Epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKI), bronchioloalveolar carcinoma 318 Epidermis/epidermal appendages, tumors of 577 – 9 Epidermoid cyst, testicular 766 Epidermolysis bullosa (EB), pediatric squamous cell carcinoma 812, 813 Epiglottis 102, 103 Epinephrine, pheochromocytoma 156, 157, 158, 789 Epirubicin nasopharyngeal carcinoma 127 recurrent/metastatic 129 prostate sarcomas 52 see also specific combinations Epirubicin, cisplatin and 5-FU (ECF), small bowel adenocarcinoma 393 – 4 Epithelial cells, thymus 237, 239 – 40 Epithelial membrane antigen (EMA), small cell undifferentiated carcinoma of prostate 41 Epithelial ovarian carcinoma (EOC) 437 Epithelial tumors cervical 508 – 11 choroid plexus 667 – 73 classification 2 desmoplastic neuroepithelial tumors 798 – 801 gastrointestinal 749 laryngeal 103 – 6 metaplastic carcinomas 181 phyllodes tumors 210 thymus 237 – 56 see also Mixed epithelial – mesenchymal tumors; specific types Epithelioid angiomyolipoma 8 Epithelioid angiosarcoma 98, 581 mediastinal tumors 735 Epithelioid hemangioendothelioma (EHE) cutaneous 581 hepatic 387 – 8 oral cavity and adjacent structures 98, 100 pulmonary 271, 271 pediatric 743
SUBJECT INDEX Epithelioid hemangioma 98 Epithelioid leiomyosarcoma, cervical 512 Epithelioid sarcoma cutaneous 580 vulvar 525 – 6 Epithelioid type morphology, gastrointestinal stromal cell tumors 419 Epithelioma, eccrine (syringoid eccrine carcinoma) 579 Epstein – Barr nuclear antigen (EBNA1) 115 Epstein – Barr Virus (EBV) bladder lymphoepithelioma 23 bone tumors of oral cavity and adjacent structures 89 Burkitt lymphoma 559 cutaneous leiomyosarcoma in children 816 lymphoepithelioma-like carcinoma 503 nasopharyngeal cancers basaloid squamous cell carcinoma 106 carcinoma 115, 129, 727 EBV-specific cytotoxic T lymphocytes 129 undifferentiated carcinomas 727 natural killer granular lymphocytes leukemia 551 primary CNS lymphoma 658 primary effusion lymphoma 559 small bowel lymphoma 396 thymoma/thymic carcinoma etiology 239, 243, 244 undifferentiated gastric carcinoma with lymphoid stroma 353 Epulis-osteoid type giant cell carcinoma 371 Erionite, mesothelioma and 280, 280 Erlotinib bronchioloalveolar carcinoma 314, 317 neuroepithelial neoplasms 677 Erythroblastic leukemia 545 – 6 Esophageal adenocarcinoma 337, 340 Esophageal cancer 337 – 51 adenocarcinomas 337, 340 carcinomas 337 – 43 mixed differentiation 341 – 2 squamous cell variants 337 – 40 mesenchymal tumors 343 – 5 miscellaneous tumor types 345 – 7 neuroendocrine tumors 342 – 3 risk factors 337 staging 338 see also specific types Esophageal pseudosarcoma 339 – 40 Esophageal sarcoma, primary 345 Esophageal small cell carcinomas 432 Esophagectomy adenosquamous carcinoma of esophagus 342 esophageal small cell carcinomas 432 Esophagogastric-duodenoscopy (EGD) 392 Esthesioneuroblastoma (ENB) 133 – 42, 729 – 30 imaging studies 136, 136 – 8 general considerations 136 – 7 local disease 137 molecular imaging 138 recurrence/surveillance 138 regional/distant disease 137 – 8 see also individual techniques pathology 134 – 5 differential diagnosis 135 electron microscopy 135, 135 immunohistochemistry 135 microscopic grading 134 – 5 morphologic characteristics 134 staging 133 – 4, 134 survival 134 treatment 138 – 40 chemotherapy for recurrence 139 – 40 early stage disease 138 historical perspective 138
locally advanced resectable disease 138 – 9 nodal disease 139 patterns of recurrence 139 radiation therapy considerations 140 surgical considerations 140 unresectable/marginally resectable disease 139 Estrogen receptor adenoid cystic carcinoma of the breast 188, 191 endometrial carcinomas 494 endometrioid adenocarcinoma 505 extra-ovarian primary peritoneal carcinomas 441 male breast cancer 203 meningioma 644 meningiomas 639 tubular carcinoma 232 Estrogens (circulating) male breast cancer 201 – 2 uterine sarcoma etiology 485 Ethnicity/race lung cancer and 321 nasopharyngeal carcinoma 114, 114 – 5 treatment and 121 nasopharyngeal carcinoma treatment 121 thymoma etiology and 239 trophoblastic disease and 532 urethral cancer 27 uterine sarcoma 485 uveal melanoma 715 Etoposide-based chemotherapy cervical small cell carcinoma 511 pulmonary carcinoid tumors 310 see also specific combinations Etoposide/cisplatin (EP) Merkel cell carcinoma (MCC) 599 thymomas/thymic carcinomas 250 Etoposide, ifosfamide and cisplatin (VIP), thymomas/thymic carcinomas 250 Etoposide, methotrexate (with folic acid), actinomycin D, and cisplatin (EMA-EP), gestational trophoblastic neoplasia 540 Etoposide, methotrexate (with folic acid), actinomycin D, cyclophosphamide and vincristine (EMA-CO), gestational trophoblastic neoplasia 539, 539, 540 Etoposide, methotrexate (with folic acid), and actinomycin D (EMA), gestational trophoblastic neoplasia 539 Etoposide/thiotepa, choroid plexus carcinoma 802 European Organisation for Research and Treatment of Cancer (EORTC) 684 anal carcinoma 407 cutaneous lymphoma 562 pleural mesothelioma prognosis 287 systemic therapy 317 thymomas/thymic carcinoma chemotherapy 250 Europe, nasopharyngeal carcinoma incidence 114 Ewing’s sarcoma cutaneous involvement 817 extraosseous (EW-PNET) 721 – 3 biology 722 cardiac 737 clinical presentation 722 differential diagnosis 722 epidemiology 721 evaluation 722 pathology 721 – 2 patterns of spread 722 pleural tumors 744 prognosis 723 treatment 722 – 3 see also Primitive neuroectodermal tumor (PNET)
SUBJECT INDEX management 90 oral cavity and adjacent structures 88, 89, 90 PNET vs. 89 EWS gene 722 Exocrine pancreatic tumors 368 – 73, 754 see also specific types External beam radiation female urethral cancer 33 malignant hemangioendothelioma 170 Extra-appendiceal mucin-secreting epithelial cells, pseudomyxoma peritonei 411 Extrafascial hysterectomy 516 Extragonadal germ cell tumors 170 Extramammary Paget’s disease 579 Extramedullary plasmacytoma oral cavity and adjacent structures 95 – 6, 96, 97 urethra 30 Extranodal marginal zone B cell MALT lymphoma 558 Extranodal NK/T cell lymphoma, nasal type 561 Extraosseous Ewing’s sarcoma see under Ewing’s sarcoma Extra-ovarian primary peritoneal carcinomas (EOPPC) 437 – 46 biology/epidemiology 437 – 9 clinical presentation 440 – 1 diagnostic presentation 440 – 1 differential diagnosis 440 historical background 437 pathology 439 – 40 prognosis 443 – 4 treatment 441, 441 – 3 drug resistance 443 Extraparenchymal lesions, Langerhans’ cell histiocytosis of CNS 611 – 2, 612 Extrapleural pneumonectomy, pleural mesothelioma 283 – 4, 284, 286 Extrauterine ESS 512 Extreme drug resistance (EDR) assays 443 Eye anatomy 712, 713, 715, 717 tumors see Ophthalmic cancers; specific types Eyelid anatomy 712 cancers 712 – 3, 713 Facial pain, nasopharyngeal carcinoma 118 Fallopian tube tumors 477 – 84 anatomy 477 metastatic tumors 482 pediatric malignancy 771 primary adenocarcinoma 477 – 81, 479 biology/epidemiology 477 – 9 clinical presentation 479 – 80 diagnostic considerations 479 – 80 histology 479 macroscopic appearance 478 pathology 478 prognosis 481 spread pattern 478 – 9 treatment 480 – 1 sarcoma 481 – 2 staging 477 trophoblastic tumors 482 Familial adenomatous polyposis colorectal cancer in children 751 – 2 hepatoblastoma 754 small bowel adenocarcinomas 392 Familial atypical (dysplastic) mole syndrome, pediatric melanoma 810 Familial cancer syndromes choroid plexus papilloma/carcinoma 668 endocrine tumors 781 – 91 adrenocortical 149, 776, 776 – 7 pheochromocytoma 155, 787 – 91
malignant cylindroma 578 phyllodes tumor of the breast 211 see also specific syndromes Familial cylindromatosis (Brooke – Spiegler syndrome), malignant cylindroma 578 Familial medullary thyroid cancer (FMTC) 165 Familial melanoma syndrome, pancreatic cancer and 367 Familial nonmedullary thyroid carcinoma 786 Familial pancreatic cancers 367, 374 Familial pheochromocytoma 155, 787 – 91 Familial polyposis coli, Carney complex and 784 Familial segregation, nasopharyngeal carcinoma 115 – 6 Family history, neuroepithelial neoplasms 675 Farnesylthiosalicyclic acid, Merkel cell carcinoma 599 Fat cell tumors see Liposarcoma Fatigue, GIST in children 753 Fat necrosis, acinar cell carcinoma syndrome 370 FDG-PET see Fluorodeoxyglucose-positron emission tomography (FDG-PET) Federation Internationale de Gynecologie et d’Obstetrique (FIGO) cervical melanoma staging 513 extra-ovarian primary peritoneal carcinomas 441 fallopian tube cancer 477, 478 gestational trophoblastic neoplasia classification 537 – 8, 538 ovarian cancer classification 447, 450, 769, 769 ovarian germ cell tumor staging 471 pregnancy and staging 517 uterine sarcoma classification 489 vulvar squamous cell carcinoma 522 Female(s) adrenogenital syndrome 145 gastric carcinoid tumors, type 1, 358 reproductive system tumors cervix see Cervical tumors fallopian tube see Fallopian tube tumors ovarian see Ovarian cancer pediatric malignancy 766 – 71 uterus see Uterine tumors see also specific tumors/locations urethra anatomy 27 surgery 31 Feminization, adrenocortical tumors 143, 145, 146, 146 children/adolescents 778 Fertility, chemotherapy and 473 Fertility-compromising surgery, embryonal rhabdomyosarcoma 512 Fertility preservation borderline ovarian tumor treatment 450, 452 cervical tumors 514 embryonal rhabdomyosarcoma 512 hydatidiform mole treatment 536 stromal tumors of the ovary and 456, 457, 459, 460, 461, 462, 464 Fetal adrenal cortex 776 Fetal alcohol syndrome, adrenocortical tumors 777 Fetal pulmonary adenocarcinoma 739 – 40, 740 Fetus nitrosourea exposure, neuroepithelial neoplasms 675 – 6 partial hydatidiform moles 534, 535 Fiberglass, mesothelioma and 280 Fibroadenoma breast carcinosarcoma 219
833 phyllodes tumor vs. 209, 212 Fibroblast growth factors (FGFs) basic (bFGF), meningiomas 639 neuroepithelial neoplasms 676 – 7 Fibrohistiocytic tumors 580 – 1 Fibrolamellar carcinoma 387 Fibroma cardiac 736, 737 ovarian 461 Fibromatosis gastrointestinal stromal cell tumors 420 – 1 phyllodes tumor of the breast 212 Fibronectin, thymomas/thymic carcinomas and 238 Fibrosarcoma cutaneous, pediatric malignancy 815 esophageal 345 infantile, kidney 762 meningeal 629, 629 – 30 classification 629 prognosis 635 radiation-induced 627 – 8 radiation therapy 634 radiography 628, 629 oral cavity and adjacent structures 98 management 100 prognosis 97 ovarian 461 pediatric malignancy 770 pulmonary 264, 266 – 7, 267 pediatric malignancy 742, 742 Fibrothecoma, ovarian 461 Fibrous histiocytoma, pediatric bronchus 737, 738 see also Malignant fibrous histiocytoma (MFH) Fibrous skin lesions 579 – 80 Fibrous xanthosarcoma see Malignant fibrous histiocytoma (MFH) Fibroxanthoma, atypical 581 FIGO see Federation Internationale de Gynecologie et d’Obstetrique (FIGO) Fine needle aspiration (FNA) biopsy/cytology ameloblastoma 724 breast carcinosarcoma 222 male 203 metaplastic carcinoma 183 phyllodes tumor of the breast 213 lung primary lung sarcomas 274 primary pulmonary lymphoma 260 – 1 risks 247 thymomas/thymic carcinomas 247 thyroid medullary thyroid carcinoma 166 pediatric cancers 787 secondary malignancies 171 FLI1 gene, extraosseous Ewing’s sarcoma 722 Florid mesothelial hyperplasia 440 Flow cytometry, thymomas/thymic carcinoma 244 – 5 Fludarabine, Waldenstr¨om macroglobulinemia 571 Fluorodeoxyglucose-positron emission tomography (FDG-PET) adrenocortical tumors 152 – 3, 153, 779, 780 bronchioloalveolar carcinoma 316 esthesioneuroblastoma 137 – 8 gastric lymphomas 362 imatinib mesylate 423 meningioma 642 Merkel cell carcinoma 597 non-Hodgkin’s lymphoma, breast 197 pheochromocytomas 158, 158 – 9 thymomas/thymic carcinomas 247 5-Fluorouracil (5FU) appendiceal adenocarcinoma 413
834 5-Fluorouracil (5FU) (cont.) appendiceal carcinoid tumors 415 CAP and, adenoid cystic carcinoma of the lung 70 colon/rectum carcinoid tumors 404 colorectal carcinoma in children 752 – 3 mucoepidermoid tumors 332 nasopharyngeal carcinoma 126, 127 recurrent/metastatic 129 pseudomyxoma peritonei 411 pulmonary carcinoid tumors 310 small bowel adenocarcinoma 393 – 4 squamous cell carcinoma cervix/vagina 34 streptozotocin and in islet cell tumors 377 topical, squamous cell carcinoma of the conjunctiva 714 urachal cancer 25 urethral cancer 34 see also specific combinations 5-Fluorouracil (5FU), doxorubicin, and cyclophosphamide (FAC) breast carcinosarcoma 226 cisplatin and, mucoepidermoid tumors 332 Flushing, carcinoid syndrome 750 Follicular carcinomas, differentiated thyroid carcinoma 786 Follicular dendritic cells (FDCs) 563 – 4 Follow-up gestational trophoblastic neoplasia 540 hydatidiform mole 536 Fontana – Masson method 39 Foster Kennedy’s syndrome, meningioma association 641 Frantz tumor, pediatric malignancy 754 – 5, 755 Freedom from local recurrence (FLR), nasopharyngeal carcinoma treatment 121 French-American-British (FAB) classification, acute erythroblastic leukemia 545 French Sarcoma Group, imatinib mesylate 423 Fulguration, urethral cancer 31 Gallbladder carcinoma 383 Gallbladder tumors 383 – 5 adjuvant therapy 383 – 4, 384 advanced disease 384 epidemiology/risk factors 383 pathology 383 presentation/diagnosis 383 staging 384 surgical resection 383 uncommon histologies 384 – 5 see also specific types Gallstones, gallbladder tumors and 383 Gamma knife radiation therapy, meningioma 643, 644 Gangliocytomas 686 Gangliogliomas, desmoplastic 798 – 9 Ganglion cell neoplasms 686 – 7, 687 Ganglioneuroblastoma, mediastinal tumors 735, 735, 736 Ganglioneuroma, mediastinal 735 Gardner syndrome colorectal carcinoma 751 thyroid cancer 786 Gastrectomy, gastric stromal tumors 361 Gastric carcinoma with lymphoid stroma (GCLS) 353, 353 small bowel metastases 398 Gastric choriocarcinoma 355 – 6 Gastric lymphoma 361 – 3 clinical presentation 362 MALT 558 pathology 361 – 2 prognosis 363 staging 362, 362 treatment 362 – 3
SUBJECT INDEX Gastric small cell carcinomas 432 – 3 Gastrinomas MEN1 and 782 Zollinger – Ellison syndrome 374 Gastrin secretion, Zollinger – Ellison syndrome 374 – 5 Gastrointestinal autonomic nerve tumors 753 – 4 Gastrointestinal bleeding/hemorrhage gastric choriocarcinoma 355 – 6 gastric lymphomas 362 Gastrointestinal mucinous borderline ovarian tumors 448 – 9, 449 Gastrointestinal small cell carcinoma 430 – 5 clinical manifestations 430 epidemiology 430 histological characteristics 431 molecular characteristics 431 – 2 specific sites, characteristics/treatment 432 – 3 staging/workup 430 – 1 treatment 432 see also specific types/locations Gastrointestinal stromal tumor (GIST) 343 – 5, 344, 418 – 28 anatomy 418 clinical presentation/diagnosis 344, 421 – 2 colorectal in children 753 – 4 differential diagnosis 420 – 1 epidemiology 418 etiology/pathology 344 – 5 historical background 418 intestinal 405 – 6 mast cell leukemia 551 natural history/treatment 345 pathology 418 – 21 gross features 418 – 9 malignancy criteria 421 microscopic differential diagnosis 420 – 1 microscopic features 419, 420 molecular pathology 421 prognosis 426 – 7 small bowel 396 – 7, 397 treatment 422 – 6 chemotherapy 422 combination therapies 426 imatinib mesylate see Imatinib mesylate local therapy 422 radiation therapy 422 sunitinib malate (SU11248, Sutent) 397, 425 surgery 422 tyrosine kinase inhibitors 426 Gastrointestinal toxicity, imatinib mesylate 423 Gastrointestinal tract tumors carcinoid see Carcinoid tumors epidemiology 749 hepatobiliary 383 – 90 pediatric malignancy 749 – 59 smooth muscle 753 stromal mesenchymal 753 see also specific types/locations Gefitinib bronchioloalveolar carcinoma 314, 317 large cell neuroendocrine carcinoma of the lung 303 mucoepidermoid carcinomas 333 pleural mesothelioma 285 thymomas/thymic carcinoma treatment 252 Gemcitabine mucoepidermoid carcinomas 333 pleural mesothelioma 285 see also specific combinations Gemcitabine/cisplatin (GC), urethral cancer 34 Gemistocytic astrocytoma 685, 685 Genetic abnormalities see Molecular biology
Genetic instability Carney complex 784 pheochromocytoma 157 Genetics see Molecular biology Genetic testing, multiple endocrine neoplasia (MEN) 166 Genitourinary small cell carcinoma 46 Genitourinary tumors benign mesothelioma 288 male reproductive system 764 – 6 pediatric malignancy see under Pediatric malignancy Peutz – Jeghers syndrome and 785 see also specific types/locations Genomics, esthesioneuroblastoma 135 – 6 Geographic variation adrenocortical tumors in children 775, 776 male breast cancer 201 trophoblastic disease 532 Germ cell theory 649 Germ cell tumors classification 2, 649 female 768 gastric 355 – 6 mediastinal 734 nongerminomatous see Nongerminomatous germ cell tumors primary 514 primary intracranial see Primary intracranial germ cell tumors sites of origin 649 see also specific types/locations Germinoma (seminoma) mediastinal 734 pediatric malignancy 765 primary intracranial 649 chemotherapy/multimodal therapy 654 epidemiology 650 pathology 650 prognosis 654 radiation therapy 653 surgery 653 tumor markers 652 Gestational trophoblastic diseases 532 – 42 chemotherapy 532, 536 diagnosis 535 hydatidiform moles 535 – 6 prenatal ultrasonography 535 timing of 533, 535 epidemiology 532 – 3 etiology 532 – 3 fallopian tube 482 hCG production 532, 534, 535 – 6, 537 historical background 532 hydatidiform mole see Hydatidiform mole neoplastic see Gestational trophoblastic neoplasia (GTN) outcome 532 pathology 533 – 4 Gestational trophoblastic neoplasia (GTN) 536 – 7 chemotherapy treatment 532, 538 – 40 combination chemotherapy 539, 539 – 40 high-risk metastatic disease 539, 539 – 40 low-risk metastatic disease 539 nonmetastatic disease 538 – 9 PSTTs 540 secondary malignancy and 540 secondary treatment 540 single-agent 538 – 9 choriocarcinoma see Choriocarcinoma classification/staging 537 – 8, 538, 540 – 1 clinicopathologic definition 532, 536 diagnosis 537 metastases 537, 538 etiology 533 molar pregnancy 533, 536 follow-up 540
SUBJECT INDEX hCG production 532, 534, 537, 539 hysterectomy 539, 540 invasive mole 532, 536 – 7 metastases 534, 536 – 7, 538 outcome/prognosis 532 placental-site trophoblastic tumor (PSTT) 532, 534, 537, 540 radiation therapy 539, 540 risk stratification 538, 541 GETT classification system, thymomas 238, 248 GFAP-containing cytoplasm 690 Ghost tumor 661 – 2 Giant cell angioblastoma 582 Giant cell carcinoma (pleomorphic adenocarcinoma) oral cavity and adjacent structures 88, 90 pancreatic 370 – 1 Giant cell fibroblastoma (GCF) 580 Giant congenital nevus, pediatric melanoma 810 – 1, 811 Gingiva epithelioid hemangioendothelioma 98 melanoma 91 non-Hodgkin’s lymphoma vs. necrotic gingivitis 95 Gingivitis, non-Hodgkin’s lymphoma vs. 95 Glandular metaplasia bladder adenocarcinoma 21 urethra 28 Glandular tumors, cervical 504 – 8 Glassy cell carcinoma, cervical 509 Gleevec see Imatinib mesylate Glial-derived nerve growth factor (GDNF), MEN type 2 , 783 Glial progenitor cells 689 – 90 Glioblastoma (GBM) 674, 681 – 3 cerebral hemispheres 685 imaging 680 survival 678 Gliofibroma (desmoplastic glioma), pediatric 800 – 1, 801 Glioma 674 – 94 biology 675 – 7, 676 brain stem 687 – 8 clinical features/natural history 677 – 8 epidemiology 674 – 5 high-grade 679 – 80 histologic type 674 meningeal sarcomas vs. 630 metastases 678 pathological anatomy/neuroimaging 678 – 81, 679 pediatric malignancy 799 resistance to chemotherapy 676 spinal cord 688 Glivec see Imatinib mesylate Glomangiosarcoma (malignant glomus tumor) 582 Glomus jugulare tumors see Paraganglioma Glomus tumors 271 malignant (glomangiosarcoma) 582 Glottis 102 Glucagon, islet cell tumors 756 Glucagonoma 375 Glucocorticoids, Cushing’s syndrome and 143 Glutathione S-transferase M1 gene (GSTM1), nasopharyngeal carcinoma 115 Gluten-sensitive enteropathy (celiac disease) 392, 396 Goblet cell carcinoids, appendix 413 Gonadal function, late sequelae of chemotherapy 473 Gonadal (sex cord) stromal tumors 70 – 4, 456 history 455 Leydig cell tumors 71, 71 – 2, 456, 765, 784, 785 ovarian see Stromal tumors of the ovary
Sertoli cell tumors 72 – 4, 455, 456, 765, 784, 785 sex cord tumor with annular tubules (SCTAT) 455, 462 – 3 Gonadoblastoma, children (prepubertal) 764 – 5 Gorlin’s syndrome (nevoid basal cell carcinoma syndrome: NBCCS) 695, 811 – 2, 812 GPI-80 expression, thymomas/thymic carcinoma 245 Grading systems see Tumor classification/staging systems Granulocyte-colony stimulating factor (GSF), appendiceal carcinoid tumors 415 Granulocyte-macrophage colony stimulating factor (GM-CSF), pediatric melanoma 811 Granulocytic sarcoma 512 – 3, 528 French-American-British (FAB) subtype 770 ovarian involvement, pediatric malignancy 770 small bowel 398 Granulomas, Langerhans’ cell histiocytosis of CNS 610 – 1 Granulosa stromal cell tumors of the ovary 456, 459 – 61 adult granulosa cell tumors 459 – 60 cellular fibromas/fibrosarcomas 461 fibromas 461 fibrothecomas 461 juvenile granulosa cell tumors 460 recurrence/metastases 459 – 60 thecomas 461 Grave’s disease, thymic hyperplasia 240 Gray matter lesions, Langerhans’ cell histiocytosis of CNS 611, 611 Great vessels, primary neoplasms 736 – 7 Grimelius staining small cell carcinomas of the gastrointestinal tract 431 small cell undifferentiated carcinoma of prostate 39 Groin node dissection, vulvar squamous cell carcinoma 523 Gross tumor volume (GTV) coverage, nasopharyngeal carcinoma 123 Ground glass opacities (GGOs), bronchioloalveolar carcinoma 316 Growing teratoma syndrome 653 Growth disturbances adrenocortical tumors 778 Carney complex and 784 Growth factors adrenocortical tumors 149 meningiomas 639 MEN type 2 783 phyllodes tumor of the breast 210 T-large granular lymphocyte leukemia 551 see also specific growth factors Growth hormone, MEN1 and 782 gsp (GNAS1) proto-oncogene, Carney complex and 784 Gynandroblastoma, ovarian 463 Gynecologic malignancy see specific types/locations Gynecologic Oncology Group (GOG) borderline ovarian tumor therapy 451 – 2 cervical adenosquamous carcinoma 508 extra-ovarian primary peritoneal carcinoma 442 ovarian germ cell tumors 472 stromal ovarian tumor treatment 457 – 8 uterine sarcoma therapy 492, 493 Gynecomastia male breast cancer association 202 male breast cancer vs. 202 – 3
835 Hair follicles, tumors 577 Hair matrix, tumors 577 Hairy cell leukemia 547 – 9 Hassall’s corpuscles 239 Headache chordomas 617 neuroepithelial neoplasms 677 primary CNS lymphoma 659 Head and neck cancer esthesioneuroblastoma see Esthesioneuroblastoma (ENB) larynx 102 – 12 oral cavity and adjacent structures 87 – 101 pediatric malignancies 721 – 31 extraosseous Ewing’s sarcoma 721 – 3 odontogenic tumors 723 – 31 pediatric malignancy thyroid carcinoma 786 see also specific types/locations Head trauma, neuroepithelial neoplasms 675 Heart pleural mesothelioma and 282 tumors see Cardiac tumors Heavy chain disease (HCD) 571 – 2 Heavy metal exposure 27 Helicobacter pylori, 396 colorectal lymphoma 406 eradication 363 gastric lymphomas 361, 363 stomach cancers 352 Hemangioendothelioma epithelioid see Epithelioid hemangioendothelioma (EHE) Kaposiform 582 malignant see Angiosarcoma retiform (hobnail) 582 Hemangioma, epithelioid 98 Hemangiopericytoma (HPC) clitoris/vulva 766 – 7 lung 267, 267, 743 meningeal 646 oral cavity and adjacent structures 97 pediatric malignancy clitoris/vulva 766 – 7 lung parenchyma 743 uterine 493 Hemangiosarcoma see Angiosarcoma Hematogenous spread phyllodes tumor of the breast 213 thymomas/thymic carcinomas 245 Hematological disorders cancer of oral cavity and adjacent structures 95 – 7 diagnosis 96 leukemia see Leukemia lymphoma see Lymphoma pediatric see under Pediatric malignancy prostatic involvement 57 thymoma and 246, 246 vaginal tumors 528 Hematological toxicity, imatinib mesylate 423 Hematopoietic neoplasm classification 2 Hematoxylin and eosin (H&E), primary CNS lymphoma 658 Hematuria, renal cell carcinoma in children 760 Hemicolectomy appendiceal carcinoid 751 carcinoid tumors 403 Hemiscrotectomy, malignant mesothelioma of the tunica vaginalis 77 Hemoptysis, primary pulmonary lymphoma 259 Hemorrhage adrenocortical carcinoma 144 gastrointestinal gastric choriocarcinoma 355 – 6 gastric lymphomas 362 intratumoral 271
836 Hemorrhage (cont.) malignant fibrous histiocytoma (MFH) 633 primary lung sarcomas 272 thymoma/thymic carcinoma 240, 242 see also Bleeding Hendrickson and Kempson classification, uterine sarcomas 487 Hepatectomy, bile duct tumors 386 Hepatic angiosarcoma 387 Hepatic cystadenomas 388 Hepatic epithelioid hemangioendothelioma 387 – 8 Hepatic resection, fibrolamellar carcinoma 387 Hepatitis C infection, hepatic tumors 388 Hepatobiliary tumors 383 – 90 Hepatoblastoma 388 Hepatocellular carcinoma (HCC) 387 mixed 387 pediatric malignancy 749 Hepatocyte growth factor (HGF; scatter factor: SF), neurocutaneous melanosis and 605 Hepatoid adenocarcinoma 353 – 4 Hepatosplenic T cell lymphoma 562 Her-2/neu gene/protein male breast cancer 203 vulvar squamous cell carcinoma 522 Hereditary breast-ovarian cancer (HBOC) syndrome 439 Hereditary nonpolyposis colon cancer (HNPCC) Carney complex gene and 784 colorectal cancer in children 752 small bowel adenocarcinomas 392 5-HIAA, carcinoid tumors 377 Hidradenocarcinoma 578 Hidradenoma, malignant 578 Highly active antiretroviral therapy (HAART), primary CNS lymphoma 657, 664 Hilar node metastases, lung mucoepidermoid tumors 331 Hilar (Klatzkin) tumors 385 Hirschsprung’s disease 783 Hirsutism, stromal tumors of the ovary 456 Histamine release syndrome acute basophilic leukemia 547 mast cell leukemia 551 – 2 Histiocyte Society classification of LCH by MRI 611 pathogenesis of LCH 610 prognosis of LCH 613 Histiocytoid cardiomyopathy (cardiac hamartoma) 736, 737 Histiocytoma fibrous, pediatric bronchus 737, 738 malignant fibrous see Malignant fibrous histiocytoma (MFH) plexiform fibrous 580 – 1 Histiocytosis Langerhan’s cell see Langerhans’ cell histiocytosis (LCH) pediatric, malignant 813 Histiocytosis X see Langerhans’ cell histiocytosis (LCH) HIV/AIDS-associated lymphoma 565 non-Hodgkin’s lymphoma, oral cavity 95 primary CNS lymphoma 657, 658, 664 pathogenesis 658 – 9 therapeutic outcomes 661 treatment efficacy 664 primary pulmonary lymphoma 258 primary renal T cell lymphoma of childhood 762 HIV infection/AIDS anal carcinoma 407 cancers of oral cavity and adjacent structures bone tumors 89 Kaposi’s sarcoma 89, 96, 97 non-Hodgkin’s lymphoma 95
SUBJECT INDEX cervical tumors 516 drug treatment 97 gastrointestinal tumors 753 Kaposi’s sarcoma see Kaposi’s sarcoma (KS) Langerhans’ cell histiocytosis (LCH) 610 leiomyosarcoma cutaneous 816 meningeal 632 lymphoma and see HIV/AIDS-associated lymphoma pediatric malignancy cutaneous leiomyosarcoma 816 gastrointestinal tumors 753 lung parenchyma 742 primary renal T cell lymphoma of childhood 762 sebaceous carcinoma of the eyelid 712 small bowel sarcoma 396 squamous cell carcinoma of the conjunctiva 714, 715 Hobnail (retiform) hemangioendothelioma 582 Hodgkin’s disease (HD) 564 – 5 cutaneous, pediatric malignancy 813 primary pulmonary see under Primary pulmonary lymphoma (PPL) small bowel 396 Hormonal therapy adenoid cystic carcinoma of the breast 191 borderline ovarian tumors 451, 452 estrogen in prostate cancer, cancer due to 201, 202 male breast cancer etiology 201, 202 treatment 205 – 6 meningioma 644 metaplastic breast carcinoma 185 phyllodes tumor of the breast 214 primary adenocarcinoma of fallopian tube 480 – 1 recurrent meningioma 644 small cell undifferentiated carcinoma of prostate 45 stromal tumors of the ovary 458, 458, 460, 462 tamoxifen see Tamoxifen uterine sarcomas 492, 493 Hormone(s) disturbance see Endocrine disturbance islet cell tumors 756, 756 large cell neuroendocrine cancer (LNEC) of the lung 301 small cell carcinoma of the pancreas 372 see also specific hormones Horseshoe kidney 9, 10 Host factors, uveal melanoma 715 – 6 HRPT2 gene mutations, parathyroid carcinoma 175 Human achaete-scute homolog (hASH1) 135 Human chorionic gonadotrophin (hCG) 533 gestational trophoblastic diseases 532 choriocarcinoma 534 hydatidiform moles 534, 535 – 6 PSTT 537 treatment effects 539 ovarian germ cell tumors 471 thyroid-stimulating effects 534 Human herpes virus-6 (HHV-6), primary CNS lymphoma 659 Human herpes virus-8 (HHV-8), retiform hemangioendothelioma 582 Human immunodeficiency virus (HIV) see HIV infection/AIDS Human leukocyte antigen (HLA) alleles nasopharyngeal carcinoma 115 – 6 thymoma etiology and 239 Human pancreatic peptide (hPP), islet cell tumors 756
Human papillomavirus (HPV) 501 anal carcinoma 406 cervical adenoid basal carcinoma 509 cervical carcinosarcoma 511 cervical tumors 516 cervical verrucous carcinoma 502 pediatric squamous cell carcinoma 812 squamous cell carcinoma of the conjunctiva 713 – 4 transitional cell carcinoma of prostate 54 urethral cancer 27 verrucous carcinoma of the larynx 104 vulvar squamous cell carcinoma 521 Human telomerase reverse transcriptase (hTERT) chordoma prognosis and 616 meningioma 641 pheochromocytoma 157 Human T-lymphotrophic virus type 1 (HTLV-1) 561 – 2 cutaneous lymphoma in children 814 H¨urthle cell carcinoma 167 – 8 clinical presentation/diagnosis 167 total thyroidectomy vs. lobectomy 167 – 8 treatment 167 – 8 Hyaline cartilage, mesenchymal chondrosarcoma 630 Hybrid oncocytic tumor (HOT) 6 Hydatidiform mole 532, 534 – 6 complete 533 – 4 clinical presentation 534 – 5 diagnosis 535 pathologic features 533 – 4 diagnosis 535 – 6 timing of 533, 535 epidemiology 532 etiology 532 – 3 follow-up 536 hCG production 534, 535 – 6 karyotypes 533, 534 malignant potentials 536 partial 534 clinical presentation 535 diagnosis 535 pathologic features 534 post-molar GTN 536 see also Gestational trophoblastic neoplasia (GTN) treatment 536 Hydrocephalus choroid plexus papilloma/carcinoma 667, 668, 801 medulloblastoma/PNET 697 neurocutaneous melanosis 606 pineoblastoma 804 primary intracranial germ cell tumors 651 17α-hydroxylase, adrenocortical tumors 151 5-Hydroxytryptamine (5-HT; serotonin) 308 Hydroxyurea chronic neutrophilic leukemia 543 recurrent meningioma 644 Hyoid 102, 103 Hyperaldosteronism with low plasma renin see Conn’s syndrome Hypercalcemia clear cell ovarian carcinoma 452 MEN1 782 osteosclerotic myeloma 572 parathyroid carcinoma 175, 175 small cell ovarian carcinoma, children 175 Hypereosinophilic syndrome 544 – 5 Hyperfractionated radiation, glioblastoma 682 Hyperglycemia, glucagonoma 375 Hypernephroma see Renal cell carcinoma (RCC) Hyperparathyroidism MEN1 782 parathyroid carcinoma 174, 178
SUBJECT INDEX Hyperparathyroidism, and jaw tumor syndrome (HPJTS) 175 endocrine cancers 781 Hyperprolactinemia, male breast cancer association 202 Hypertension adrenocortical tumors adrenogenital syndrome 145 children/adolescents 778 Conn’s syndrome 147 pheochromocytoma 143, 155, 158, 787 – 8 surgery and 159 uterine sarcoma etiology 485 Hyperthyroidism complete hydatidiform moles 534 Grave’s disease and thymic hyperplasia 240 Hypoglycemia, insulinoma 375 Hypokalemia, Conn’s syndrome 147, 152 Hypophyseal vestige 113 Hypophysectomy, male breast cancer therapy 206 Hypothalamic disturbances, craniopharyngiomas 707 Hypothalamic – pituitary axis, Langerhans’ cell histiocytosis (LCH) of CNS 612, 612, 613 Hypothyroidism, medulloblastoma 702 Hysterectomy abdominal 501, 513, 516 aborted 516 bilateral salpingo-oophorectomy and see Total abdominal hysterectomy with bilateral salpingo-oophorectomy (TAH/BSO) cervical adenoid cystic carcinoma 510 cervical melanoma 513 ‘cut-thru,’ 515 – 6 extrafascial 516 gestational trophoblastic neoplasia 539, 540 hydatidiform mole treatment 536 radical 501, 510, 516 vaginal 501 Ifosfamide mesna and in recurrent meningioma 644 thymomas/thymic carcinomas 250, 251 Ileum 391 Imaging adrenocortical tumors 152, 152 – 3, 153 children/adolescents 779 – 80 borderline ovarian tumors 449 breast carcinosarcoma 223, 224 chordomas 618, 618 – 20 choroid plexus papilloma/carcinoma 667, 669 colon/rectum carcinoid tumors 403 – 4 craniopharyngiomas 706, 707 – 8 esthesioneuroblastoma 136 – 8 meningiomas 641 – 2 parathyroid carcinoma 176 pheochromocytoma 158, 158 – 9, 790 – 1 pleuropulmonary blastoma, pediatric 740 primary lung sarcomas 273 – 4 pseudomyxoma peritonei 411 renal angiomyolipoma 8 stromal tumors of the ovary 456 thymomas/thymic carcinomas 247 see also specific modalities Imatinib, bevacizumab and sunitinib (SU-011248) gastric endocrine cell proliferations 359 gastric stromal tumors 361 Imatinib mesylate 427 adenoid cystic carcinoma of the lung 325 chordoma therapy 622 chronic eosinophilic leukemia 545 chronic neutrophilic leukemia 544
dermatofibrosarcoma protuberans 580, 592 esophageal adenoid cystic carcinoma 341 gastrointestinal stromal tumors 345, 361, 422 – 4 intestinal leiomyosarcoma 405 progression and 424 – 5 resistance 425 small bowel GIST 397 hypereosinophilic syndrome 545 mast cell leukemia 552 phyllodes tumor of the breast 214 pleural mesothelioma 285 resistance 424, 425, 545 therapeutic response 424 thymomas/thymic carcinoma 646 Imatinib Target Exploration Consortium Study B2225, dermatofibrosarcoma protuberans 592 Imiquimod cream, extramammary Paget’s disease 579 Immune derangements, hairy cell leukemia 548 Immune reaction nasopharyngeal carcinoma 117 T-large granular lymphocyte leukemia 551 Immunocompromised host cervical tumors 516 primary CNS lymphoma, pathogenesis 658 – 9 primary effusion lymphoma 559 see also Immunodeficiency; Immunosuppression; specific causes Immunodeficiency pediatric melanoma 810 pediatric squamous cell carcinoma 812 primary CNS lymphoma 657 thymic dysplasia 240 thymoma and 246, 246 Immunodeficiency associated lymphoproliferative disorder (IALD) 565 Immunoglobulin A (IgA), nasopharyngeal carcinoma 115 Immunoglobulin D (IgD) myeloma 570 Immunoglobulin G (IgG), nasopharyngeal carcinoma 115 Immunoglobulin M (IgM) paraprotein, lymphoplasmacytic lymphoma 557 Immunohistochemistry (IHC) breast cancer non-Hodgkin’s lymphoma 196 tubular carcinoma 231, 232 cervical cancer mucinous adenocarcinomas 504 serous papillary adenocarcinoma 507 cutaneous malignancy dermatofibrosarcoma protuberans 589 melanoma, oral cavity 91 Merkel cell carcinoma 595, 595, 596 gastrointestinal malignancy esophageal carcinosarcoma 339 gastric choriocarcinoma 355 gastric stromal tumors 360 meningeal sarcomas chondrosarcomas 631 fibrosarcoma 629 leiomyosarcoma 632 malignant fibrous histiocytoma 633 rhabdomyosarcoma 631, 633 sarcomatosis 633 mesothelioma 281 metanephric tumors 12 neurological malignancy chordomas 616 choroid plexus papilloma/carcinoma 668 – 9 primary CNS lymphoma 658 pulmonary malignancy adenoid cystic carcinoma 322 angiosarcoma 270
837 fibrosarcoma 267 Kaposi’s sarcoma 270 – 1 large cell neuroendocrine cancer of the lung 300 – 1 leiomyosarcoma 266 malignant fibrous histiocytoma 269 primary lung sarcoma 266, 274 synovial sarcoma 268 – 9 Immunology, nasopharyngeal carcinoma 118 Immuno-proliferative small intestinal disease (IPSID) 395 – 6 Immunosuppression cervical lymphoma 513 Merkel cell carcinoma etiology 596 pediatric squamous cell carcinoma 812 – 3 primary pulmonary lymphoma 258 sebaceous carcinoma and 577 sebaceous carcinoma of the eyelid 712 Immunotherapy primary lung sarcomas 275 renal tumors, collecting duct (of Bellini) carcinomas (CDCs) 3 uterine sarcoma treatment 493 see also specific agents Immunovirological tests, nasopharyngeal carcinoma 118 “Incidentalomas,” adrenal 143, 160, 160 Infantile fibrosarcoma (IFS) kidney 762 Infantile pancreatic adenocarcinoma (pancreatoblastoma) 754, 755, 755 Infants, medulloblastoma 698 Infectious agents nongastric MALT lymphoma 558 vulvar squamous cell carcinoma 521 see also specific infections/organisms Infertility chemotherapy and 473 fertility-compromising surgery 512 see also Fertility preservation primary adenocarcinoma of fallopian tube 477 Inflammatory myofibroblastic tumor (IMT), pediatric airway 737 lung parenchyma 743, 743, 744 Inguinofemoral lymphadenectomy, vulvar squamous cell carcinoma 523 Inhibin, Leydig cell tumors 71 INI1 gene/protein, atypical teratoid/rhabdoid tumors (AT/RT) 806 Innervation, larynx 103 Insular carcinoma 169 Insulin insulinoma, levels in 375 islet cell tumors 756, 756 phyllodes tumor of the breast 210 Insulin-like growth factor-1 receptor (IGF-1R), adrenocortical tumors 149 Insulin-like growth factor-2 (IGF-2), adrenocortical tumors 149, 777 Insulinoma 375 Intensity-modulated radiation therapy (IMRT) chordomas 621 glioblastoma 682 nasopharyngeal carcinoma 123, 124, 124 – 6 Interferon(s) cervical melanoma 513 gastric endocrine cell proliferations 359 interferon alpha see Interferon-α (IFN-α) Interferon-α (IFN-α) appendiceal carcinoid tumors 415 colon/rectum carcinoid tumors 404 giant cell angioblastoma 582 hairy cell leukemia 549 interferon α-2b meningioma 644 – 5, 645 pediatric melanoma 811 topical, squamous cell carcinoma of the conjunctiva 714
838 Interferon-α (IFN-α) (cont.) meningioma 644 – 5 Merkel cell carcinoma (MCC) 599 pancreatic carcinoid 377 pleural mesothelioma 286 toxicity 645 Intergroup Rhabdomyosarcoma Study (IRS) classification system 79 embryonal rhabdomyosarcoma 512 Grouping System 528 mediastinal tumors 735 sarcoma treatment 52 pathology 721 – 2 prognostic factors 78 surgical treatment 79 Intergroup trial, nasopharyngeal cancer 127 Interleukin-2 (IL-2) pediatric melanoma treatment 811 thymomas/thymic carcinoma treatment 252 Interleukin-8 (IL-8), mesothelioma angiogenesis 282 Interleukin-10 (IL-10), primary CNS lymphoma 659 Interleukin-12 (IL-12), pediatric melanoma treatment 811 Intermediate (pseudostratified) cuboidal cells 114 Intermediate trophoblastic cells 533, 534, 537 International Extranodal Lymphoma Study Group (IELSG), testicular non-Hodgkin’s lymphoma (NHL) 66, 68, 70 International Federation of Gynecology and Obstetrics (FIGO) see Federation Internationale de Gynecologie et d’Obstetrique (FIGO) International Mesothelioma Interest Group (IMIG), staging system 283, 283 International Pediatric Adrenocortical Tumor Registry (IPACTR) 145 International Society of Pediatric Oncology (SIOP), paratesticular rhabdomyosarcoma chemotherapy 80 International Union Against Cancer (UICC) nasopharyngeal carcinoma 120 urethral cancer 30 Intestinal cells of Cajal 396, 397 Intestinal gastrointestinal stromal cell tumors (GISTs) 405 – 6 Intestinal leiomyosarcoma 405 – 6 Intestinal-type adenocarcinoma 506 – 7 Intracranial cavity chordomas 617, 617 primary germ cell tumors 649 – 56 Intracranial pressure, increased craniopharyngiomas 707 pineoblastomas 803 primary intracranial germ cell tumors 651 see also Hydrocephalus Intraductal papillary-mucinous tumors (IPMT) of the pancreas 371, 372 Intralobar nephrogenic rests (ILNR), adult Wilms tumor (nephroblastoma) 13 Intratumoral hemorrhage 271 Intravascular bronchoalveolar tumor, pediatric malignancy 743 Intravascular large B cell lymphoma 559 Intravascular lymphomatosis (angiotropic lymphoma) 258 Intravenous leiomyomatosis 494 Intraventricular radiation, primary intracranial germ cell tumors 653 Invasive carcinoma cervical, pregnancy and 517, 518 cribriform breast carcinomas 187, 230, 233 adenoid cystic carcinoma vs. 189 – 90 ductal breast carcinoma 222
SUBJECT INDEX Invasive mole (gestational trophoblastic) 532, 536 – 7 Iodine-131-metaiodobenzylguanidine, pulmonary carcinoid tumors 309 Ionizing radiation exposure cancer-induction see Radiation-induced cancers effect on mucosa of sinonasal tract 128 Iridectomy, melanoma 716 Irinotecan hydrochloride (CPT-11), colorectal carcinoma in children 753 Iris 715 melanoma 715, 716 Irradiation therapy see Radiation therapy Ischemia, mesenteric infarction 394 Islet cell tumors 373 – 4 etiology 374, 756 hormone production 756 incidence 373 metastases 376 nonfunctioning 375 – 6 pathology 373 – 4, 374 pediatric malignancy 755, 756 syndromes 374, 374 – 6, 756 treatment 376 – 7 types 368 Jaundice extrapulmonary small cell carcinomas of the pancreas 372 pancreatic cystadenocarcinoma 368 – 9 Jaw pain, bone tumors 88 Jejunum 391 adenocarcinoma 393 small cell carcinomas 433 Jugular foramen syndrome, nasopharyngeal carcinoma 118 Juvenile granulosa cell tumors of the ovary 771 of the testis 766 Juxtaglomerular cell tumor, renin-secreting 762, 763 Kadish staging, esthesioneuroblastoma 133 – 4 Kaposiform hemangioendothelioma 582 Kaposi’s sarcoma (KS) 270 – 1 antiretroviral therapy (HAART) 275 cancers of oral cavity and adjacent structures 89, 96, 97 Kaposiform hemangioendothelioma vs. 582 small bowel 396 Karnofsky performance status (KPS) brain stem gliomas 687 medulloblastoma 701 neuroepithelial neoplasms 677 primary CNS lymphoma 660 Karyotyping see Cytogenetics Kasabach – Merritt syndrome, Kaposiform hemangioendothelioma 582 Keratinizing squamous cell carcinoma (KTSC), thymic carcinoma 242 – 3, 243 Keratinocytes, pediatric squamous cell carcinoma 813 Keratin, thymic carcinoma 242, 244 17-Ketosteroids (17-KS), adrenocortical tumors 778, 779 Ki67 expression adrenocortical tumors 150 adult granulosa cell tumors 459 Kidney horseshoe 9, 10 stones, renal cell carcinoma in children 760 tumors see Renal tumors Kiel pediatric Tumor Registry 749 c-KIT gene/protein
AMN107 426 breast carcinosarcoma 227 gastrointestinal stromal cell tumors 418, 421, 421 germline mutations 405 imatinib mesylate and 422 – 4, 424, 424, 425 Merkel cell carcinoma 584 PKC412 426 sunitinib malate and 424, 425, 427 VEGFR 426 see also Imatinib mesylate; Sunitinib malate (SU11248, Sutent) Klatzkin tumors 385 Klinefelter’s syndrome male breast cancer association 202 primary intracranial germ cell tumors 649 – 50 KOC overexpression, pancreatic mucinous adenocarcinoma 371 – 2 K-RAS mutations, bronchioloalveolar carcinoma 314 Laboratory evaluation adenoid cystic carcinoma of the lung 324 adrenocortical tumors 151 – 2 children/adolescents 778 – 9 borderline ovarian tumors 450 cancers of oral cavity and adjacent structures 89, 96 gastrointestinal stromal cell tumors 421 pheochromocytoma 157 – 8 children/adolescents 789 – 90 dopamine-secreting 790 drugs affecting 158, 790, 790 renal failure 790 see also specific tests Lacrimal caruncle tumors 577 Lacrimal gland tumors 577 Lactate dehydrogenase (LDH) bone tumors 89 Burkitt lymphoma 559 hematologic malignancy 96 Lactation, ovarian cancer risk 447 Laminin, thymomas/thymic carcinomas and 238 Langerhans’ cell histiocytosis (LCH) 610 bone, oral cavity and adjacent structures 89 of central nervous system 610 – 3 clinical features/diagnosis 611 differential diagnoses 611, 612 MRI-based classification 611 – 2 pathology 610 – 1 prognosis 613 radiographic features 611, 611 – 2, 612 ‘risk organ’ involvement 613 treatment 612 – 3 Laparoscopy gastric endocrine cell proliferations 359 gastrointestinal stromal cell tumors 422 lymphadenectomies, cervical tumors 514 – 5, 515 primary adenocarcinoma of fallopian tube 480 stromal tumors of the ovary 456, 457 Laparotomy second-look ovarian stromal tumors 469 uterine sarcomas 490 stromal tumors of the ovary 456 – 7 Large bowel cancer 402 Large cell calcifying Sertoli cell tumors (LCCSCT) 74 Carney complex and 784 Large cell carcinoma (LCC) 300 classification 298 differential diagnosis 302 large cell neuroendocrine carcinoma (LNEC) 298, 299
SUBJECT INDEX differential diagnosis 302 lung see Large cell neuroendocrine carcinoma (LNEC) of the lung neuroendocrine differentiation 298, 299 with neuroendocrine differentiation (LCC-NED) 298, 299, 302 with neuroendocrine morphology (LCNEM) 298, 299 clinical characteristics 302 histological features 300 large cell neuroendocrine carcinoma (LNEC) vs. 302 thymic carcinoma 243 see also specific types Large cell neuroendocrine carcinoma (LNEC) of the lung 298 – 306, 299 biology/molecular biology 298 – 9 classification/definition 298 clinical characteristics 301, 302 combined 299 – 300 diagnosis/differential diagnosis 302 criteria 299 epidemiology 298 – 9 pathology 299 – 302 cytology 301 electron microscopy 300 – 1 gross 299 histologic features 299 – 300, 300 immunohistochemistry 300 – 1 prognosis 304 recurrence/metastases 304 treatment 302 – 3 chemotherapy 303 radiation therapy 303 recommendations 303 – 4 surgery 302 – 3 Laryngeal muscles 103 Laryngeal skeleton 102 – 3, 103 Laryngectomy chondrosarcoma 109 neuroendocrine laryngeal tumors 110 Larynx anatomy 102 – 3, 103 functions 102 histology 103 innervation 103 tumors 102 – 12 epithelial origin 103 – 6 mesenchymal origin 108 – 9 nasopharyngeal carcinoma and 118 neuroendocrine origin 109 – 10 salivary gland origin 106 – 8 see also specific types Laser therapy adenoid cystic carcinoma of the lung 325 cervical tumors 514 esophageal melanoma 346 pulmonary carcinoid tumors 309 retinoblastoma 718 urethral cancer 31 Lauren classification, stomach cancers 352 Leiomyoblastoma 494 Leiomyoma malignant transformation 485 uterine 485 benign metastasizing 494 Leiomyomatosis disseminated peritoneal 494 intravenous 494 Leiomyosarcoma cardiac 737 cervical 511 clitoris 767 colorectal 404 – 5 cutaneous 816 esophageal 343 hepatic 388 intestinal 405 – 6 meningeal 632, 632 radiation therapy 634
oral cavity and adjacent structures 97, 97 ovarian 770 pediatric malignancy clitoris/vulva 767 cutaneous 816 lung parenchyma 742 ovarian 770 renal 762 prostate 47 – 8 pulmonary 264 parenchyma, pediatric 742 primary sarcomas 266 renal 11, 762 small bowel 396, 397 – 8 superficial 582 uterine 485 chemotherapy 492, 492 – 3, 493 myomectomy 490 pathology 486 – 7, 487 symptoms/signs 486 vaginal 528 vulvar 525 – 6, 767 Lentigines, Carney complex and 783 Leptomeninges embryology 626 melanoma 608 metastases 678 pineoblastomas 803 pleomorphic xanthoastrocytoma 799 primary CNS lymphoma 658 Leu-2 expression, thymomas/thymic carcinomas and 238 Leu-3 expression, thymomas/thymic carcinomas and 238 Leukemia 543 – 54 acute myeloid 545 – 7, 546 B cell 547 – 50, 548 congenital 814 – 5 cutaneous involvement (leukemia cutis), children 814 – 5 myeloproliferative disease 543 – 5 natural killer (NK) cells 550 – 2 ovarian involvement, children 770 T cell 550 – 2 testicular involvement, children 765 treatment 815 see also specific types Leukemia cutis, children 814 – 5 Leukemoid reactions, chronic neutrophilic leukemia 543 Leukocoria 717, 718 Leukocyte alkaline phosphatase (LAP), chronic neutrophilic leukemia 543 Leukoencephalopathy 663 Leydig cells 70 hyperplasia 71 pediatric malignancy 765 tumors 71, 71 – 2, 456, 463 Carney complex and 784 Peutz – Jeghers syndrome and 785 Lhermitte’s phenomenon 688 Li – Fraumeni syndrome adrenocortical tumors 149, 776, 776 choroid plexus papilloma/carcinoma 668 phyllodes tumor of the breast 211 Lipid (steroid) cell tumors, ovarian 455, 456, 463 Lipoblastoma, cardiac 736 Lipomas, thymus 237 Liposarcoma cutaneous 583, 816 esophageal 345 meningeal 633 – 4 oral cavity and adjacent structures 97 – 8, 98 Liver disease, male breast cancer and 201 metastases appendiceal carcinoid tumors 415 carcinoid 394
839 pulmonary carcinoid tumors 311 primary tumors 387 – 8 see also specific types transplants, fibrolamellar carcinoma 387 see also entries beginning hepato-/hepatic Liver flukes, bile duct tumors 385 Lobectomy LNEC 302 – 3 pediatric, inflammatory myofibroblastic tumor 743 primary pulmonary lymphoma 261 total thyroidectomy vs. H¨urthle cell carcinoma 167 – 8 Localized fibrous mesothelioma 288 Loss of heterozygosity (LOH) adrenocortical tumors 150, 775, 777 breast carcinosarcoma 226 Carney complex and 785 glioblastoma 683 meningioma 640 – 1 Merkel cell carcinoma 596 mesothelioma 282 Peutz – Jeghers syndrome 785 – 6 sarcomatoid carcinoma 18 thymomas/thymic carcinomas and 238, 239 Zollinger – Ellison syndrome 374 Low malignant potential tumors, ovarian see Borderline ovarian tumors Lumpectomy (breast) adenoid cystic carcinoma 191 male breast cancer 204 metaplastic carcinoma 184 tubular carcinoma 232 Lung cancers see Lung cancer fibrosis 473 function tests 323 see also entries beginning pulmonary Lung cancer bronchial adenomas 329 pediatric malignancy 738 bronchioloalveolar carcinoma see Bronchioloalveolar carcinoma burden 321 carcinoid tumors see Pulmonary carcinoid tumors differential diagnoses 322 epidemiology 321 large cell 298 LNEC 298 – 306 mucoepidermoid tumors 329 – 35 non-small cell 298, 321, 329 primary adenoid cystic carcinoma 321 – 8 primary lymphoma see Primary pulmonary lymphoma (PPL) primary melanoma 293 – 7 primary sarcomas see Primary pulmonary sarcoma (PPS) small cell 321 TNM classification 321 WHO classification 298 see also specific types Lung Cancer Study Group (LCSG), bronchioloalveolar carcinoma 316 Lung consolidation, primary pulmonary lymphoma 259, 260 Lung parenchyma, pediatric neoplasms 739 – 43, 740 Lung transplantation, bronchioloalveolar carcinoma 316 Lye ingestion, verrucous carcinoma 338 Lymphadenectomy ameloblastic carcinoma 727 carcinoid tumors 403 esophageal carcinosarcoma 340 extra-ovarian primary peritoneal carcinomas 444 stomach cancers 352 vulvar squamous cell carcinoma 523
840 Lymphadenopathy bilateral inguinofemoral 523 cervical, esthesioneuroblastoma 137 immunoglobulin D myeloma 570 Lymphangiosarcoma see Angiosarcoma Lymphatic drainage, vulvar melanoma 525 Lymphatic mapping, vulvar squamous cell carcinoma 523 Lymph node involvement adenocarcinoid tumors of appendix 413 adenoid cystic carcinoma of the breast 190, 191 angiomatoid malignant fibrous histiocytoma 580 apocrine carcinoma 578 appendiceal carcinoid tumors 750 breast carcinosarcoma 225 cervical melanoma 513 cervical small cell carcinoma 511 cervical villoglandular papillary adenocarcinoma 505 cutaneous angiosarcoma 581 epithelioid sarcoma 580 esophageal carcinosarcoma 340 gallbladder tumors 383 gastric carcinoid tumors 358 – 9 large cell neuroendocrine carcinoma (LNEC) of the lung 303 lung mucoepidermoid tumors 331 male breast cancer 202 prognosis and 204, 204 malignant mesothelioma of the tunica vaginalis 76 Merkel cell carcinoma 594 – 5, 597 – 8 mesonephric carcinoma 508 nasopharyngeal carcinoma 122 – 3 paratesticular rhabdomyosarcoma 79, 81 parathyroid carcinoma 176 pediatric thyroid cancer 787 primary adenocarcinoma of fallopian tube 478 – 9 renal angiomyolipoma 7, 8 sebaceous carcinoma of the eyelid 712 stromal tumors of the ovary 457 testicular NHL 69 thymomas/thymic carcinomas 245 tubular carcinoma 231, 233 urethral cancer 27 vulvar squamous cell carcinoma 523 Lymph node smear, blastic NK cell lymphoma 560 Lymphoblastic lymphoma cutaneous lymphoma in children 813 mediastinal 733 Lymphocyte-depleted classical Hodgkin’s lymphoma (LDHL) 565 Lymphocyte-rich classical Hodgkin’s lymphoma (LECHL) 564 – 5 Lymphocytes B lymphocytes see B-cells (lymphocytes) primary CNS lymphoma 658 T lymphocytes see T cells (lymphocytes) Lymphocytic lobulitis, non-Hodgkin’s lymphoma (NHL), breast 196 Lymphoepithelioma bladder 23 – 4 nasopharyngeal carcinoma 118 Lymphoepithelioma-like carcinoma (LELC) cervical 503 cutaneous 579 gastric 353, 353 Lymphoepithelioma-like squamous thymic carcinoma (LETC) 243, 243, 244 Lymphoid cells, thymus 237 – 8, 240 Lymphoid neoplasms classification 2 see also specific types/locations Lymphoma 555 – 68 B cell neoplasms 557 – 60 bile duct tumors 385, 387
SUBJECT INDEX biological aspects 556 cervical involvement 493, 513 – 4 CNS involvement see Primary central nervous system lymphoma (PCNSL) colorectal involvement 406 cutaneous involvement 562 – 4, 813 – 4, 814 diagnosis 96 extranodal 257 differential diagnosis 273 Langerhans cell histiocytosis vs. 89 oral cavity tumors 99 historical background 555 HIV-related see HIV/AIDS-associated lymphoma management 96 – 7 ocular involvement 658, 659, 661, 663 – 4 oral cavity involvement 95, 96, 99 ovarian involvement 770 pancreatic involvement 373 pediatric malignancy cutaneous 813 – 4, 814 ovarian 770 penile 764 renal 762 testicular 765 – 6 penile involvement 764 pulmonary involvement 257 primary lymphoma see Primary pulmonary lymphoma (PPL) renal involvement 10, 762 small bowel cancer and 391, 395 – 6 testicular involvement 66 – 70, 67, 765 – 6 thyroid 169 uterine involvement 493 see also specific types Lymphomatoid granulomatosis, primary pulmonary lymphoma 259, 260 Lymphoplasmacytic lymphoma (LPL) 557 Lymphoreticular malignancies, mediastinal 733 Lymphoscintigraphy, Merkel cell carcinoma 598 Lynch syndrome I see Hereditary nonpolyposis colon cancer (HNPCC) Lynch syndrome II, pancreatic cancer and 367 MacLeod’s syndrome, adenoid cystic carcinoma vs. 323 Macroglobulinemia cutis, Waldenstr¨om macroglobulinemia 571 Macrophages, Langerhans’ cell histiocytosis (LCH) 610 Maffucci’s syndrome, juvenile granulosa cell tumors 460 Magnetic resonance angiography (MRA) extraosseous Ewing’s sarcoma 722 meningioma 642 nasopharyngeal carcinoma 119 Magnetic resonance cholangiopancreatography (MRCP), bile duct tumors 385 Magnetic resonance imaging (MRI) breast cancer breast carcinosarcoma 223 metaplastic breast carcinoma 183 non-Hodgkin’s lymphoma 197 phyllodes tumor of the breast 213 computed tomography (CT) vs. esthesioneuroblastoma 137 endocrine tumors adrenocortical tumors 152, 152, 779 – 80 medullary thyroid carcinoma 166 pheochromocytoma 158, 159, 790 – 1 primary thyroid lymphoma 169 extraosseous Ewing’s sarcoma 722 FLAIR images intracranial primary germ cell tumors 652
mass lesion 681 neuroepithelial neoplasms 680 gastrointestinal malignancy gastrointestinal stromal cell tumors (GISTs) 344 small bowel carcinoid 394, 395 small cell carcinomas 431 genitourinary malignancy prostate sarcoma 49, 50 renal angiomyolipoma 8 small cell undifferentiated carcinoma of prostate 44 urethral cancer 30 gynecological malignancy borderline ovarian tumors 449 primary adenocarcinoma of fallopian tube 480 uterine sarcoma 486 head and neck cancer bone tumors 89 – 90 esthesioneuroblastoma 137, 729 nasopharyngeal carcinoma, metastases 119 – 20, 120 oral cavity and adjacent structures 89 – 90, 93, 97 paranasal sinuses tumors 728 salivary tumors 93 soft tissue sarcomas 97 neurological malignancy atypical teratoid/rhabdoid tumors (AT/RT) 806 chordoma diagnosis 618 – 20, 619 choroid plexus papilloma/carcinoma 667, 669, 670, 801, 802 craniopharyngiomas 707 – 8 desmoplastic cerebral astrocytoma of infancy (DCAI) 798 desmoplastic glioma (gliofibroma) 801 desmoplastic infantile ganglioglioma (DIG) 798, 799 dysembryoplastic neuroepithelial tumors (DNT) 803 gliomas, high-grade 680 intracranial germ cell tumors 650, 651, 652, 652 intracranial primary germ cell tumors 652 Langerhans’ cell histiocytosis (LCH) 611, 611 – 2, 612 medulloblastoma/PNET 697 melanotic meningeal lesions 605, 606, 608, 807, 807 meningeal sarcomas 628, 628, 629, 632, 632, 634 meningioma 641 – 2 neuroepithelial neoplasms 679 pilocytic/pilomyxoid astrocytoma 686 pineoblastoma 803, 804 pleomorphic xanthoastrocytoma (PXA) 799, 800 primary CNS lymphoma 659, 660 spinal cord gliomas 688 in pregnancy 159 T1-weighted chordoma diagnosis 619 intracranial primary germ cell tumors 652 Langerhans’ cell histiocytosis of CNS 611, 611, 612, 612 meningeal sarcomas 628, 628, 629, 632 meningioma 641 – 2 T2-weighted chordoma diagnosis 619, 620 meningeal sarcomas 628, 632 meningioma 641 – 2 thoracic tumors adenoid cystic carcinoma of the lung 323 pediatric malignancy 732 pleural mesothelioma 282
SUBJECT INDEX primary melanoma of lung 295 thymomas/thymic carcinomas 247 Magnetic resonance spectroscopy (MRS) neuroepithelial neoplasms 681 primary CNS lymphoma 660 Male(s) adrenogenital syndrome 145 breast cancer see Male breast cancer (MBC) neuroepithelial neoplasms 675 reproductive system tumors pediatric malignancy 764 – 6 see also specific tumors/locations urethra anatomy 27 surgery 31 33 Male breast cancer (MBC) 201 – 8 adenoid cystic carcinoma of the breast 187 associated conditions 201 – 2 classification/staging 203 – 4 clinical presentation/diagnosis 202 – 3 differential diagnosis 202 – 3 epidemiology 201 etiology/biology 201 – 2, 202, 206 historical background 201 metastatic 203, 205 – 6 non-Hodgkin’s lymphoma 195 phyllodes tumors 209, 213 prognosis 203 – 4 axillary node status and 204, 204 metastatic disease 205 TNM classification 203 – 4 treatment 204 – 6 localized disease 204 – 5 metastatic disease 205 – 6 recommendations 206 Malignancy posttransplantation see Transplantation-related malignancy secondary see Secondary malignancies thymoma and 246, 246 Malignant acrospiroma (hidradenocarcinoma; malignant hidradenoma) 578 Malignant ameloblastoma 723 – 6 biology/patterns of spread 724 clinical presentation 724 differential diagnosis 725 embryology 723 epidemiology 723 evaluation 724 – 5 pathology 724 treatment 725 – 6 Malignant cylindroma 578 Malignant eccrine poroma (porocarcinoma) 578 Malignant eccrine spiradenoma 578 – 9 Malignant fibrous histiocytoma (MFH) 267 angiomatoid 580 cutaneous 580 dermatofibrosarcoma protuberans and 590 esophageal 345 hepatic 388 meningeal 633, 633 myxoid (myxofibrosarcoma) 580 oral cavity and adjacent structures 97, 98, 99 management 100 prognosis 97 pediatric malignancy 580 lung parenchyma 742, 743 pulmonary 269, 742, 743 renal 11 vulvar 525 – 6 Malignant fibrous xanthoma see Malignant fibrous histiocytoma (MFH) Malignant glomus tumor (glomangiosarcoma) 582 Malignant granular cell tumor (myoblastoma) (MGCT) 585 Malignant hemangioendothelioma see Angiosarcoma
Malignant hidradenoma (hidradenocarcinoma; malignant acrospiroma) 578 Malignant histiocytosis, pediatric 813 Malignant lymphocytes, primary CNS lymphoma 658 Malignant melanoma see Melanoma Malignant mesothelioma (MM) 279 – 92, 440 extra-ovarian primary peritoneal carcinomas vs. 437 pericardial 287 peritoneal 287 pleural see Pleural mesothelioma of the tunica vaginalis (MMTV) 75 – 7, 287 clinical features 76 epidemiology 75 etiology 75 investigations/staging 76 pathology 75 – 6 pediatric malignancy 766 recommendations 77 treatment 76 – 7 Malignant mixed mesodermal tumors (MMMTs), ovarian 452 – 3 Malignant mixed M¨ullerian tumor of fallopian tubes, children 771, 771 Malignant peripheral nerve sheath tumor (MPNST; neural sarcoma) 271, 271 cutaneous 584 – 5 pediatric malignancy 817 Malignant schwannoma, esophageal 345 Malignant syringoma (microcystic adnexal carcinoma) 579 Malignant teratoma 170 MALT lymphomas see Mucosa-associated lymphoid tissue (MALT) lymphomas Mammography breast carcinosarcoma 223, 224 metaplastic breast carcinoma 183 non-Hodgkin’s lymphoma 196 – 7 phyllodes tumor 213 tubular carcinoma 231 Manchester Children’s Tumour Registry (CTR), rhabdomyosarcoma 77 Mandible, bone tumors 87, 88, 89 Mantle cell leukemia 549 – 50 Mantle cell lymphomas, small bowel 396 Mapping biopsy, sebaceous carcinoma of the eyelid 713 Marginal zone lymphoma, cutaneous lymphoma in children 813 Masaoka staging system, thymomas 238, 246, 247 – 8 Mast cell leukemia 551 – 2 Mast cells, thymus 239 Mastectomy adenoid cystic carcinoma of the breast 191 breast carcinosarcoma 224, 225 male breast cancer 204, 206 metaplastic breast carcinoma 184 phyllodes tumors 214, 216 tubular carcinoma 232 Maternal age, trophoblastic disease and 532 – 3 Matrix metalloproteinases (MMPs) dermatofibrosarcoma protuberans (DFSP) 589 thymomas/thymic carcinoma 245 uterine sarcoma 486 Maxilla, bone tumors 87 M-BACOP regimen, testicular non-Hodgkin’s lymphoma 70 McCune – Albright syndrome 781 Median time to progression (MTP), meningioma 642 Mediastinal large B cell lymphoma 558 – 9 Mediastinal non-Hodgkin’s lymphoma (NHL) 258 Mediastinoscopy, primary lung sarcomas 274 Mediastinum
841 masses, treatment recommendations 252, 252 – 3 pleural mesothelioma 282 thymomas/thymic carcinoma 245 Mediterranean lymphoma, small bowel 396 Mediterranean sea coast, nasopharyngeal carcinoma 114 Medullary thymomas 240, 241, 242 Medullary thyroid carcinoma (MTC) 165 – 7 calcitonin elevation, management 167 clinical presentation/diagnosis 166 hereditary features 165 – 6 MEN2 and 783 nonoperative treatment 167 treatment 166 – 7 Medulloblastoma 695 – 704 adults 699 – 702 recurrence risk 701 biology/epidemiology 695 – 6 clinical presentation 695 – 7 neuroepithelial neoplasms 678 pediatric malignancy 698 – 9, 799 treatment 697 – 702 see also Primitive neuroectodermal tumor (PNET) ‘Medusa appearance,’ serous borderline tumors 448, 448 Megakaryoblastic leukemia, acute 546 – 7 Megavoltage therapy, chordomas 621 Meibomian glands 712 carcinoma 577 – 8 Melanin 605 Melanocytes aberrant migration 294 meningeal 605 uveal 715 Melanocytic meningioma 607 Melanocytoma, meningeal 605, 607 – 9 Melanogenic metaplasia 294 Melanoma 273 central nervous system (CNS) 811 cervical 513 choroid plexus 715, 716 ciliary body 715, 716 congenital 810 desmoplastic 91 gallbladder, primary 385 gastrointestinal colon/rectum/anus 407 – 8 esophageal 346 metastatic 398 small bowel 396 lung, primary 293 – 7 clinical presentation/diagnostic considerations 294 diagnostic criteria 293 histogenesis 294 incidence 293 – 4 metastasis patterns 295 pathology 294 – 5, 295 prognosis 296 treatment 295 management 91, 811 meningeal, primary 605, 607, 608 pediatric malignancy 806 – 7 metastases 91 meningeal 608, 806 – 7 small bowel 398 vulvar 525 misdiagnosis 91, 811 ophthalmic 715 iris 715, 716 uveal see Uveal melanoma oral cavity and adjacent structures 91 pediatric malignancy 810 – 1, 811 precursor lesions 810, 811 presentation 91, 811 vaginal 527 – 8 vulvar 524 – 5 clinical presentation 524
842 Melanoma (cont.) prognostic factors 524 – 5 treatment 525 Melanosis 346 Melanotic lesions epididymis 766 melanocytoma 695, 697 – 9 melanoma see Melanoma melanosis 346 neurocutaneous see Neurocutaneous melanosis (NCM) meningeal 605 – 9, 806 – 7 pediatric 806 – 7, 810 – 1 see also specific lesions Melanotic neuroectodermal tumor of infancy (MNTI), epididymis 766 Melena, GIST in children 753 Men see Male(s) MEN 1 gene, large cell neuroendocrine cancer of the lung 299 Meningeal sarcomas 626 – 37 angiosarcoma 633 chondrosarcoma 630 – 1, 634 classification/terminology 626, 627, 629 differential diagnosis/misdiagnosis 628, 630, 631 embryology 626 epidemiology 626 – 7, 627 etiology/models 627 – 8, 634 familial 628 fibrosarcoma 629, 629 – 30, 634, 635 gross pathology 628 – 9, 631, 632, 633 leiomyosarcomas 632, 632, 634 liposarcoma 633 – 4 malignant fibrous histiocytoma 633, 633 mesenchymal chondrosarcoma 630, 630, 630, 634 metastases 631 MRI staging 634 osteosarcoma 634 pediatric 627 presentation 628 prognosis 635 radiologic features 628, 628, 629, 631 rhabdomyosarcoma 631, 632, 635 sarcomatosis 632 – 3, 633 treatment 634 – 5 chemotherapy 634 – 5 radiation therapy 634 surgical 634 see also specific types Meningeal sarcomatosis 632 – 3, 633 menin gene/protein 782 Meninges embryology 626 hemangiopericytoma 646 melanotic lesions 605 – 9 melanocytic meningioma 607 melanocytoma 605, 607 – 9 neurocutaneous melanosis 605 – 7, 806 – 7 pediatric malignancy 806 – 7 primary melanoma 605, 607, 806 – 7 meningioma see Meningioma oral cavity rhabdomyosarcomas and 99 sarcomas see Meningeal sarcomas tumors classification 626, 638, 639, 639, 640, 646 see also specific tumors/lesions Meningioma 638 anaplastic (malignant) 638 – 48 cytogenetics 640 – 1 initial management 642 – 3 pathology 640 prognosis 645 – 6 recurrent disease management 643 – 5 anatomy 638 angioblastic (hemangiopericytoma) 646 atypical 638 – 48 cytogenetics 640 – 1
SUBJECT INDEX initial management 642 – 3 pathology 640 prognosis 645 – 6 recurrent disease management 643 – 5 benign 638 classification/terminology 638, 639, 639, 640, 646 clinical features 641, 641, 645 – 6 cytogenetics 640 – 1 diagnosis/imaging 641 – 2 epidemiology 638 – 9 etiology 640 – 1 growth pattern en plaque, 638 ‘mushrooming,’ 642 management chemotherapy 643 initial 642 – 3 medical/biologic therapy 638, 644 – 5, 645 metastases 645 radiation therapy 638, 642 – 3, 643, 644, 646 recommendations 645 recurrent disease 643 – 5 surgical resection 638, 642, 642, 644, 646 median time to progression 642 melanocytic 607 metastases 645 molecular biology 639, 641 neurological syndromes associated 641 papillary 640 pathology 639 – 40 radiation-induced 640 recurrence 642, 643 – 4, 645 rhabdoid 640 trauma and 640 Meningitis, neoplastic 608 Merkel cell carcinoma (MCC) 583 – 4, 594 – 603 aggressiveness 595 biological features 594 – 7 chemotherapy 598 – 9 clinical presentation 583, 583, 594, 595 diagnostic evaluation 597 differential diagnosis 583, 595, 596 epidemiology 583, 594 etiology 583, 596 intermediate (solid) type 595, 595 metastases common sites 594 lymph nodes 594 – 5, 597 – 8 molecular biology 596 – 7 neuroendocrine origin 594, 599 pathology 583 – 4, 595, 595 – 6 electron microscopy 584, 596 immunohistochemistry 583, 584, 595, 595, 596 “small blue cells,” 595, 596, 599 small cell (diffuse) type 595, 595 – 6 staging/prognosis 584, 594 – 5, 597 trabecular type 595, 596 treatment 584, 597 – 9 biologic agents 599 chemoradiation therapy 598 chemotherapy 584, 598, 598 – 9 radiation therapy 584, 598 sentinel lymph node biopsy 584, 597 – 8 stage and 597 surgical 584, 597 tumor appearance 594, 595 Merkel cells 594 Mesenchymal neoplasms classification 2, 387 – 8 esophageal 343 – 5 gastrointestinal 749, 753 laryngeal tumors 108 – 9 meningeal chondrosarcoma 630, 630, 630, 634, 635
uterine sarcoma 494 see also Mixed epithelial – mesenchymal tumors; specific types Mesenteric infarction 394 Mesonephric carcinoma, cervical 508, 508 Mesothelioma 75, 279 – 92 angiogenesis 282, 285 benign (localized fibrous) 288 biology 281 – 2 cytogenetics 281 – 2 differential diagnosis 281 epidemiology 279 – 81, 280 etiology 279 – 81 asbestos-related 279, 280, 280 familial aspects 280 – 1 non-asbestos-related 279 – 81 viral infection 281 histopathology 281 diagnostic 282 – 3 imaging 282 malignant see Malignant mesothelioma (MM) pleural see Pleural mesothelioma prognosis 287 treatment 283 – 7, 284, 285, 286 tunica vaginalis see under Malignant mesothelioma (MM) Metaiodobenzylguanidine (MIBG) scans colon/rectum carcinoid tumors 403 – 4 pheochromocytoma 158 Metanephric tumors 11 – 2 adenofibroma 11, 12 adenoma 11, 11, 12 background 11 classification 2 clinical presentation 12 pathology 11, 11 – 2 stromal (MST) 11, 12 pediatric malignancy 762 treatment/prognosis 12 Metanephrines, pheochromocytoma analysis 158, 789 – 90 Metaplastic carcinoma of the breast 181 – 6 adenosquamous carcinoma 181, 182, 182 – 3 biology 181 biphasic 181, 182, 184, 218 see also Breast carcinosarcoma classification/staging 181 epidemiology 181 fibromatosis-like (spindle cell) 181, 182, 182, 184, 223, 224 grade 183 historical background 181 monophasic 181, 182 see also Spindle cell (sarcomatoid) carcinoma nomenclature/definitions 104, 181 pathology 182, 182 – 3, 183 presentation/diagnosis 183 prognosis 183 – 4 recurrence/metastases 184 with sarcomatous metaplasia 183, 183 treatment 181, 184 – 5 see also specific types Metastases/metastatic disease see individual tumors/tumor types Metastatic sarcoma 273 Metatypical carcinoma see Basaloid squamous cell carcinoma Methotrexate gestational trophoblastic neoplasia 538, 539 hydatidiform mole treatment 536 medulloblastoma 700 neurotoxicity 663 ocular lymphoma 664 primary CNS lymphoma 662 see also specific combinations
SUBJECT INDEX Methotrexate, actinomycin D, and cyclophosphamide (MAC), gestational trophoblastic neoplasia 539 Methotrexate, vinblastine, doxorubicin, cisplatin (MVAC) collecting duct (of Bellini) carcinomas (CDCs) 3, 4 renal medullary carcinoma 4 transitional cell carcinoma of prostate 56, 57 urethral cancer 34 Methyl nitrosoureas, islet cell tumors 377 c-Met, neurocutaneous melanosis and 605 MGMT gene promoter 676 methylation in glioblastoma 683 MIC2 antibodies, thymomas/thymic carcinoma 245 Microadenocarcinoma, pancreatic 371 Microarray analysis dermatofibrosarcoma protuberans 589 Merkel cell carcinoma 596 – 7 Microcystic adnexal carcinoma 579 Microgemistocytes 690 Microglandular adenosis (MGA), adenoid cystic carcinoma of the breast and 187, 189, 190 Micropapillary bladder cancer 21 Micropapillary serous borderline ovarian tumor 448, 448 Microvessel density (MVD), mesothelioma 282 Mifepristone (RU 486), recurrent meningioma 644 Minimal deviation adenocarcinoma (adenoma malignum) 506, 506 Miscarriage, hydatidiform moles and 533 Mismatch repair genes, mucinous carcinomas 372 Mitomycin-C pseudomyxoma peritonei 411 topical sebaceous carcinoma of the eyelid 713 squamous cell carcinoma of the conjunctiva 714, 715 Mitotane (o, p -1,1-dichloro-diphenyl-dichloroethane; DDD) 153 – 4, 154, 780 Mitoxantrone, esophageal adenoid cystic carcinoma 341 Mixed cell tumors apocrine/eccrine gland 578 – 9 pancreatic 373 thymomas 240, 242 Mixed epithelial – mesenchymal tumors metaplastic carcinomas 181 uterine sarcomas 488, 488 – 9, 489 see also specific types Mixed hepatocellular and cholangiocarcinoma 387 MMH classification, thymomas 241 – 2 MMP-1, uterine sarcoma 486 MMP-2, uterine sarcoma 486 Mohs micrographic surgery (MMS) atypical fibroxanthoma 581 dermatofibrosarcoma protuberans 580, 590 – 1, 816 malignant eccrine poroma 578 Merkel cell carcinoma 597 microcystic adnexal carcinoma 579 pediatric basal cell carcinoma 812 plexiform fibrous histiocytoma 581 sebaceous carcinoma of the eyelid 713 sebaceous gland carcinoma 578 squamous cell carcinoma of the conjunctiva 714 superficial leiomyosarcoma 582 trichilemmal carcinoma 577 Molar pregnancy see Hydatidiform mole Molecular biology adrenocortical tumors 149 – 50 bile duct tumors 385
breast carcinosarcoma 226 – 7 chordomas 616 endometrial carcinomas 494 HNPCC 752 large cell neuroendocrine carcinoma (LNEC) of the lung 298 – 9 male breast cancer 202, 206 meningioma 639, 641 Merkel cell carcinoma (MCC) 596 – 7 multiple endocrine neoplasia (MEN) Carney complex 784 – 5 type 1 (MEN1) 782 type 2 (MEN2) 783 nasopharyngeal carcinoma 115 – 6 neuroepithelial neoplasms 675 – 6, 676 pancreatic cancers 367, 371 pheochromocytoma 155, 157, 789, 789 phyllodes tumor of the breast 210 – 1, 214 primary intracranial germ cell tumors 649 – 50 renal tumor diagnosis 1 – 2 retinoblastoma 717 rhabdoid tumor of the kidney 761 thyroid carcinoma, pediatric 786 tuberous sclerosis complex 7 uterine sarcoma 485 – 6 uveal melanoma 715 – 6 von Hippel-Lindau syndrome (disease) 786 Molecular imaging, esthesioneuroblastoma (ENB) 138 Molecular markers see Tumor markers Monoclonal gammopathy of undetermined significance (MGUS) 570 M protein, nonsecretory multiple myeloma 569 – 70 MTX, T-large granular lymphocyte leukemia 551 MUC1 gene, micropapillary bladder cancer 21 Mucin leakage, cervical mucinous adenocarcinomas 504 Mucinous adenocarcinomas appendiceal 412 cervical 504 pancreatic 371 – 2 Mucinous borderline ovarian tumors endocervical-like 449 gastrointestinal 448 – 9, 449 pediatric malignancy 769 Mucinous (adenocystic) carcinoma, cutaneous 579 Mucinous cystadenomas, pancreatic 368 Mucinous neoplasms appendiceal 412 cervical 504 cutaneous 579 ovarian 448 – 9, 449, 452, 769 pancreatic 368, 371 – 2 pseudomyxoma peritonei 411 renal 5 see also specific types Mucinous spindle cell carcinomas 5 Mucinous tubular cell carcinomas 5 Mucoepidermoid carcinoma (MEC) 171, 341 – 2 anatomy 329 – 30 breast, WHO classification 181 cervical 509 classification 329 historical background 329 lung 329 – 35, 330 anatomy 329 – 30 behavior/treatment 332 – 3 biology/epidemiology 330 clinical presentation/diagnosis 331 historical background 329 importance of grading 331 – 2 metastases 331 – 2 pathology 330, 330 – 1
843 prognosis 333 recommendations 333 – 4 oral cavity and adjacent structures diagnosis 89 – 90 presentation 88 salivary glands see under Salivary gland tumors pediatric malignancy bronchus 738, 738 trachea 737 – 8 thymus 243 – 4 Mucoid material, voiding in urachal cancer 24 Mucosa-associated lymphoid tissue (MALT) lymphomas 169 breast lymphomas 194 behavior/treatment 197 – 8 cytogenetics 196 immunohistochemistry 196 incidence 195 pathology 195 prognosis 198 extranodal marginal zone B cell lymphoma 558 high-grade 363 low-grade 362, 362, 363 pulmonary lymphomas 258 treatment/prognosis 261 – 2 small bowel lymphomas 396 Mucous cyst (“ranula”) of salivary gland 91, 93 Mucus antigen positivity, cervical mucinous adenocarcinomas 504 Muir – Torre syndrome, sebaceous gland carcinoma 577 M¨ullerian adenosarcoma cervical 511 – 2 pediatric malignancy 767 uterine, pediatric malignancy 768 M¨ullerian epithelium, primary adenocarcinoma of fallopian tube 479 M¨ullerian inhibiting factor (MIF) 438 M¨ullerian system, extra-ovarian primary peritoneal carcinomas 438 Multimodality treatment non-Hodgkin’s lymphoma (NHL), testicular 67 – 8 pleural mesothelioma 285 – 6 primary intracranial germ cell tumors 654 stromal tumors of the ovary 457 – 8 thymomas/thymic carcinomas 250 – 2 see also Chemoradiation therapy Multiple endocrine neoplasia (MEN) 165 – 6, 781 – 2 Carney complex see Carney complex (CNC) definitions 781 endocrine tumors in children 781 – 91 related syndromes 781 – 2 type 1 (MEN 1) 781, 782 carcinoid tumors 377 clinical presentation 782 gastric carcinoid tumors 358 islet cell tumors 374, 376, 756 molecular genetics 782 type 2 (MEN 2) see Multiple endocrine neoplasia type 2 (MEN 2) see also Cowden syndrome; Peutz – Jeghers syndrome; von Hippel – Lindau syndrome (disease) Multiple endocrine neoplasia type 2 (MEN 2) 781 carcinoid tumors 377 clinical presentation 782 – 3 molecular genetics 374, 783 pheochromocytoma 155, 783, 788 type 2A (MEN 2A) 783 islet cell tumors 374 pheochromocytoma 788 type 2B (MEN 2B) 783 childhood prostate cancer 764
844 Multiple endocrine neoplasia type 2 (MEN 2) (cont.) pheochromocytoma 788 Multiple myeloma nonsecretory 569 – 70 primary plasma cell leukemia 549 solitary plasmacytomas, after 572 Mural sarcomas 270 Muscle tumors, cutaneous manifestations 582 – 3 ‘Mushrooming,’ meningioma 642 Myasthenia gravis (MG) thymic hyperplasia and 240 thymomas and 245, 246 prognosis 253 c-myc oncogene medulloblastoma in children 698 meningioma 641 phyllodes tumor of the breast 211 rhabdoid tumor of the kidney and 761 Mycosis fungoides (MF) 562 – 3 cutaneous lymphoma in children 813, 814, 814 Myeloablative chemotherapy, medulloblastoma 701 Myeloma, jaw 88 Myeloproliferative disease 543 – 5 chronic eosinophilic leukemia (CEL) 544 – 5 chronic neutrophilic leukemia (CNL) 543 – 4 hypereosinophilic syndrome 544 – 5 Myoblastoma, malignant granular cell 585 Myoepithelial carcinoma, salivary glands 92 Myofibroblastic tumors 579 – 80 Myoid marker reactivity, gastrointestinal stromal cell tumors 419 Myomectomy, uterine sarcomas 490 Myxofibrosarcoma (myxoid malignant fibrous histiocytoma) 580 Myxoid chondrosarcoma, laryngeal 109 Myxoid dermatofibrosarcoma protuberans 580 Myxoid liposarcoma, oral cavity and adjacent structures 98, 98 Myxoid malignant fibrous histiocytoma (myxofibrosarcoma) 580 Myxoid smooth muscle tumors, uterine sarcoma 487 Myxomas cardiac 736, 783 Carney complex 783 – 4 Nasopharyngeal carcinoma biopsy 119 classification 116 – 8 differential diagnosis 728 immunological aspects 118 management/prognosis 121 – 8 non-endemic populations 113 – 32 anatomy of nasopharynx 113 – 4 clinical presentation 118 – 21 diagnosis 118 – 21 epidemiology/etiology 113 – 6 management of recurrence/metastases 128 – 9 pathology 116 pretreatment evaluation/follow up 128 staging 120 – 1, 121 routes of spread 123 squamous carcinoma 116, 116 undifferentiated 116 – 7, 117 Nasopharyngeal lymphoid tissue 113 Nasopharynx gross anatomy 113 microscopic anatomy 113 National Cancer Data Base, nasopharyngeal carcinoma 121 National Institutes of Health (NIH)
SUBJECT INDEX Gastrointestinal Stromal Tumors 421 staging for gestational trophoblastic neoplasia 537 National Wilms Tumor Study Group (NWTSG) adult Wilms tumor treatment 13 recommendations 14 clear cell carcinoma vs. 761 Natural killer (NK) cells granular lymphocytes leukemia 551 leukemia 550 – 2 neoplasms 560 – 2 T cell (NK-T) lymphomas, oral cavity 95 Nd:YAG laser therapy, esophageal melanoma 346 Neck dissection, ameloblastic carcinoma 727 Necrolytic migratory erythema, glucagonoma 375 Necrosis adenoid cystic carcinoma of the breast 190 – 1 adrenocortical carcinoma 144 malignant fibrous histiocytoma (MFH) 633 thymoma/thymic carcinoma 239, 240, 242 Necrotic gingivitis, non-Hodgkin’s lymphoma vs. 95 Needle biopsy, transitional cell carcinoma of prostate 55 Neonates primary germ cell tumors 650 thymus 238 Neoplastic meningitis 608 Nephrectomy adult Wilms tumor 14 collecting duct (of Bellini) carcinomas (CDCs) 3 renal lymphoma 10 Nephroblastic tumors, classification 2 Nephroblastoma see Wilms tumor (nephroblastoma) Nephrogenic adenofibroma, pediatric malignancy 763 Nephrogenic adenomas, urinary tract 28 Nephrogenic rests, adult Wilms tumor (nephroblastoma) 13 Nephroma, congenital mesoblastic 761 – 2 Nerve root compression, chordoma 617, 618 Neural antigens, gastrointestinal stromal cell tumors 419 Neural crest melanin containing cells 605 meningeal development 626 Merkel cells 594 Neuroblastoma mediastinal metastases 735 ovarian metastases 735 skin metastases 816 testicular metastases 765 Neurocutaneous melanosis (NCM) 605 – 7 associated disorders 605 biology 605 clinical features 606, 806 – 7 diagnosis/diagnostic criteria 606 – 7, 806 pathology 605 – 6, 806 pediatric malignancy 806 – 7 melanoma 810 – 1 treatment/prognosis 607, 807 Neuroectodermal tumors rhabdoid tumor of the kidney and 761 skin see Merkel cell carcinoma see also specific tumors/syndromes Neuroendocrine carcinoma of the larynx 109 – 10 Neuroendocrine tumors carcinoid see Carcinoid tumors classification 2, 109 colon/rectum carcinoid tumors 401, 402 esophageal 342 – 3 gastric 357 – 60 treatment 359 large cell lung cancer 298
laryngeal tumors 109 – 10 skin see Merkel cell carcinoma (MCC) small cell 109 see also specific types Neuroepithelial tumors 674 – 94 biology 675 – 7 clinical features 677 – 8 desmoplastic 798 – 801, 799 features 801 epidemiology 674 – 5 incidence 674 Merkel cell carcinoma see Merkel cell carcinoma natural history 677 – 8 pathological anatomy/neuroimaging 678 – 81 prevalence 675 prognosis 678, 678 see also individual types Neuroepithelioma see Primitive neuroectodermal tumor (PNET) Neurofibroma, mediastinal 735 Neurofibromatosis type I (NF1) see von Recklinghausen disease (neurofibromatosis type I; NF1) type II (NF2), meningioma and 641 clitoris/vulva malignancy, pediatric 767 Neurofibrosarcoma, cutaneous involvement in children 817 Neurogenic neoplasms, mediastinal 735 Neuroglia, meningeal fibrosarcoma 630 Neurological development, craniospinal radiation and 654 Neurological examination chordoma 617, 617 – 8 craniopharyngiomas 708 Langerhans’ cell histiocytosis (LCH) of CNS 611 Neurological syndromes, meningioma associated 641 Neurologic malignancy see Central nervous system (CNS) Neurologic symptoms/signs choroid plexus papilloma/carcinoma 669 meningeal sarcomas 628 meningioma associated 641, 641, 645 – 6 neuroepithelial neoplasms 677 small cell undifferentiated carcinoma of prostate 43 Neuromuscular disorders, thymoma and 246, 246 Neuron-specific enolase (NSE) esthesioneuroblastoma 135 small cell undifferentiated carcinoma of prostate 41, 42 Neuropeptide Y (NPY), pheochromocytoma 156 Neuropsychological assessment, medulloblastoma 701 – 2 Neurosecretory granules, colon/rectum carcinoid tumors 401, 402 Neurotoxicity, primary CNS lymphoma 663 Neutropenia, T-large granular lymphocyte leukemia 551 Neutrophilia, chronic neutrophilic leukemia 543 Nevi, congenital see Congenital nevi Nevoid basal cell carcinoma syndrome (NBCCS; Gorlin’s syndrome) 811 – 2, 812 NF1 gene peripheral nerve sheath tumors 585 pheochromocytoma 155 NF2 gene, meningioma and 641 Nitrosourea-based chemotherapy fetal exposure, neuroepithelial neoplasms 675 – 6 glioblastoma 683 medulloblastoma 700 neurocutaneous melanosis (NCM) 607
SUBJECT INDEX spinal cord gliomas 689 N-myc, medulloblastoma/primitive neuroectodermal tumor (PNET) 696 N -nitrosamines, stomach cancers 352 Nodal marginal zone B lymphoma 558 Nodal metastasis see Lymph node involvement Nodular lymphocyte predominant Hodgkin’s lymphoma (NLPHL) 564 Noguchi classification, bronchioloalveolar carcinoma 314 Nonendometrioid carcinoma 488 Nonfunctioning islet cell tumors (NIT) 375 – 6 Nongastric MALT lymphoma 558 Non-gastrointestinal stromal cell tumors, primary esophageal sarcomas 345 Nongerminal sarcoma, mediastinal tumors 734 Nongerminomatous germ cell tumors 649 epidemiology 650 primary intracranial chemotherapy/multimodal therapy 654 pathology 650 – 1 prognosis 654 radiation therapy 653 – 4 surgery 653 Non-Hodgkin’s lymphoma (NHL) breast see Non-Hodgkin’s lymphoma (NHL), breast cervical lymphoma vs. 513 classification 195 cutaneous, pediatric malignancy 813 – 4 extranodal 194, 257 see also specific sites historical aspects 194 mediastinal 258 oral cavity and adjacent structures 95, 95 – 7 bone destruction 88 diagnosis 96 differential diagnosis 95, 96 HIV/AIDS and 95 management 96 – 7 presentation 95, 96 pediatric malignancy 749 cutaneous 813 – 4 testicular 69 primary bile duct tumors 387 primary pulmonary see under Primary pulmonary lymphoma (PPL) testicular see Non-Hodgkin’s lymphoma (NHL), testicular tracheobronchial 258 Non-Hodgkin’s lymphoma (NHL), breast 194 – 200 biology 194 – 5 Burkitt-type 195, 196, 198 classification/staging 195, 197 clinical presentation/diagnosis 196 – 7 cytogenetics 196 definition 194 epidemiology 194 – 5, 195 historical background 194 immunohistochemistry 196 lymphocytic lobulitis and 196 male 195 MALT-type 194 behavior/treatment 197 – 8 histopathology 195, 196 incidence 195 prognosis 198 misdiagnosis 196 pathology 195 – 6 primary (PBL) 194 prognosis 198 recommendations 198 secondary 194 treatment 197 – 8 Non-Hodgkin’s lymphoma (NHL), testicular 66 – 70
age distribution 66, 67, 69 classification/staging 67, 69 prognosis and 68, 68 clinical features 69 epidemiology 66 – 7, 67, 67 etiology 67 history 66 pathology 67 pediatric malignancy 765 – 6 prognosis 66, 67 – 9, 68 relapse rates/patterns 67 – 9 sites of relapse 68 – 9, 69 treatment chemotherapy 70 multimodal 67 – 8 radiation therapy 69 – 70, 70 recommendations 70 surgery 69 Nonkeratinizing squamous cell carcinoma, thymic carcinoma 243 Nonsecretory multiple myeloma 569 – 70 Non-small cell lung cancer (NSCLC) 298, 321, 329 mesothelioma vs. 281 prognosis 304 see also specific types Norepinephrine, pheochromocytoma and 156, 157, 158, 789 Notochord, chordoma origin 614 Nuclear imaging esthesioneuroblastoma, regional/distant disease 137 medullary thyroid carcinoma (MTC) 166 pheochromocytoma 159 see also specific techniques Nuclear pleomorphism, adrenocortical carcinoma 144 Nucleoside analogs hairy cell leukemia 549 Waldenstr¨om macroglobulinemia 571 Nulliparous women ovarian cancer risk 447 phyllodes tumor of the breast 209 Oat cell-type tumors, small cell carcinomas of the gastrointestinal tract 431 Ober classification, uterine sarcomas 486 Obesity male breast cancer and 201 uterine sarcoma etiology 485 Occupational factors male breast cancer 202 mesothelioma 279, 280 OCT 3/4, dysgerminoma 468 Octreoscan pheochromocytoma 158 Zollinger – Ellison syndrome 374 Octreotide acetate (Sandostatin ) carcinoid syndrome diarrhea 395 colon/rectum carcinoid tumors 403 hepatic carcinoid 388 islet cell tumors 376, 756 large cell neuroendocrine carcinoma of the lung 303 thymomas/thymic carcinoma treatment 252 Ocular lymphoma 663 – 4 primary CNS lymphoma 658, 659 radiation therapy 661 Ocular surface squamous neoplasia (OSSN) 713 Odontogenic carcinoma 88 management 90 metastases 88 Odontogenic tumors 88, 723 – 31, 725 diagnosis 89 – 90 presentation 88 see also specific types Olfactory groove, meningioma 641
845 Olfactory neuroblastoma (ONB) see Esthesioneuroblastoma (ENB) Oligocilia, adenoid cystic carcinoma 322 Oligodendroglial neoplasms 689, 689 – 91 see also specific types Oligodendroglioma 688 low-grade 690 – 1 Ollier’s disease, juvenile granulosa cell tumors 460 Oncocytic adrenocortical tumors 149 Oncocytic carcinoma gastric 356 – 7 pancreatic 372 – 3 Oncocytic cardiomyopathy (cardiac hamartoma) 736, 737 Oncocytoma (oncocytic adenoma) 5 – 7 background 5 clinical presentation 6 – 7 differential diagnosis 6 metastatic 7 origins 6 pathology 5 – 6, 6 treatment/prognosis 7 Oncocytosis 6 Oncoplastic procedures, phyllodes tumor of the breast 213 – 4 Oophorectomy prophylactic, extra-ovarian primary peritoneal carcinomas 439 unilateral, borderline ovarian tumors 450 Ophthalmic cancers 712 – 20 anatomical classification 712, 713 conjunctival 713, 713 – 5 eyelid 712 – 3, 713 melanoma 715 iris 715, 716 uveal see Uveal melanoma ocular lymphoma 661 optic pathway glioma 685 retinoblastoma see Retinoblastoma see also specific types/locations Opisthorchis viverrini, bile duct tumors 385 Opportunistic infection, hairy cell leukemia 548 Optic pathway glioma (OPGs) 685 Oral cavity cancers 87 – 101, 579 bone tumors 87 – 90, 88, 89 hematologic 95 – 7, 96 melanoma 91 metastases 89, 91 salivary gland tumors see Salivary gland tumors soft tissue sarcomas 97, 97 – 100, 98, 99 squamous cell carcinoma 87 verrucous carcinoma 87, 88 see also specific types/locations Oral cavity care, nasopharyngeal carcinoma 128 Oral contraceptives hydatidiform moles and 533 ovarian cancer risk 447 Orbit, cancers 713 Orchidectomy adenocarcinoma of the rete testes 75 Leydig cell tumors 72 male breast cancer therapy 205 – 6 malignant mesothelioma of the tunica vaginalis 76, 77 non-Hodgkin’s lymphoma 69 paratesticular rhabdomyosarcoma 79, 81 Sertoli cell tumors 73 Orexin, adrenocortical tumors 149 Organ transplantation bone marrow, pineoblastoma 804 liver transplants, fibrolamellar carcinoma 387 lung transplantation, bronchioloalveolar carcinoma 316 Merkel cell carcinoma etiology 596
846 Organ transplantation (cont.) posttransplantation lymphoproliferative disease 10, 396 Ossifying renal tumor of infancy 762 Osteolysis, meningeal sarcomas 628 Osteomyelitis Langerhans cell histiocytosis vs. 89 oral cavity lymphoma vs. 96 Osteosarcoma esophageal 345 meningeal 634 oral cavity and adjacent structures 88, 89, 90 renal involvement 11 pediatric malignancy 763 Osteosclerotic myeloma 572 – 3 Otitis media chronic 128 nasopharyngeal carcinoma 118 Ovarian cancer borderline tumors see Borderline ovarian tumors classification/staging 769, 769 endometriosis and 449 epithelial 447 – 54 germ cell tumors see Ovarian germ cell tumors (OGCT) metastases appendiceal adenocarcinoma 413 small bowel 398 pediatric malignancy 768 – 71 risk factors 447 staging 769, 769 stromal tumors see Stromal tumors of the ovary see also specific types Ovarian cyst(s) complete hydatidiform moles and 535 dermoid, germ cell tumors 467 – 8 Ovarian cystectomy borderline ovarian tumors 450 stromal tumors of the ovary and 457 Ovarian germ cell tumors (OGCT) 467 – 76 chemotherapy 472 – 3 late sequelae 473 management of residual/recurrent disease 472 – 3 toxicity 473 clinical features 470 – 1 historical background 467 malignant 468 – 70 mixed 469 pathology 467 – 70 dermoid cysts 467 – 8 surgical evaluation 471 – 2 surgical staging 472 Ovarian stromal sarcoma (endometrial type), pediatric malignancy 770 Ovine pulmonary adenomatosis 313 – 4 p16INK4 gene/protein adrenocortical tumors 149 large cell neuroendocrine cancer of the lung 299 malignant peripheral nerve sheath tumors 585 pancreatic cancer and 367 primary intracranial germ cell tumors 650 p21 gene/protein, adrenocortical tumors 149 p53 gene/protein adrenocortical tumors 149, 150 pediatric malignancy 775, 776, 776 atypical fibroxanthoma and 581 breast carcinosarcoma 226 bronchioloalveolar carcinoma 318 chordomas 616 large cell neuroendocrine cancer of the lung 299
SUBJECT INDEX malignant peripheral nerve sheath tumors 585 meningioma 641 Merkel cell carcinoma 596 neuroepithelial neoplasms 676 overexpression, extra-ovarian primary peritoneal carcinomas 440 pheochromocytoma 157 phyllodes tumor of the breast 211 primary adenocarcinoma of fallopian tube 477 – 8 thymomas/thymic carcinoma 245 uterine sarcoma 485 vulvar squamous cell carcinoma 522 p63 gene/protein, breast cancer adenoid cystic carcinoma 189 carcinosarcoma 226 – 7 tubular carcinoma 231, 232 p73 gene/protein, Merkel cell carcinoma 596 Paclitaxel extra-ovarian primary peritoneal carcinomas 443 mucoepidermoid tumors 332, 333 see also specific combinations Paclitaxel, carboplatin and etoposide (PCE), small cell carcinomas of the gastrointestinal tract 432 Paget’s disease extramammary 579 perianal region 408 urethra 30 male breast cancer and 202 Palliative therapy bile duct tumors 386 colon/rectum carcinoid tumors 404 gastric endocrine cell proliferations 360 laryngeal neuroendocrine tumors 110 pancreatic carcinoid 377 paranasal sinuses tumors 728 phyllodes tumor of the breast 216 pleural mesothelioma 284 primary CNS lymphoma 660 primary lung sarcomas 275 small cell undifferentiated carcinoma of prostate 45 Pamidronate, primary plasma cell leukemia 549 Pancoast tumor, adenoid cystic carcinoma vs. 323 Pancreas, functions 373 Pancreatectomy, cystadenocarcinoma 369 Pancreatic adenosquamous carcinoma (adenocanthoma) 369 – 70 Pancreatic carcinoma cystic 371 small bowel metastases 398 Pancreatic cholera syndrome 375 Pancreatic cystadenocarcinoma 368 – 9 Pancreatic cystic neoplasms 371 – 2 Pancreatic ductal adenocarcinoma 367 – 8, 368 pediatric malignancy 754, 755 Pancreaticduodenectomy 755 bile duct tumors 386 Pancreatic lymphoma 373 Pancreatic microadenocarcinoma 371 Pancreatic mixed cell tumors 373 Pancreatic oncocytic carcinoma 372 – 3 Pancreatic pleomorphic adenocarcinoma 370 – 1, 371 Pancreatic polypeptide 376 Pancreatic sarcoma 373 Pancreatic small cell carcinomas 433 Pancreatic tumors 367 – 82 background 367 – 8 classification 367, 368 diagnosis 368 differential diagnoses 369, 371 endocrine 367, 373 – 8, 754 epidemiology 367
etiology 367, 371, 374 exocrine 367, 368 – 73, 754 familial associations 367, 374 metastases 368, 376, 398 mixed cell 373 pediatric malignancy 754 – 6, 755 acinar cell carcinoma 754 ductal adenocarcinoma 754 incidence 754, 755 islet cell tumors 755, 756, 756 management 755 – 6 pancreatoblastoma 754, 755 solid-pseudopapillary tumor 754 – 5, 755 small cell carcinomas 433 survival 367 symptoms/signs 368 see also specific types Pancreatitis, pancreatic cancer and 367, 371 Pancreatoblastoma 372, 754, 755, 755 Panmyelosis with myelofibrosis, acute 547 Panniculitic lymphoma, cutaneous lymphoma in children 814 Panorex roentgenogram, ameloblastoma 724 Papanicolaou (Pap) test 501, 526 adenoma malignum (minimal deviation adenocarcinoma) 506 cervical adenoid basal carcinoma 509 primary adenocarcinoma of fallopian tube 480 uterine sarcoma 486 Papillary adenocarcinoma, aggressive digital 578 Papillary adenoma 2 Papillary and cystic carcinomas, pancreatic 371, 372 Papillary carcinomas differentiated thyroid carcinoma 786 gallbladder tumors 384 Papillary craniopharyngiomas 707, 708 Papillary meningioma 640 Papillary mesothelioma 288 Papillary serous ovarian carcinoma (PSOC) 437 Papillary squamotransitional carcinoma 503 Papilloma choroid plexus 667 – 73 urethra 28 Pap smear/test see Papanicolaou (Pap) test Parafollicular C cell, medullary thyroid carcinoma (MTC) 165 Paraganglioma 155, 361 dopamine-secreting 790 familial 788 laryngeal 109 molecular genetics 157, 789 Parametrectomy, radical 516 Paraneoplastic symptoms/syndrome acinar cell carcinoma 370 pleural mesothelioma 282 primary lung sarcomas 272 small cell undifferentiated carcinoma of prostate 43 thymoma 246, 246 Parapharyngeal space, soft tissue sarcomas 97 Paratesticular rhabdomyosarcoma 77 – 81 clinical features 78 epidemiology 77 history 77 investigations 78 – 9 pathology 77 – 8 prognosis 78 treatment 79 – 81 chemotherapy 80 – 1 outcomes 81 radiation therapy 79 – 80 recommendations 81 surgical 79 Paratesticular tumors 66 – 85 embryology 66, 67
SUBJECT INDEX see also specific types Parathyroid carcinoma 174 – 9 anatomy/physiology 174 clinical presentation 175 – 6 diagnosis 178 epidemiology 174 genetics 175 operative findings 176 pathology 174 – 5 prognosis 177 – 8 recurrence/survival rates 178 treatment 177, 178 Parathyroid disease, MEN 2 and 783 Parathyroid hormone (PTH) islet cell tumors 756 parathyroid carcinoma 176, 176 small cell undifferentiated carcinoma of prostate 38 Parenchymal lesions, Langerhans’ cell histiocytosis of CNS 612 Parenchyma-sparing surgery, pulmonary carcinoid tumors 309 Paresthesias, nasopharyngeal carcinoma 118 Parietal cell carcinoma, gastric 356 – 7 Parinaud’s syndrome, primary intracranial germ cell tumors 651 Parous women, ovarian cancer risk 447 Paroxysmal hypertension, pheochromocytoma 787 – 8 Pediatric malignancy cancer diagnoses 798 cutaneous/subcutaneous tumors 810 – 8 basal cell carcinoma 811 – 2, 812, 812, 813 dermatofibrosarcoma protuberans 579, 590, 815 – 6, 816 Ewing’s sarcoma 817 fibrosarcoma 815 leiomyosarcoma 816 leukemia 814 – 5 liposarcoma 816 lymphoma 813 – 4, 814 malignant fibrous histiocytoma (MFH) 580 malignant histiocytosis 813 malignant peripheral nerve sheath tumors 817 melanoma 810 – 1, 811 neuroblastoma 816 rhabdomyosarcoma 815 squamous cell carcinoma 812, 812 – 3 synovial sarcoma 817 endocrine tumors 775 – 97, 782 adrenocortical tumors 144, 145, 775 – 81 adrenogenital syndrome 145 Cushing’s syndrome 144, 144, 778 MEN and related syndromes 781 – 91 pheochromocytoma 787 – 91 thyroid nodules and 786 – 7 gastrointestinal tract tumors 749 – 59 carcinoid tumors 749 – 51 colorectal carcinoma 751 – 3 GIST 753 – 4 hepatoblastoma 388 pancreatic 372, 754 – 6 small bowel 392 genitourinary tumors 760 – 74 bladder tumors 763 – 4 non-Hodgkin’s lymphoma 69 prostate sarcomas 49 renal tumors 760 – 3 rhabdomyosarcoma 77 testicular tumors 763, 764 – 6 Wilms tumor (nephroblastoma) 1, 12 Xp11 translocations 5 gynecologic malignancy alveolar soft part sarcoma 767, 768, 768 cervical tumors 512, 767 – 8 ovarian tumors 460, 768 – 71 head and neck cancer 721 – 31
extraosseous Ewing’s sarcoma 721 – 3 odontogenic tumors 723 – 31 thyroid carcinoma 786 hematologic malignancy acute megakaryoblastic leukemia 547 non-Hodgkin’s lymphoma 69, 749, 813 – 4 neurological malignancy 798 – 809, 799 atypical teratoid/rhabdoid tumors 805 – 6 brain stem 681, 687 chordomas 614 choroid plexus papilloma/carcinoma 667, 670, 801 – 2, 802 craniopharyngiomas 705 desmoplastic 798 – 9, 798 – 801, 799, 799 – 800, 800, 800 – 1, 801 dysembryoplastic 802 – 3, 803 ependymoblastoma 805 ependymoma 678, 689, 691 epidemiology 798, 799 gliofibroma 800 – 1 glioma 687 infantile astrocytoma/ganglioglioma 798 – 9, 799 medulloblastoma 698 – 9 melanocytic 806 – 7 meningeal 627 neuroepithelial neoplasms 675 pineoblastomas 803 – 4 pleomorphic xanthoastrocytoma 799 – 800, 800 primary germ cell tumors 650 primitive neuroectodermal tumor (PNET) 696 retinoblastoma 717, 804 – 5 rhabdomyosarcoma see Rhabdomyosarcoma (RMS) sequelae/adverse affects of craniospinal radiation 654 thoracic tumors 732 – 48 airways 737 – 9 chest wall 744 diaphragm 744 epidemiology 732 histology 732 mediastinal 733 – 7 parenchymal 739 – 43 pleura 744 symptomatology 732 see also specific tumors/locations Pediatric Oncology Group (POG) choroid plexus carcinoma 802 desmoplastic infantile astrocytoma/ganglioglioma 799 medulloblastoma 700 pineoblastoma 804 Pellagra, carcinoid syndrome 750 Pelvic exenteration, embryonal rhabdomyosarcoma 512 Pelvic imaging studies, primary adenocarcinoma of fallopian tube 480 Pelvic irradiation, embryonal rhabdomyosarcoma 512 Pelvic lymphadenectomy, vulvar squamous cell carcinoma 523 Pelvic lymph nodes, urethral cancer 27 Pelvic mass, clear cell ovarian carcinoma 452 Pelvic pain borderline ovarian tumors 449 endometrioid ovarian carcinoma 452 Pelvic tuberculous, primary adenocarcinoma of fallopian tube 477 Pelvic ultrasonography, prostate sarcomas 49 Pemetrexed, bronchioloalveolar carcinoma 318 Penectomy, radical, urethral cancer 33 Penile carcinoma, pediatric malignancy 764 Penis Paget’s disease 579
847 pediatric malignancy 764 Pentostatin 549 Perianal region, Paget’s disease 408 Pericardium mesothelioma 287 primary neoplasms 736 – 7 teratomas 737 Perilobar nephrogenic rests (PLNR), adult Wilms tumor (nephroblastoma) 13 Perineal leiomyosarcoma, pediatric malignancy 767 Periodic – acid Schiff (PAS) staining, mucin, adenoid cystic carcinoma 321 Periodic Cushing’s syndrome (PCS) 784 Periorbital ecchymosis, neuroblastoma 816 Peripheral primitive neuroectodermal tumor (pPNET), cervical 514 Peritoneal mesothelioma 287 Peritoneal serous borderline tumors (PSBTs) 439 Perivascular epithelioid cell tumor (PECOMA), uterine 489 Perivascular tumors cutaneous 582 uterine 489 Personality changes, neuroepithelial neoplasms 677 Peter MacCallum Cancer Centre, nasopharyngeal carcinoma 122, 123, 124 – 5 Peutz – Jeghers syndrome (PJS) 785 – 6 Carney complex and 784, 785 clinical features 785 colorectal cancer in children 751 endocrine cancers in children 781, 785 – 6 molecular genetics 785 – 6 sex cord tumor with annular tubules (SCTAT) 455, 462 small bowel adenocarcinomas 392 testicular tumors 72, 765, 785 Phenoxybenzamine, pheochromocytoma surgery and 159 Pheochromocytes 156 Pheochromocytoma 155 – 60 benign 155, 156 catecholamine production 143, 155, 157 – 8, 789 – 90 children/adolescents 787 – 91 clinical presentation 155 – 6, 787 – 8 emergency situations 788 definition 787 diagnosis 157 – 9 biochemical 157 – 8, 789 – 90, 790 dopamine-secreting tumors 790 histologic 156 – 7 imaging 158, 158 – 9, 790 – 1 in renal failure 790 screening, indications for 157 differential diagnosis 157 dopamine-secreting 790 etiology 155 extra-adrenal (paragangliomas) 155 familial 155, 783, 787 – 91 genetics 155, 789, 789 incidence 788 – 9 malignant 155 histologic diagnosis 156 – 7 metastases 157, 789 prognosis 157 rate/incidence 159, 789 misdiagnosis 157 molecular biology 155, 157, 789, 789 neuroectodermal origins 155 other cancer associations 159 pathology 156, 156 – 7 pregnancy and 159 – 60 staging 159 treatment 159 – 60 Photodynamic therapy (PDT) bile duct tumors 386
848 Photodynamic therapy (PDT) (cont.) pediatric basal cell carcinoma 812, 813 pleural mesothelioma 286 small bowel metastases 398 Phyllodes tumor of the breast 209 – 17, 219 benign 209, 211, 211 – 2, 215 biology 209 – 10 borderline tumors 210, 212, 215 – 6 clinical presentation 213 diagnosis 213 differential diagnosis 209, 212 epidemiology 209, 210 fibroadenomas and 209, 212 males 209, 213 malignant 209, 212, 212, 213, 214, 215 malignant transformation 210 metastases 209 – 10, 213, 215, 216 molecular biology 210 – 1, 214 pathology 211 – 3 biological markers 212 – 3 histologic 211, 211 – 2, 212, 216 macroscopic 211, 211 sarcomatous differentiation 212 prognosis 215 – 6 recurrence 210, 212, 213, 215 treatment 213 – 5 chemotherapy 214 hormonal/biologic therapy 214 metastatic disease 215 radiation therapy 214 recommendations 216 recurrent disease 215 surgical 213 – 4 tumor development model 211 Phyllodes tumors 209 breast see Phyllodes tumor of the breast prostate 48 treatment 51 Physaliphorous (bubble-bearing) cells, chordomas 615, 615 Pilocytic astrocytoma 685 – 6, 686 spinal cord 688 Pilomatrix carcinoma (pilomatricarcinoma) 577 Pilomyxoid astrocytoma 685 – 6 Pinealoma, primary germ cell tumors 649 Pineal tumors pediatric malignancy 799, 803 – 4, 804 primary germ cell tumors 649 clinical presentation 651, 651 Pineoblastoma, pediatric malignancy 803 – 4, 804 Pituitary adenoma, Carney complex and 784 “Pituitary sarcomas,” 627 – 8 PKC412 426 Placenta development 533 disease see Gestational trophoblastic diseases transplacental melanoma transmission 810 Placental alkaline phosphatase (PLAP) dysgerminoma 468 intracranial germ cell tumors 650 Placental-site trophoblastic tumor (PSTT) 532, 537 pathology 534 treatment 540 Planning target volume (PTV), nasopharyngeal carcinoma 123 Plaque radiation therapy, retinoblastoma 718 Plasma cell dyscrasias, uncommon presentations 569 – 74 Plasma cell leukemia 569 primary 549 Plasmapheresis, Waldenstr¨om macroglobulinemia 571 Plasminogen activator inhibitor type I, meningiomas 639 Platelet-derived growth factor (PDGF) dermatofibrosarcoma protuberans 579, 589, 592
SUBJECT INDEX gastric stromal tumors, treatment targets 361 Platelet-derived growth factor receptor (PDGFR) gastrointestinal stromal cell tumors 421, 421 imatinib mesylate 422 – 4, 424, 425 meningiomas 639 PKC412 426 sunitinib malate 425 VEGFR 426 Platinum based regimens 442 extra-ovarian primary peritoneal carcinomas 444 ovarian germ cell tumors 472 – 3 primary adenocarcinoma of fallopian tube 481, 481 pulmonary carcinoid tumors 310 resistance 473 urethral cancer 34 see also specific drugs/drug combinations Platinum, gemcitabine, and a taxane (GCT), urethral cancer 34 Pleomorphic adenocarcinoma, pancreatic 370 – 1 Pleomorphic carcinoma see Spindle cell (sarcomatoid) carcinoma Pleomorphic liposarcoma, oral cavity and adjacent structures 98, 98 Pleomorphic rhabdomyosarcoma 269 Pleomorphic xanthoastrocytoma (PXA) 686, 686 pediatric 799 – 800, 800, 801 Pleural effusions, pleural mesothelioma 282 Pleural mesothelioma 282 – 7 clinical presentation 282 evaluation/staging 282 – 3, 283, 287 management 283 – 7 chemotherapy 284 – 5, 285, 286 combined modality 285 – 6 intracavity therapy 286 – 7 novel targeted therapy 285 photodynamic therapy 286 – 7 radiation therapy 284 surgical 283 – 4, 284 prognosis/survival 283, 287 Pleurectomy, pleural mesothelioma 283, 284, 285 – 6 Pleuropulmonary blastoma (PPB), pediatric 732, 740, 740, 741, 742 Plexiform fibrous histiocytoma 580 – 1 Pneumonectomy extrapleural, pleural mesothelioma 283 – 4, 284, 286 large cell neuroendocrine carcinoma of the lung 302 – 3 primary pulmonary lymphoma 261 Polio vaccination, neuroepithelial neoplasms 675 Polyembryoma 469 Polymorphous low-grade (terminal duct) salivary gland carcinoma 94 presentation 92 – 3 treatment 95 Polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy, and skin changes (POEMS) syndrome 572 – 3 Polypeptide hormones, small cell undifferentiated carcinoma of prostate 41 Polyploid squamous cell carcinoma see Spindle cell (sarcomatoid) carcinoma Polypoid carcinoma, esophagus 339 Pontine gliomas, pediatric malignancy 687 Porocarcinoma (malignant eccrine poroma) 578 Poroma, malignant eccrine (porocarcinoma) 578 Positron emission tomography (PET)
adrenal tumors adrenocortical tumors 152 – 3, 153, 779, 780 “incidentalomas,” 160 pheochromocytoma 158, 158 – 9, 791 cancers of oral cavity and adjacent structures hematological cancers 96 salivary gland tumors 93 colon/rectum carcinoid tumors 403 – 4 esthesioneuroblastoma 137 – 8 FDG-PET see Fluorodeoxyglucose-positron emission tomography (FDG-PET) gastric lymphomas 362 gastrointestinal stromal cell tumors 422 in children 753 genitourinary tumors small cell undifferentiated carcinoma of prostate 44 urethral cancer 30 imatinib mesylate 423 Merkel cell carcinoma 597 neurological tumors meningioma 642 neuroepithelial neoplasms 681 pleomorphic xanthoastrocytoma 800 primary CNS lymphoma 660 non-Hodgkin’s lymphoma, breast 197 primary adenocarcinoma of fallopian tube 480 sunitinib malate 425 thoracic tumors adenoid cystic carcinoma of the lung 323 bronchioloalveolar carcinoma 316 pleural mesothelioma 282 primary melanoma of lung 295 pulmonary carcinoid tumors 309 thymomas/thymic carcinomas 247 Postoperative cerebellar mutism 701 Postoperative tumor production, meningeal sarcomas 628 Posttransplantation cancers see Transplantation-related malignancy Posttransplantation lymphoproliferative disease (PTLD) 10, 565 primary pulmonary lymphoma 258 small bowel lymphoma 396 Precocious puberty, adrenogenital syndrome 145 Precursor B cell lymphoblastic lymphoma 557 cutaneous lymphoma in children 813 Precursor T lymphoblastic leukemia 560 Precursor T lymphoblastic lymphoma 560 Prednisolone, thymomas/thymic carcinoma treatment 252 Preeclampsia, hydatidiform moles and 535 Pregnancy cervical tumors 19, 515, 516 – 7 granulosa cell tumors 459 molar see Hydatidiform mole ovarian cancer risk 447 ovarian germ cell tumors 471 pediatric thyroid cancer recurrence 787 pheochromocytoma and 159 – 60 renal angiomyolipoma and 9 transplacental melanoma transmission 810 trophoblastic diseases and see Gestational trophoblastic diseases Primary aldosteronism see Conn’s syndrome Primary central nervous system lymphoma (PCNSL) 657 – 66, 663 AIDS-related see under HIV/AIDS-associated lymphoma clinical presentation 659 diagnostic considerations 659 – 60 epidemiology 657 evaluation 659 imaging 660, 660 neurotoxicity 663
SUBJECT INDEX pathogenesis, immunocompromised patient 658 – 9 pathology 657 – 9 recurrent disease 663 symptoms/signs 659 treatment 660 – 3 chemotherapy 661 prognostic factors 660 radiation therapy 661 supportive care 660 surgery 661 Primary cutaneous anaplastic large cell lymphoma (PCALCL) 563 Primary effusion lymphoma 559 Primary intracranial germ cell tumors 649 – 56 classification 649, 650 clinical presentation 651, 651 diagnosis 651 – 3, 652, 652, 654 differential diagnosis 651 epidemiology 650 historical background 649 pathogenesis 649 – 50 pathology 650 – 1 prognosis 654 recommendations 654 relapse rates 653, 654 treatment chemotherapy 654 combined modality therapy 654 radiation therapy 653 – 4 surgery 653 types 649 Primary lung sarcoma see Primary pulmonary sarcoma (PPS) Primary pigmented nodular adrenocortical disease (PPNAD) 783, 784 Primary pulmonary lymphoma (PPL) 257 – 63 classification 257, 258 – 9 clinical presentation 043:259, 260 diagnosis 260 – 1 differential diagnosis 257 epidemiology 257 – 8 Hodgkin’s disease 257, 258 epidemiology 257 pathology/classification 258 symptoms/signs 259, 260 treatment/prognosis 262 imaging 259, 259 – 60, 260 immunosuppression and 258 non-Hodgkin’s lymphoma 257 clinical features 259, 259, 260 epidemiology 257 grading 258 – 9 MALT-type 258, 261 – 2 treatment/prognosis 261 – 2 pathology 258 – 9 staging 261, 261 treatment/prognosis 261 – 2 Primary pulmonary sarcoma (PPS) 264 – 78 clinical presentation 272 diagnostic evaluation 273 – 4 differential diagnosis 272, 272 epidemiology/etiology 264 fine needle aspiration (FNA) 274 histological subtypes 265 pathology 264 – 72 prognostic factors 274 treatment 274 – 5 see also individual tumors/locations Primary renal lymphoma (PRL) 10 Primary sclerosing cholangitis, bile duct tumors 385 Primitive neuroectodermal tumor (PNET) 695 – 704 adults 702 biology/epidemiology 695 – 6 clinical presentation 695 – 7 cutaneous 585
differential diagnosis 697 Ewing’s sarcoma vs. 89 meningeal rhabdomyosarcoma vs. 631 extraosseous Ewing’s sarcoma as 722 management 90 neuroepithelial neoplasms 678 oral cavity and adjacent structures 88, 89, 90 treatment 697 – 702 see also Ewing’s sarcoma; Medulloblastoma PRKAR1A gene Carney complex and 784 – 5 pediatric thyroid cancer and 786 Procarbazine CCNU, and vincristine (PCV) in glioblastoma 683 primary CNS lymphoma 662 Progesterone receptor adenoid cystic carcinoma of the breast 188 endometrial carcinomas 494 endometrioid adenocarcinoma 505 extra-ovarian primary peritoneal carcinomas 441 male breast cancer 203 meningioma 639, 644 Prolactinomas, MEN1 782 Proliferation markers adrenocortical tumors 150 dermatofibrosarcoma protuberans 589 meningioma 640, 644 pheochromocytoma 157 thymomas/thymic carcinoma 245 PROMACE-MOPP regimen, testicular non-Hodgkin’s lymphoma 70 Prostate cancer 38 – 65 antecedent adenocarcinoma 40, 40 breast metastases 203 carcinoid 45 – 6 estrogen therapy and male breast cancer 201, 202 hematologic malignancy involvement 57 pediatric malignancy 764 phyllodes tumors 209 primary lymphoma 57 – 9 clinical presentation 58 prognosis 58 recommendations 59 treatment 58 – 9 sarcomas 46 – 53 clinical characteristics 49 incidence 46 pathology 46 – 9 prognosis 50 recommendations 52 – 3 treatment 50 small cell undifferentiated (neuroendocrine) carcinoma 38 – 46 transitional cell carcinoma (TCC) 53 – 7 clinical features 55 conclusions/recommendations 56 – 7 diagnosis 55 incidence 53 – 4 management 56 pathogenesis/etiology 54 pathology 54 – 5 prognosis 55 – 6 Prostatectomy primary lymphoma of prostate 58 radical, small cell undifferentiated carcinoma of prostate 44 Prostate-specific antigen (PSA) 764 small cell undifferentiated carcinoma of prostate 44 Prostatic acid phosphatase (PAP) 764 Protein kinase A (PKA), Carney complex and 784 – 5 Proteomics, esthesioneuroblastoma 135 Proton irradiation, chordomas 621 – 2 Psammoma bodies
849 adenoid cystic carcinoma of the breast 190 – 1 metanephric tumors 12 Pseudoadenomatous basal cell carcinoma see Adenoid cystic carcinoma Pseudocapsule, chordomas 615 Pseudocarcinoma see Spindle cell (sarcomatoid) carcinoma Pseudocarcinosarcoma see Spindle cell (sarcomatoid) carcinoma Pseudolymphoma 257 Pseudomyxoma peritonei (PMP), appendix 411 Pseudopapillary pancreatic tumor, pediatric malignancy 754 – 5, 755, 755 management 755 – 6 Pseudopheochromocytoma 787 – 8 Pseudosarcoma breast 183, 183 larynx 104 see also Spindle cell (sarcomatoid) carcinoma Pseudostratified columnar cells 113 – 4 Psoralen and ultraviolet light (PUVA) therapy, cutaneous lymphoma in children 814 Psychiatric symptoms, craniopharyngiomas 707 PTEN gene/protein Cowden disease 786 male breast cancer 202 neuroepithelial neoplasms 676 Pth-like substance, small cell undifferentiated carcinoma of prostate 41 Pubarche, adrenocortical tumors and 778 Pulmonary adenoid cystic carcinoma see Adenoid cystic carcinoma of the lung Pulmonary adenomatosis see Bronchioloalveolar carcinoma Pulmonary and Mediastinal Pathology Registry of the Armed Forces Institute of Pathology, mucoepidermoid lung cancer 330 Pulmonary arterial sarcoma 270 Pulmonary arterial tree sarcoma 270 Pulmonary blastoma, differential diagnosis 273 Pulmonary carcinoid tumors 307 – 12 biology 307 biotherapy 310 central 308 chemotherapy 310 clinical features 308 diagnosis 308 – 9 epidemiology 307 immunohistochemistry 308 management 309 – 11 metastatic disease 309 – 10 local treatment 311 microscopically 308 pathology 307 – 8 surgery 309 Pulmonary chondromata 361 Pulmonary endodermal tumor resembling fetal lung (fetal pulmonary adenocarcinoma) 739 – 40, 740 Pulmonary fibrosis, bleomycin 473 Pulmonary function tests (PFTs), adenoid cystic carcinoma of the lung 323 Pulmonary hypertension, primary lung sarcomas 272 Pulmonary infiltrates, primary pulmonary lymphoma 259, 260 Pulmonary lymphoma, primary see Primary pulmonary lymphoma (PPL) Pulmonary melanoma see under Melanoma Pulmonary nodules, primary pulmonary lymphoma 259, 259, 260 Pulmonary sarcoma, primary see Primary pulmonary sarcoma (PPS)
850
SUBJECT INDEX
Pure red cell aplasia (PRCA), thymoma and 245, 246 Purine analogs, mantle cell leukemia 550 PUVA therapy, cutaneous lymphoma in children 814 Pyramidal tract dysfunction, chordomas 617 Quadrangular membrane
102
RAD001 426 Radial scars, tubular carcinoma and 230, 231 Radiation-induced cancers medulloblastoma/PNET 696 meningioma 640 sarcomas 49, 98, 485, 627 – 8, 634 see also Fibrosarcoma sebaceous carcinoma 577 see also Radiation therapy, sequelae/adverse effects Radiation portal design, glioblastoma 682 Radiation therapy breast cancers adenoid cystic carcinoma 191 carcinosarcoma 226 male breast cancer 205 metaplastic carcinoma 184 non-Hodgkin’s lymphoma 197 phyllodes tumor 214 tubular carcinoma 233 craniospinal limitations 700 – 1 medulloblastoma in adults 700, 701 medulloblastoma in children 699 primitive neuroectodermal tumor (PNET) 697 sequelae/adverse affects in children 654 cutaneous malignancy angiosarcoma 581 dermatofibrosarcoma protuberans 580, 591 Merkel cell carcinoma 584, 598 pediatric basal cell carcinoma 812 endocrine malignancy adrenocortical 154 – 5 parathyroid carcinoma 177 gastrointestinal malignancy esophageal adenoid cystic carcinoma 341 esophageal melanoma 346 – 7 gastrointestinal stromal cell tumors 345, 422 hepatoid adenocarcinoma 354 small bowel adenocarcinoma 393 genitourinary malignancy primary lymphoma of prostate 58 prostate sarcomas 51 small cell undifferentiated carcinoma of prostate 44 transitional cell carcinoma of prostate 56, 57 urethral cancer 33 see also renal tumors; testicular/ paratesticular tumors (below) gynecological malignancy borderline ovarian tumors 451, 452 cervical lymphoma 514 cervical melanoma 513 cervical small cell carcinoma 511 cervical verrucous carcinoma 503 dysgerminoma 474, 474 endometrial carcinoma 495 gestational trophoblastic neoplasia 539 gestational trophoblastic neoplasia (GTN) 540 primary adenocarcinoma of fallopian tube 480 stromal tumors of the ovary 457 – 8, 459, 462
uterine sarcoma 490, 491 head and neck cancer ameloblastoma 725 esthesioneuroblastoma 138 – 9, 140, 729 extraosseous Ewing’s sarcoma 723 see also laryngeal tumors; nasopharyngeal carcinoma; oral cavity (below) laryngeal tumors adenoid cystic carcinoma 107 chondrosarcomas 109 mucoepidermoid carcinoma 108 neuroendocrine tumors 110 spindle cell carcinoma 105 verrucous carcinoma 104 nasopharyngeal carcinoma 121 – 2 dose/fractionation 126 philosophy/technique 122 – 6 racial factors 121 recurrent/metastatic 128 – 9 neurological malignancy anaplastic oligodendroglioma 691 astrocytoma 684 atypical teratoid/rhabdoid tumors 806 chordomas 621 – 2 choroid plexus carcinoma 670 – 1, 802 craniopharyngiomas 709 glioblastoma 682, 684 gliomas 688, 689 hemangiopericytoma 646 Langerhans’ cell histiocytosis 613 medulloblastoma in adults 700 medulloblastoma in children 698 – 9 melanocytic lesions 807 meningeal sarcomas 634 meningioma 638, 642 – 3, 643, 644, 646 neurocutaneous melanosis 607 neuroepithelial neoplasms 675 oligodendroglioma, low-grade 690 optic pathway glioma 685 pilocytic/pilomyxoid astrocytoma 686 pineoblastoma 804 primary CNS lymphoma 661 primary intracranial germ cell tumors 653 – 4 primitive neuroectodermal tumor 702 primitive neuroectodermal tumor (PNET) 702 see also ophthalmic tumors (below) ophthalmic tumors ocular lymphoma 661 optic pathway glioma 685 retinoblastoma 718 sebaceous carcinoma of the eyelid 713 uveal melanoma 716 oral cavity and adjacent structures bone tumors 90 hematologic malignancy 96 melanoma 91 salivary tumors 94, 95 soft tissue sarcomas 99, 100 pediatric malignancy basal cell carcinoma 812 extraosseous Ewing’s sarcoma 723 medulloblastoma in children 698 – 9 sequelae/adverse effects 654 pregnancy and 517 renal tumors adult Wilms tumor 13 – 4 carcinoid 10 collecting duct (of Bellini) carcinomas (CDCs) 3 renal medullary carcinoma 4 sarcomas 11 sequelae/adverse effects cancer induction see Radiation-induced cancers craniospinal in children 654 effect on mucosa of sinonasal tract 128 radionecrosis 682
spontaneous abortions 517 testicular/paratesticular tumors adenocarcinoma of the rete testes 75 Leydig cell tumors 72 malignant mesothelioma of the tunica vaginalis 76 NHL 70 Non-Hodgkin’s lymphoma 69 – 70 rhabdomyosarcoma 79 – 80 Sertoli cell tumors 73 thoracic tumors adenoid cystic carcinoma of the lung 325 large cell neuroendocrine carcinoma of the lung 303 mediastinal tumors 733 mucoepidermoid lung cancer 332, 333 pleural mesothelioma 284, 285 – 6 primary lung sarcomas 275 primary pulmonary lymphoma 262 pulmonary carcinoid tumors metastasis 311 thymomas/thymic carcinomas 248, 249, 250 – 2 see also Adjuvant therapy; individual disorders; specific techniques Radiation Therapy Oncology Group (RTOG) 661 anal carcinoma 407 Radioactive iodine treatment, thyroid carcinoma 167, 169, 787 Radiofrequency ablation (RFA) gastric endocrine cell proliferations 359 islet cell tumors 376 pancreatic carcinoid 377 pulmonary carcinoid tumor metastases 311 Radiographic studies adenoid cystic carcinoma of the lung 323 breast carcinosarcoma 223, 224 bronchioloalveolar carcinoma 316 cancers of oral cavity and adjacent structures 87, 89 esophageal 346 gastrointestinal stromal cell tumors 421 lung mucoepidermoid carcinoma 331 lung sarcomas, primary 265 nasopharyngeal carcinoma 119 non-Hodgkin’s lymphoma, breast 196 – 7 pancreatic cystadenocarcinoma 369 primary lung sarcomas 274 small bowel cancer 391, 393, 393 tubular carcinoma 231 see also specific modalities Radioimmunotherapy hematologic malignancies 96 see also specific agents Radioiodinated metaiodobenzylguanidine (131 I-MIBG), esthesioneuroblastoma 138 Radionecrosis 682 Radionuclide scans pheochromocytoma 158 Zollinger – Ellison syndrome 374 Radioresistance, chordomas 621 Radiosurgery chordomas 622 craniopharyngiomas 709 meningioma 643, 644 Radiotherapeutics, pheochromocytoma 159 Radiotherapy see Radiation therapy “Ranula” (mucous cyst) of salivary gland 91, 93 ras oncogene 42 adrenocortical tumors 149 breast carcinosarcoma 227 meningioma 641 pediatric thyroid cancer 786 Rathke’s pouch, craniopharyngiomas 705 RB (retinoblastoma) gene see Retinoblastoma (RB) gene/protein
SUBJECT INDEX Reactive etiologies, meningeal sarcomas 628 Reactive lymphadenopathy, osteosclerotic myeloma 572 Rectal bleeding, anal carcinoma 406 Rectal tumors carcinoid tumors 401, 751 clinical presentation 402 – 3 management 403 metastatic disease 403 – 4 pathology 401 – 2 prognosis 404 malignant melanoma 407 – 8 sarcoma 404 – 6 small cell carcinomas 433 Renal cell carcinoma (RCC) classification/staging 2, 760, 761 incidence 1 molecular diagnosis 1 – 2 pediatric malignancy 760, 761 secondary thyroid malignancies 171 see also specific types Renal cell carcinoma marker (RCC Ma) 1 Renal epithelial tumors of uncertain malignant potential 2 Renal failure chemotherapy-induced 598 pheochromocytoma diagnosis 790 Renal hemangiopericytoma (HPC), pediatric malignancy 763 Renal lymphoma, primary childhood 762 Renal medullary carcinoma 4 pediatric malignancy 763 Renal pelvis tumors, pediatric malignancy 763 Renal tumors 1 – 17 classification 1, 2 incidence 1 metastatic 763 molecular diagnosis 1 – 2 pediatric malignancy 760 – 3 see also individual types Renin-secreting juxtaglomerular cell tumor 762, 763 Reproductive system tumors pediatric 764 – 71 see also specific types/locations Respiratory insufficiency, complete hydatidiform moles and 534 Respiratory tree, pediatric sarcoma 739 Rete testes, adenocarcinoma 74 – 5 Retiform hemangioendothelioma 582 Retina anatomy 717 cancers 713, 717 – 8 Retinoblastoma 717 – 8 screening 805 susceptibility gene 676 trilateral 804 – 5 Retinoblastoma (RB) gene/protein 676, 717 chordomas 616 large cell neuroendocrine cancer (LNEC) of the lung 299 Merkel cell carcinoma 596 Retinocytoma 717 RET proto-oncogene adrenocortical tumors 149 islet cell tumors 374 MEN2 and 374, 782 – 3 pediatric thyroid cancer 786 pheochromocytoma 155 Retrograde urethrography, urethral cancer 30 Retroperitoneal lymph node dissection (RPLND) adenocarcinoma of the rete testes 75 Leydig cell tumors 72 malignant mesothelioma of the tunica vaginalis 76 paratesticular rhabdomyosarcoma 79, 81 Sertoli cell tumors 73, 74
Retropharyngeal lymph node metastasis 118 – 9 Revised European American Classification of the Lymphoid Neoplasms (REAL), lymphomas 555 Rhabdoid cells 805 Rhabdoid meningioma 640 Rhabdoid tumors CNS 805 – 6 kidney (RTK) 761, 805 Rhabdomyoma, cardiac 736 Rhabdomyosarcoma (RMS) adult vs. pediatric malignancy 100 classification 78, 78 TNM 79, 79 clitoris 767 cutaneous 582 – 3, 815 diagnosis/differential diagnosis 78, 99 esophageal 345 etiology 77 incidence 77 mediastinal 735 – 6 meningeal 631, 632 chemotherapy 635 prognosis 635 metastases 631 oral cavity and adjacent structures presentation 98 – 9 prognosis 97 treatment 100 ovarian 100, 770 paratesticular see Paratesticular rhabdomyosarcoma pediatric malignancy 77 adults vs. 100 chest wall 744 clitoris/vulva 767 cutaneous 582, 815 lung parenchyma 739, 742 oral cavity and adjacent structures 98 – 9, 99 ovarian 99, 770 penile 764 penile 764 prostate 47, 51 pulmonary 264, 269 – 70, 739, 742 renal 11, 761 uterine 488 – 9 vulva 767 Rituximab (anti-CD20 monoclonal antibody) gastric lymphomas 363 hairy cell leukemia 549 mantle cell leukemia 550 Waldenstr¨om macroglobulinemia 571 Robson’s classification, renal cell carcinoma in children 760 Round cell tumors oral cavity and adjacent structures 89, 90 see also specific types Rous sarcoma virus, meningeal sarcomas 627 Royal Marsden group, mantle cell leukemia 550 Sacral chordomas 614 clinical features 617, 617 surgical resection 621 Sacral nerve root compression, chordoma 617 Salivary duct carcinoma 92 Salivary gland tumors 91 – 5 acinic cell tumors 92 adenocarcinoma (NOS) 92, 93 adenoid cystic carcinoma 91, 94, 107, 188, 321, 329 behavior/treatment 94 – 5, 107 – 8, 325 biology/epidemiology 106 pathology 106 presentation 92
851 TNM staging 92 anatomy 91 basal cell carcinoma 92 benign lesions 91, 92 biopsy 91, 93 carcinoma ex pleomorphic carcinoma 92 carcinomas 91, 92 cystic 91, 95 diagnosis 93 ductal carcinoma 92 laryngeal tumors and 106 – 8, 107 malignant 91 – 3 high-grade cancers 91 – 2 low-grade cancers 92, 92 – 3 metastases 92, 93 mixed tumors 91, 92 mucoepidermoid carcinoma 91, 93, 107, 329 behavior/treatment 94 – 5, 108, 332, 332 biology/epidemiology 106 pathology 106 – 7, 330 presentation 92 myoepithelial carcinoma 92 polymorphous low-grade (terminal duct) carcinoma 92 – 3, 94, 95 presentation 91 – 3 prognosis 92, 93 squamous cell carcinoma 92, 92 staging 92, 93 treatment 93 – 5 undifferentiated carcinoma 92 Salpingo-oophorectomy bilateral with hysterectomy see Total abdominal hysterectomy with bilateral salpingo-oophorectomy (TAH/BSO) dysgerminoma 474 unilateral, stromal tumors of the ovary 457, 463 see also Fertility preservation ‘Salt-and-pepper’ chromatin, medullary thyroid carcinoma (MTC) 166, 166 Salvage therapy, anal carcinoma 407 Sandostatin see Octreotide acetate (Sandostatin ) Sarcoidosis, solitary urethral 28 Sarcoma 169 – 70 bile duct 385 bladder 25 cervical 511 – 3, 767 colorectal 404 – 6 cutaneous 579 – 80, 581, 582 – 3 definition 626 fallopian tube 481 – 2 gallbladder tumors 385 management 92, 99 – 100 mediastinal 735 – 6 meningeal see Meningeal sarcomas metastatic 273 oral cavity and adjacent structures bone 88, 89, 89, 90 hematologic 96, 97 soft tissue 97, 97 – 100, 98, 99 ovarian 770 pancreatic 373 pediatric malignancy alveolar soft part 767, 768, 768 cervix 767 lung parenchyma 742 ovarian 770 renal 762 respiratory tree 739 uterus 768 vagina 768, 768 prostate 46 – 53 pulmonary lung parenchyma 742 primary see Primary pulmonary sarcoma (PPS) respiratory tree 739
852 Sarcoma (cont.) radiation-induced 49, 98, 485, 627 – 8, 634 see also Fibrosarcoma renal 10 – 1, 762 small bowel 396 – 8 synovial 737 uterine see Uterine sarcoma vaginal 528, 768, 768 vulvar 525 – 6 see also specific types/locations Sarcoma botryoides (SB) cervical 511, 512 pediatric malignancy 767 – 8 Sarcomatoid carcinoma (pleomorphic adenocarcinoma) 503 – 4 bladder 18 – 9 pancreas 370 – 1 thymus 244 Sarcomatoid mesotheliomas 265 Sarcomatoid renal cell carcinoma, lung metastases 273 Sarcomatosis, meningeal 632 – 3, 633 Sarcomatous overgrowth, phyllodes tumor of the breast 212 Sarcomatous transformation, dermatofibrosarcoma protuberans 589 – 90 Scatter factor (hepatocyte growth factor; HGF; SF), neurocutaneous melanosis and 605 Schiller – Duvall bodies, yolk sac tumor 468 Schwannoma malignant clitoris/vulva, pediatric malignancy 767 cutaneous, pediatric malignancy 817 mediastinal 735 Sciatica, sacral chondroma vs. 617 Sclerosing/syringomatous sweat gland carcinoma (microcystic adnexal carcinoma) 579 Screening, retinoblastoma 805 Scrotum malignant mesothelioma of the tunica vaginalis (MMTV) 76 Paget’s disease 579 rhabdomyosarcoma 78 Sebaceous carcinoma 577 – 8 of the eyelid 712 – 3, 713 Secondary malignancies breast 347, 347 non-Hodgkin’s lymphoma (NHL) 194 hairy cell leukemia 548 heart 737 late sequelae of chemotherapy 473 thyroid 171 Second-look laparotomy, ovarian germ cell tumors 472 Second malignant neoplasms (SMNs) medulloblastoma 702 retinoblastoma 718 SEER see Surveillance Epidemiology and End Results (SEER) program Segmentectomy, primary pulmonary lymphoma 261 Seizures desmoplastic glioma (gliofibroma) 801 dysembryoplastic neuroepithelial tumors 803 pleomorphic xanthoastrocytoma 800 primary CNS lymphoma 659 Selective venous sampling, adrenocortical tumors 153 Sellar craniopharyngiomas 705 Seminoma see Germinoma (seminoma) Sentinel lymph node biopsy/dissection apocrine carcinoma 578 breast carcinosarcoma 225 epithelioid sarcoma 580 history 597 – 8 male breast cancer 204 – 5
SUBJECT INDEX Merkel cell carcinoma 584, 597 – 8 Serotonin metabolite 5HIAA, appendiceal carcinoid tumors 414 Serous borderline ovarian tumors (SBTs) 447 – 8, 448 pediatric malignancy 769 Serous cystadenomas, pancreatic 368 Serous papillary adenocarcinoma 507 – 8 Serpentines, mesothelioma and 279 Sertoli cells 70 – 1 Sertoli cell tumors ovarian stromal tumors 455, 456, 461 – 2 pediatric malignancy 765 Carney complex and 784 Peutz-Jeghers syndrome and 785 testicular/paratesticular tumors 72 – 4 clinical features 73 epidemiology 72 etiology 72 investigations 73 large cell calcifying 74 pathology 72 – 3, 73 treatment 73 – 4 Sertoli – Leydig tumors see Sertoli – stromal cell tumors Sertoli – stromal cell tumors classification 461 ovarian 455, 456, 461 – 2 Serum protein electrophoresis (PEP) syndrome 572 – 3 Sex cord tumors see Gonadal (sex cord) stromal tumors Sex cord tumor with annular tubules (SCTAT) 455, 462 – 3 Sex hormone receptors estrogen see Estrogen receptor meningiomas 639 progesterone see Progesterone receptor Sex hormones adrenogenital syndrome 145 – 7 biochemical analysis 151 – 2 steroidogenesis 150 – 1 see also individual hormones Sex ratio, chordomas 614 Sexual abnormalities, adrenogenital syndrome 145 Sexually transmitted disease, anal carcinoma 406 S´ezary syndrome (SS) 563 Sickle cell disease/trait, renal medullary carcinoma 4, 763 Signet-ring cell adenocarcinoma 506 – 7 Silicone breast implants, cancer and 196, 219 Simpson’s scale of surgical resection, meningioma 642, 642 Single-photon emission computerized tomography (SPECT) meningioma 642 primary CNS lymphoma 660 Sinonasal tract, irradiation effects 128 Sinonasal undifferentiated carcinomas (SNUC) 133, 727, 727 – 8 Sipple’s syndrome (MEN type IIa), islet cell tumors 374 Skene’s glands, female urethral adenocarcinoma 29 Skin lesions acinar cell carcinoma syndrome 370 atypical fibroxanthoma 581 Carney complex and 783, 784 congenital nevi see Congenital nevi cutaneous angiosarcoma 581 dermatofibrosarcoma protuberans 580, 590, 590, 815, 816 epithelioid hemangioendothelioma 581 leiomyosarcoma 816 leukemia cutis in children 814 – 5 lymphoma in children 814, 814 malignant histiocytosis 813
malignant peripheral nerve sheath tumors 817 neuroblastoma 816 pediatric malignancies 810 – 8 superficial leiomyosarcoma 582 trichilemmal carcinoma 577 see also specific lesions Skull base, chordomas clinical features 617, 617 surgical resection 620 – 1 Skull osteolysis, meningeal sarcomas 628 “Small blue cells,” Merkel cell carcinoma (MCC) 595, 596, 599 Small bowel 391 cancer see Small bowel cancer dysplasia 392 Small bowel cancer 391 – 400 adenocarcinoma 392 – 4, 393 associated conditions 392, 396 biology/pathology adenocarcinoma 392 – 3 carcinoid 394 lymphoma 395 – 6 sarcoma 396 – 7 carcinoid tumors 394 – 5, 395 carcinoid syndrome and 391, 392 differential diagnosis 394 clinical presentation 391 diagnosis 391 – 2, 393, 393, 394, 395 incidence 391 lymphoma 391, 395 – 6 management adenocarcinoma 393 – 4 carcinoid tumors 394 – 5 lymphoma 396 metastatic disease 398 sarcoma 397 – 8 metastatic 398 sarcoma 396 – 8 GIST 396 – 7, 397 non-GIST 396, 397 – 8 see also specific tumor types Small cell-based chemotherapy, pulmonary carcinoid tumors 310 Small cell carcinoma 302, 321 differential diagnosis 273 large cell neuroendocrine carcinoma (LNEC) vs. 302 extrapulmonary bile duct 385 bladder 19 – 21 cervix 510 – 1 esophageal, poorly differentiated 343 ovary 343, 343, 470 pancreas 372 skin see Merkel cell carcinoma (MCC) gastrointestinal see Gastrointestinal small cell carcinoma neuroendocrine differentiation 109 differential diagnosis 596 skin see Merkel cell carcinoma (MCC) see also specific types Small cell undifferentiated (neuroendocrine) carcinoma of prostate (SCUCP) 38 – 46 clinical presentation 43 – 4 conclusions/recommendations 46 cytology 39 – 40 diagnosis 44 histogenesis 43 pathology 38 – 43, 39 clinical application of tumor markers 41 – 2 cytogenetic abnormalities 42 DNA content 42 expression of tumor markers 41 oncogenes 42 – 3 prognosis 46 treatment 44
SUBJECT INDEX SMARCB1 gene, medulloblastoma in children 698 Smoking bronchioloalveolar carcinoma 313 large cell neuroendocrine cancer (LNEC) of the lung 298 lung cancer and 321 lung mucoepidermoid tumors 330, 331 vulvar squamous cell carcinoma 521 see also Lung cancer; Tobacco Smooth muscle cells, meningeal leiomyosarcoma 632 Soft tissue tumors cutaneous 579 – 84 fat cell (liposarcoma) 583 fibrohistiocytic 580 – 1 fibrous/myofibroblastic 579 – 80 muscular 582 – 3 neuroectodermal see Merkel cell carcinoma (MCC) perivascular 582 vascular 581 – 2 oral cavity and adjacent structures 97 – 100 see also Sarcoma; specific types/locations Solitary fibrous tumor (SFT) differential diagnosis 273 lung 267 – 8, 268 Solitary plasmacytomas 572 Somatostatin analogs octreotide see Octreotide acetate (Sandostatin ) prostate carcinoid 46 carcinoid syndrome diarrhea 377, 395 colon/rectum carcinoid tumors 403 gastric endocrine cell proliferations 359 islet cell tumors 376, 756, 756 receptors, meningiomas 639 Somatostatinoma 375 Somatostatin receptors (SSRs), pulmonary carcinoid tumors 309 Sonic Hedgehog-patched (PTCH) pathway, medulloblastoma/primitive neuroectodermal tumor (PNET) 696 Southwest Oncology Group (SWOG) 684 Sphenoid wing meningioma 641 en plaque growth pattern 638 Spinal cord compression, chordomas 618 gliomas 688 medulloblastoma/PNET 695 – 7 Spindle cell (sarcomatoid) carcinoma 265 breast (fibromatosis-like) 181, 223, 224 pathology 182, 182 prognosis 184 WHO classification 181 laryngeal tumors 104 – 5, 105 nomenclature/definitions 104, 181, 218, 222 see also Carcinosarcoma Spindle cell epithelial tumors of thymiclike epithelium (SETTLE) 240 Spindle cells atypical, chordomas 615 cutaneous fibrosarcoma in children 815 dermatofibrosarcoma protuberans (DFSP) 589, 815 – 6 morphology, gastrointestinal stromal cell tumors 419 mucinous carcinomas (renal) 5 squamous cell carcinoma of the conjunctiva 714 Spindle cell squamous carcinoma 339 Spindle cell thymomas 240, 241, 242 Spiradenoma, malignant eccrine 578 – 9 Spitz nevus, melanoma vs. 811 Splenectomy chronic neutrophilic leukemia 544 T-large granular lymphocyte leukemia 551 Waldenstr¨om macroglobulinemia 571
Splenic lymphoma with villous lymphocytes (SLVL) 557 – 8 Splenic marginal zone lymphoma 557 – 8 Spontaneous abortions, radiation therapy 517 Sputum collection, lung sarcomas, primary 265 Squamous cell carcinoma (SCC) 170 basaloid see Basaloid squamous cell carcinoma (BSCC) bile duct tumors 386 bladder 23 breast tumors, WHO classification 181 cervical 502 – 4 conjunctival 713 – 5, 714 cutaneous pediatric malignancy 812 – 3 predisposing factors 812, 812 endometrial 495 – 6 esophageal 341 – 2 gallbladder tumors 384 gastric 354 – 5 keratinizing, thymic carcinoma 242 – 3, 243 laryngeal 102, 105, 106 lymphoepithelioma-like, thymic carcinoma 243, 243, 244 mucin-secreting component 341 nasopharyngeal 116, 116 nonkeratinizing, thymic carcinoma 243 oral cavity 87 salivary glands 92, 92 ovarian 467, 771 pediatric malignancy cutaneous 812 – 3 ovarian 771 small bowel 392 spindle cell variant see Spindle cell (sarcomatoid) carcinoma urethral 28, 29 vaginal 526 – 8 vulvar 521 – 4 Squamous cell carcinoma related oncogene (SCCRO) 314 Squamous cells 114 Squamous metaplasia 294 Staging systems see Tumor classification/staging systems Standardized uptake volume (SUV), pleural mesothelioma 282 Steatorrhea, somatostatinoma 375 Stereotactic radiosurgery (SRS) craniopharyngiomas 709 meningioma 643, 644 Sterility, late sequelae of chemotherapy 473 Steroid (lipid) cell tumors, ovarian 455, 456, 463 Steroid drugs see Corticosteroids Steroid hormones, adrenocortical tumors and steroidogenesis 150 – 1 Steroidogenesis, adrenocortical tumors and 150 – 1 Steroidogenic factor-1 (SF-1), adrenocortical tumors 151 pediatric malignancy 775 Stewart – Treves syndrome, cutaneous angiosarcoma 581 STK11/LKB1 gene, Peutz-Jeghers syndrome 786 Stomach cancers 352 – 66 adenosquamous carcinoma 354 – 5 endocrine cell proliferations 357 – 60 gastric lymphomas 361 – 3 hepatoid adenocarcinoma 353 – 4 oncocytic gastric carcinoma 356 – 7 parietal cell carcinoma 356 – 7 primary germ tumors 355 – 6 squamous cell carcinoma 354 – 5 stromal tumors 360 – 1 undifferentiated carcinoma with lymphoid stroma 353, 353
853 see also entries beginning gastro-/gastric; individual tumors Streptozotocin adrenocortical tumor treatment 154 islet cell tumors 377 pulmonary carcinoid tumors 310 Stridor, adenoid cystic carcinoma of the lung 323 Stromal luteoma, ovarian 463 Stromal tumors 360 – 1 Carney triad 360 – 1 clinical presentation 360 gastric, prognosis 361 metanephric 11, 12 ovarian see Stromal tumors of the ovary pathology 360 phyllodes tumors 210 treatment 361 uterine 487, 487 – 8, 488, 489 see also specific types/locations Stromal tumors of the ovary 455 – 66 anatomy/pathology 455 biology 455 classification 455, 456 clinical presentation/diagnosis 456 endometrial type ovarian stromal sarcoma, children 770 endometrium and 456 epidemiology 455 general treatment guidelines 456 – 8 fertility preservation 456, 459, 460, 461, 462, 464 inadequate staging 457 non-surgical/mulitmodal options 457 – 8, 458 surgery 456 – 7 granulosa stromal cell tumors 456, 459 – 61 gynandroblastoma 463 historical background 455 imaging 456 prognosis 463 – 4 Sertoli-stromal cell tumors (androblastomas) 456, 461 – 2 Sertoli – stromal cell tumors (androblastomas) 455 sex cord tumor with annular tubules (SCTAT) 455, 462 – 3 staging 457 steroid (lipid) cell tumors 455, 456, 463 see also specific tumors Stromatosis, ovarian, pediatric malignancy 770 SU11248 see Sunitinib malate (SU11248, Sutent) Subarachnoid space leptomeningeal melanoma dissemination 608 metastases, medulloblastoma/PNET 695, 697 Subcutaneous nevi (SN), pediatric basal cell carcinoma 812 Subcutaneous panniculitis-like T cell lymphoma (SPLTL) 562, 563 Subglottis 102 Succinate dehydrogenase (SDH), pheochromocytoma 155, 788, 789 Sunitinib malate (SU11248, Sutent) 424, 425, 427 colon/rectum carcinoid tumors 404 gastrointestinal stromal cell tumors 42503410 small bowel 397 Superficial perineal leiomyosarcoma, pediatric malignancy 767 Superior orbital fissure syndrome, meningioma associated 641 Supraglottis 102 Suprasellar tumors craniopharyngiomas 705
854 Suprasellar tumors (cont.) primary germ cell tumors, presentation 651, 651 Suramin, thymomas/thymic carcinomas 250, 251 Surgical debulking see Cytoreduction Surgical mortality, glioblastoma 682 Surgical therapy see individual techniques; specific conditions Surveillance Epidemiology and End Results (SEER) program 313 adrenocortical tumors 775 appendiceal tumors 410 carcinoid 414 asbestos-related mesothelioma 280 borderline ovarian tumors 447, 450 breast carcinosarcoma 218, 220 – 1, 223, 224 childhood cancer adrenocortical tumors 775 brain tumors 798 colorectal cancer 753 colorectal cancer 751 carcinoid tumors 401 in pediatric malignancy 753 male breast cancer 202, 203 Merkel cell carcinoma 594 stromal tumors of the ovary 455 uterine sarcoma 485 SV40 virus medulloblastoma/PNET 696 mesothelioma etiology 281 neuroepithelial neoplasms 675 Sweat gland tumors 578, 579 Swedish Cancer Registry, parathyroid carcinoma 175 Swyer – James syndrome, adenoid cystic carcinoma vs. 323 Syed – Neblett template, female urethral cancer 33 Synchronous tumors 358 Syncytiotrophoblast 533 Syncytiotrophoblast cells 469 dysgerminoma 468 Syncytiotrophoblastic giant cells, choriocarcinoma 469 Synovial sarcoma 268 – 9, 269, 737 cutaneous involvement, children 817 esophageal 345 pediatric, lung parenchyma 743 renal involvement 11 Syringoid eccrine carcinoma (eccrine epithelioma) 579 Syringoma, malignant (microcystic adnexal carcinoma) 579 Systemic therapy see Biologic agents; Chemotherapy; specific drugs SYT-SSX fusion gene, synovial sarcoma 269 Tactile hair disc of Pinkus 594 Takatsuki syndrome 572 – 3 Tamoxifen breast carcinosarcoma 225 – 6 male breast cancer treatment 205 metaplastic breast carcinoma 181, 184 M¨ullerian adenosarcoma, cervical 512 pleural mesothelioma 286 recurrent meningioma 644 Targeted therapy adenoid cystic carcinoma of the lung 325 breast carcinosarcoma 226 – 7 bronchioloalveolar carcinoma 318 gastric stromal tumors 361 nasopharyngeal carcinoma 129 pleural mesothelioma 285 Taxanes, nasopharyngeal carcinoma 127 T cell leukemia 550 – 2 T cell lymphoblastic lymphoma (T-LBL) 557 T cell lymphomas 560 – 2
SUBJECT INDEX cutaneous lymphoma in children 814 mature 561 – 2 oral cavity and adjacent structures 95 precursor 560 primary renal of childhood 762 small bowel 396 see also specific types T cells (lymphocytes) EBV-specific in nasopharyngeal carcinoma 129 primary CNS lymphoma 658 thymus and 237 T(1;22)(p13;p13) chromosomal translocation 547 Technetium-99m-ethylcysteinate dimer (99m Tc-ECD), esthesioneuroblastoma 138 Teeth bone tumors 88 development, ameloblastoma 723 oral cavity lymphoma 96 see also Dentists/dental examination Tegafur, nasopharyngeal carcinoma 126 Telomerase, endocrine pancreatic tumors 373 Temozolomide glioblastoma 682, 683 recurrent meningioma 644 spinal cord gliomas 689 Temperature, male breast cancer association 202 Tenascin-C, breast carcinosarcoma 227 Teratoma cystic 467 gastric 356 growing teratoma syndrome 653 immature mediastinal tumors 734, 735 ovarian 470, 470 malignant 170 mediastinal 734, 735 ovarian 467, 470, 470 pericardium 737 primary intracranial 649 epidemiology 650 growing teratoma syndrome 653 pathological classification 650 surgery 653 tumor markers 652 Terminal duct carcinoma see Polymorphous low-grade (terminal duct) salivary gland carcinoma Testicle(s) atrophy, Leydig cell tumors 71 dysfunction, male breast cancer association 202 enlargement/swelling non-Hodgkin’s lymphoma 69 Sertoli cell tumors 73 pain, non-Hodgkin’s lymphoma 69 stromal cells 70 – 1 tumors see Testicular tumors Testicular tumor of the adrenogenital syndrome (TTAGS), children 766 Testicular tumors 66 – 85 children (prepubertal) 764 – 6 Carney complex and 74, 784 classification 764 Peutz – Jeghers syndrome and 72, 765, 785 staging 765 embryology 66, 67 incidence 66 metastatic 765 see also specific types Testosterone, adrenogenital syndrome 146 TFE3 gene/protein, renal cancers 1, 4 – 5 Thalidomide neurocutaneous melanosis 607 primary plasma cell leukemia 549 uterine sarcoma treatment 493
Thecomas, ovarian 461 Thiotepa, TCC in children 763 Third National Cancer Survey, appendiceal carcinoid tumors 414 Thomsen – Friedenreich antigen, sebaceous gland carcinoma 577 Thoracic tumors lung see Lung cancer pediatric see under Pediatric malignancy vascular lesions 270 see also specific types/locations Thoracotomy, thymomas/thymic carcinomas 248 Thrombocytosis, pleural mesothelioma and 287 Thrombosis/thromboembolism nonbacterial endocarditis, acinar cell carcinoma 370 osteosclerotic myeloma 572 primary lung sarcomas 272 Thymic carcinoma 237 – 56 ancillary studies 244 biology 238 biopsy 247 classification/terminology 237, 238 clinical features 245 – 6 diagnosis 247 – 8 differential diagnosis 244 metastases 246, 248 molecular markers 244 – 5 pathology 242 – 4 prognosis 253, 253 staging 237, 238, 246, 247 – 8 subtypes 242 basaloid squamous 242, 242, 243 clear cell 244, 244 dedifferentiated 244 keratinizing squamous 242 – 3, 243 lymphoepithelioma-like squamous 243, 243, 244 mucoepidermoid 243 – 4 nonkeratinizing squamous 243 sarcomatoid 244 undifferentiated 244 transformation from thymoma 238 treatment 249 algorithm 252 chemotherapy 249 – 52 radiation therapy 249 recommendations 252 – 3 surgical 249 well differentiated (WDTC) 238, 241 Thymocytes 237 – 8, 239 Thymolipomas 237 Thymoma 237 – 56 associated disorders 245, 246, 246 myasthenia gravis 245, 246, 253 biology 238 – 9 biopsy 247 classification/terminology 237, 238 Bernatz classification 241 histologic 241, 241 – 2 MMH classification 241 – 2 WHO classification 237, 242 clinical features 245 – 6 cortical 240, 240 diagnosis 247 – 8 differential diagnosis 241, 273 encapsulated (benign) 237, 240, 240, 245 treatment 248 epidemiology 239 etiology 239, 243 functionality 238 invasive (malignant) 237, 245 mediastinal tumors 733, 734 metastases 245, 249 molecular markers 244 – 5 pathology 240 – 5 gross 240, 240 histologic subgroups 241, 241 – 2
SUBJECT INDEX microscopic 240, 240 – 1, 241 prognosis 253, 253 relapse/recurrence 248, 249 staging 237, 238, 247 – 8 transformation to carcinoma 238 treatment 248 – 9 algorithm 252 chemotherapy 249 – 52, 251 medical/biological 252 radiation therapy 248, 249 recommendations 252 – 3 resection (complete, incomplete) 248 – 9 type A (medullary; spindle cell) 240, 241 type AB (mixed) 240, 242 type B (bioreactive) 242 type C see Thymic carcinoma Thymus dysplasia 240 ectopic tissue 238 epithelial variations 239 – 40 hyperplasia 240 lipomas 237 nasopharyngeal carcinoma and 118 neonates 238 nonneoplastic histopathology 239, 239 – 40 postpubertal atrophy 239, 239 structure/function 237 – 8 tumors 733 carcinoid 237 carcinoma see Thymic carcinoma epithelial 237 – 56 staging 237, 238 thymoma see Thymoma Thyroglobulin measurement, H¨urthle cell carcinoma 168 Thyrohyoid membrane 102, 103 Thyroid cancer 165 – 73 anaplastic carcinoma 168 – 9 Carney complex and 784, 786 classification 165 Cowden disease and 786 differentiated carcinoma 786 familial 165 Gardner syndrome 786 lymphoma 169 medullary carcinoma see Medullary thyroid carcinoma (MTC) MEN2 and 783 neuroendocrine laryngeal tumors vs. 110 papillary carcinomas 786 pediatric 786 – 7 Peutz – Jeghers syndrome and 785 secondary malignancies 171 SETTLE 240 see also individual types Thyroid cartilage 102, 103 tumors 109 Thyroid disease cancer see Thyroid cancer complete hydatidiform moles 534 hyperthyroidism 240, 534 hypothyroidism 702 nodules pediatric, cancer and 786 – 7 Peutz – Jeghers syndrome and 785 thymic hyperplasia and 240 Thyroidectomy anaplastic thyroid carcinoma 168 medullary thyroid carcinoma 166 multiple endocrine neoplasia (MEN) 166 pediatric thyroid cancer 787 secondary thyroid malignancies 171 total, lobectomy vs. H¨urthle cell carcinoma 167 – 8 Thyroid lymphoma 169 Thyroid storm, complete hydatidiform moles 534 Thyroid transcription factor (TTF-1), pulmonary carcinoid tumors 308
T-large granular lymphocyte leukemia (T-LGL) 550, 550 – 1 TNM (tumor, nodes, metastasis) staging system 307 – 8 adenoid cystic carcinoma, salivary glands 92 esthesioneuroblastoma 134, 729 lung cancers 321 primary sarcomas 266 male breast cancer 203 – 4 Merkel cell carcinoma 584 nasopharyngeal carcinoma 120, 120 parathyroid carcinoma 175 rhabdomyosarcoma 79, 79 small cell carcinomas of the gastrointestinal tract 431 thymomas/thymic carcinomas 238, 247 – 8 Tobacco chewing, verrucous carcinoma 337 esophageal small cell carcinomas 432 medulloblastoma/primitive neuroectodermal tumor (PNET) 696 nasopharyngeal carcinoma 115 see also Smoking Tongue, sarcoma 97 Tooth see Teeth Topical chemotherapy sebaceous carcinoma of the eyelid 713 squamous cell carcinoma of the conjunctiva 714, 715 Total abdominal hysterectomy with bilateral salpingo-oophorectomy (TAH/BSO) borderline ovarian tumors 450, 451, 452 stromal tumors of the ovary 457, 460, 462 uterine sarcomas 489 – 90 Tracheal ring 103 Tracheal tumors mucoepidermoid 329 pediatric 737 – 9 Tracheobronchial non-Hodgkin’s lymphoma (NHL) 258 Transabdominal ultrasound, gallbladder tumors 383 Transarterial chemoembolization (TACE), gastric endocrine cell proliferations 359 Transbronchial biopsy, pulmonary carcinoid tumors 309 Transcription factors, adrenocortical tumors 151 Transforming growth factor β(TGFβ), adrenocortical tumors 149 Transitional cell borderline ovarian tumors 449 Transitional cell carcinoma (TCC) 18 of the bladder, children 763 – 4 of prostate see under Prostate cancer urethral 28, 29 Transoral brush biopsy, nasopharyngeal carcinoma 118 Transperitoneal needle core biopsy, transitional cell carcinoma of prostate 55 Transplantation autologous stem cell mantle cell leukemia 550 Waldenstr¨om macroglobulinemia 571 malignancy due to see Transplantation-related malignancy organ see Organ transplantation Transplantation-related malignancy chronic neutrophilic leukemia 544 lymphoproliferative see Posttransplantation lymphoproliferative disease (PTLD) Merkel cell carcinoma 596 pediatric squamous cell carcinoma 812 – 3 sebaceous gland carcinoma 577 Transpupillary thermotherapy (TTT), uveal melanoma 716 Transrectal biopsy primary lymphoma of prostate 58, 59
855 TCC of prostate 55 Transrectal ultrasonography (TRUS) guided biopsy, small cell undifferentiated carcinoma of prostate 39 prostate sarcomas 49 Transsexuals, male breast cancer and 202 Trans-Tasmanian Radiation Oncology Group (TROG), Merkel cell carcinoma trials 584 Transurethral resection (TUR) bladder sarcoma 25 of the prostate (TURP) 28 transitional cell carcinoma 55, 56 Trauma, meningioma and 640 Tremolite, mesothelioma and 280 ‘Tri-Arc’ technique, intensity-modulated radiation therapy 125, 125 Trichilemmal carcinoma 577 Trisomy, Merkel cell carcinoma 596 “Triton tumors,” 77 Trk C, medulloblastoma in children 698 Trophoblastic disease see Gestational trophoblastic diseases Trophoblasts 533 TRUS, transitional cell carcinoma of prostate 55 TSC 1 gene 7 TSC 2 gene 7 Tuberculosis adenoid cystic carcinoma and 323 primary adenocarcinoma of fallopian tube 477 Tuberculum sellae, meningioma 641 Tuberous sclerosis complex (TSC) 8 genetics 7 renal angiomyolipoma 7, 8 renal cell carcinoma in children 760 Tubular carcinoma 230 – 5 associated conditions 230, 232, 232, 233 biology 230 classification/grading 232 clinical presentation/diagnosis 231 – 3 epidemiology 230 histochemistry 231, 232 historical background 230 pathology 230 – 1, 231, 232 prognosis 232, 233 – 4 pure type 230, 232 radial scars 230, 231 recommendations 234 recurrence 234 sclerosing type 230, 231, 231 treatment 233 Tumor classification/staging systems adrenocortical tumors 153, 778 colorectal carcinoma 752, 752 gallbladder tumors 384 germ cell tumors 2, 649 primary intracranial 649, 650 gestational trophoblastic neoplasia 537 – 8, 538, 540 – 1 lung cancers 321 lymphoma(s) gastric 362, 362 non-Hodgkin’s lymphoma, breast 195, 197 non-Hodgkin’s lymphoma, testicular 67, 68, 69 primary pulmonary 257, 258 – 9, 261, 261 Merkel cell carcinoma 584 pheochromocytoma 159 pleural mesothelioma 282 – 3, 283 primary lung sarcomas 274 renal tumors 1, 2 adult Wilms tumor 13, 13 renal cell carcinoma (RCC) 2, 760, 761 rhabdomyosarcoma 78, 78, 79, 79 stromal tumors of the ovary 455, 456 testicular tumors
856 Tumor classification/staging systems (cont.) children (prepubertal) 765 malignant mesothelioma of the tunica vaginalis (MMTV) 76 NHL 67, 68, 69 rete testes adenocarcinoma 74 – 5 rhabdomyosarcoma 79, 79 thymomas/thymic carcinoma 237, 238, 246, 247 – 8 tubular carcinoma 232 uterine sarcomas 486, 487 see also specific systems Tumor embolization, pancreatic carcinoid 377 Tumorigenesis, phyllodes tumor of the breast 211 Tumor lysis syndrome 598 Tumor markers adrenocortical tumors 149 borderline ovarian tumors 450 hCG in gestational trophoblastic disease 532, 534, 535 – 6, 537 intracranial germ cell tumors 651 – 2, 652 large cell neuroendocrine carcinoma (LNEC) of the lung 302 phyllodes tumor of the breast 212 – 3 thymomas/thymic carcinoma 244 – 5 Tumor necrosis factor (TNF), Merkel cell carcinoma (MCC) 599 Tumor-node-metastasis (TNM) staging see TNM (tumor, nodes, metastasis) staging system Tumor staging systems see Tumor classification/staging systems Tunica vaginalis 75 mesothelioma see under Malignant mesothelioma (MM) Turcot’s syndrome 695, 751 Typical serous borderline ovarian tumor 448, 448 Tyrosine kinase inhibitors, bronchioloalveolar carcinoma 317 Tyrosine kinases, uterine sarcoma 486 Ulcers, subcutaneous panniculitis-like T cell lymphoma (SPLTL) 563 Ultrasonography bile duct tumors 385 breast carcinosarcoma 223, 224 male breast cancer 203 metaplastic carcinoma 183 non-Hodgkin’s lymphoma 196 – 7 phyllodes tumor 213 gallbladder tumors 383 malignant mesothelioma of the tunica vaginalis 76 ovarian tumors 449 pancreatic tumors 368 pancreatic cystadenocarcinoma 369 prenatal, gestational trophoblastic disease 535 Zollinger – Ellison syndrome 374 Ultrastructural examination see Electron microscopy Ultraviolet (UV) light atypical fibroxanthoma and 581 Merkel cell carcinoma etiology 583, 596 microcystic adnexal carcinoma 579 squamous cell carcinoma of the conjunctiva 713 – 4 trichilemmal carcinoma and 577 uveal melanoma 716 Undifferentiated carcinoma gastric with lymphoid stroma 353, 353 salivary glands 92 United Kingdom Coordinating Committee on Cancer Research (UK-CCCR), anal carcinoma 407
SUBJECT INDEX University of Texas MD Anderson Cancer Center, breast carcinosarcoma 225 – 6 Urachal cancer, bladder 24 – 5 Uracil, nasopharyngeal carcinoma 126 Ureteral tumors, pediatric malignancy 763 Urethra, anatomy 27 Urethral cancer 27 – 37 anatomy 27 epidemiology 27 investigations endoscopic examination 30 radiological studies 30 staging 30 symptoms/physical examinations 30 pathology 28 – 30 benign lesions 28 malignant lesions 28 – 30 rare histologies 30 treatment 31 – 4 combined chemotherapy 34 cytotoxic chemotherapy 33 – 4 radiation therapy 33 surgery 31, 33 see also specific types Urethrectomy, urethral cancer 31 Uric acid bone tumors 89 Burkitt lymphoma 559 hematologic malignancy 96 Urinary tract nephrogenic adenomas 28 pediatric malignancy 760 – 4 sarcoma 25 see also specific components; specific tumors/locations Urine analysis adrenocortical tumor diagnosis 151, 778 – 9 pheochromocytoma diagnosis 157 – 8 Urothelial carcinoma 18, 19, 22 US Finland phase ll, imatinib mesylate 423 Uterine papillary serous carcinoma (UPSC) 494 – 6, 507 management 495 – 6 pathology 494, 494 – 5 Uterine sarcoma 485 – 94 benign metastasizing leiomyoma 494 carcinosarcoma 488, 488 – 9, 491 – 2, 492, 492 – 3 classification/staging 486, 486, 487, 489 surgical 490 disseminated peritoneal leiomyomatosis 494 endometrial stromal sarcoma 487, 487 – 8, 488, 492 etiology 485 hemangiopericytomas 493 incidence 485 intravenous leiomyomatosis 494 leiomyoblastoma 494 leiomyosarcoma 485, 486 – 7, 487, 490, 492, 492 – 3, 493 lymphomas 493 molecular biology 485 – 6 pathology 486 – 9 endometrial stromal sarcoma 487, 487 – 8, 488 leiomyosarcoma 486 – 7, 487 mesenchymal 494 mitotic index 487, 487 mixed epithelial/mesenchymal 488, 488 – 9, 489 myxoid smooth muscle tumors 487 pediatric 768 preoperative evaluation 486 rare tumor variants 493 – 4 recurrence/metastases 489, 490, 492 following adjuvant therapy 491 symptoms/signs 486 treatment
adjuvant in stage I and II disease 490 – 2, 491 advanced (stage III & IV) disease 492, 492 – 3, 493 chemoradiation therapy 491 – 2 chemotherapy 490 – 1, 491, 492, 492 – 3, 493 hormone therapy 492, 493 immunotherapy 493 primary therapy 489 – 90 radiation therapy 490, 491 surgical 489 – 90 see also specific types Uterine tumors 485 – 500 endometrial 494 – 6 pediatric malignancy 768 sarcomas see Uterine sarcoma trophoblastic neoplasia 534 see also specific types Uvea anatomy 715 cancers 713, 715 – 7 melanocytes 715 Uveal melanoma 715 – 7 biology/epidemiology 715 – 6 environmental factors 716 host factors 715 – 6 incidence 715 clinical presentation/diagnosis 716, 716 historical background 715 pathology 716 prognosis 717 treatment 716, 717 Vaginal bleeding, abnormal endocervical type adenocarcinoma 504 hydatidiform moles 534, 535 invasive mole (gestational trophoblastic) 536 – 7 stromal tumors of the ovary 456, 459, 461, 462 uterine sarcoma 486 Vaginal brachytherapy, endometrial carcinoma 495 Vaginal radical hysterectomy 501 Vaginal radical trachelectomy, cervical tumors 514 – 5 Vaginal squamous cell carcinoma 526 – 8 clinical presentation 526 staging/prognosis 526 treatment 526 – 7 Vaginal tumors 526 – 8 adenocarcinoma 527 endodermal sinus 528 hematologic neoplasms 528 melanoma 527 – 8 pediatric malignancy 768 sarcoma 528 squamous cell carcinoma 526 – 8 see also specific types Vagus nerve, larynx 103 Valsalva maneuver 102 Vascular endothelial growth factor (VEGF) adrenocortical tumors 149 mesothelioma angiogenesis 282, 285 Vascular endothelial growth factor receptor (VEGFR) 426 neuroepithelial neoplasms 676 – 7 Vascular tumors cutaneous 581 – 2 thoracic 270 see also specific types Vasoactive intestinal peptide (VIP) islet cell tumors 375, 756 pheochromocytoma 156 Vasogenic edema, meningioma 642 Venous thrombosis, osteosclerotic myeloma 572 Ventricles (brain)
SUBJECT INDEX choroid plexus papilloma/carcinoma 667 – 73, 801 craniopharyngiomas 705 primary intracranial germ cell tumors 653 Ventriculoperitoneal (VP) shunt, medulloblastoma/PNET 697 Verrucous carcinoma 337 – 9, 338 cervical 502 – 3 clinical characteristics 338 etiology/pathology 338 laryngeal 103 – 4 etiology 103 – 4 pathology 104, 104 presentation/diagnosis 104 treatment 104 natural history 339 oral cavity 87, 88 Verrucous squamous cell carcinoma 338 Vertebrae, chordomas clinical features 617 – 8 surgical resection 621 Vestigial pharyngeal hypophysis 113 Veterans Administration Lung Study Group, small cell carcinomas of the gastrointestinal tract 431 VHL gene 786, 789 – 90 Video-assisted thoracic surgery (VATS) primary pulmonary lymphoma 261 thymomas/thymic carcinomas 248 Villoglandular papillary adenocarcinoma, cervical 505, 505 – 6 Vimentin, breast carcinosarcoma 227 Vinblastine primary CNS lymphoma 662 see also specific combinations Vinblastine, adriamycin and dexamethasone (VAD), primary plasma cell leukemia 549 Vinblastine, ifosfamide, platinum (VeIP), ovarian germ cell tumors 473 Vincristine medulloblastoma 700 paratesticular rhabdomyosarcoma 80 pregnancy 517 – 8 see also specific combinations Vincristine, actinomycin D , and cyclophosphamide (VAC) ovarian germ cell tumors 472 paratesticular rhabdomyosarcoma 80 – 1 Vincristine, bleomycin and cisplatin (PVB), ovarian germ cell tumors 472 Vincristine, doxorubicin, and cyclophosphamide /cisplatin and etoposide (VAC/PE), cervical small cell carcinoma 511 Vincristine, etoposide, and carboplatin, retinoblastoma 718 Vincristine/nimustine (ACNU), ependymoblastoma 805 Vinorelbine esophageal adenoid cystic carcinoma 341 mucoepidermoid tumors 332 VIPoma 375, 376 Viral infection(s) mesothelioma etiology 281 see also specific infections/organisms Virilism/virilization adrenocortical tumors 143, 145 biochemical diagnosis 151 – 2
children/adolescents 777, 777, 777 – 8, 778, 780 evaluation 462 stromal tumors of the ovary 456, 459, 461, 463 Visual disturbances craniopharyngiomas 707, 708 primary intracranial germ cell tumors 651 Vitamins, medulloblastoma/PNET and 696 Vocal cord 103 Von Brunn’s nests 28 von Hippel – Lindau syndrome (disease) choroid plexus papilloma/carcinoma 668 endocrine cancers in children 781, 786 pheochromocytoma 155, 156, 788 – 9, 789 epidemiology 786 molecular genetics 786 pancreatic cancer and 367 renal cell carcinoma in children 760 von Recklinghausen disease (neurofibromatosis type I; NF1) clitoris malignancy 767 malignant peripheral nerve sheath tumors 584 – 5, 817 cutaneous 584 – 5 pediatric 817 meningeal sarcomas 628 optic pathway glioma 685 pheochromocytoma 155, 788, 789 rhabdomyosarcoma 77 vulvar malignancy 767 von Willebrand’s disease, Wilms tumor 13 Vulvar tumors 521 – 6 basal cell carcinoma 525 melanoma 524 – 5 pediatric malignancy 766 – 7 phyllodes tumors 209 rare type of carcinoma 526 sarcomas 525 – 6 squamous cell carcinoma 521 – 4 clinical presentation 521 – 2, 522 microinvasive tumors 523 recurrent 524 stages l and ll tumors 523 stages lll and lV tumors 523 – 4 staging/prognosis 522, 522 treatment 522 – 4 Vulvectomy, vulvar squamous cell carcinoma 523 Waldenstr¨om macroglobulinemia 570 – 1 Waldeyer’s ring, nasopharyngeal carcinoma 118 Warty (condylomatous) carcinoma 503 Watery diarrhea, hypokalemia and achlorhydria (WDHA) syndrome 375 Well differentiated thymic carcinoma (WDTC) 238, 241 Werner’s syndrome see Multiple endocrine neoplasia (MEN), type I Wheezing, adenoid cystic carcinoma of the lung 323 Whipple procedure, cystadenocarcinoma 369 Whipple’s triad, insulinoma 375 White asbestos (chrysotile), mesothelioma and 279, 280 White matter lesions, Langerhans’ cell histiocytosis 611, 612
857 Whole-brain radiation therapy (WBRT) chemotherapy 662 primary CNS lymphoma 661 Wilms tumor (nephroblastoma) adult 12 – 4 clinical presentation 13 criteria 12 pathology 12 – 3 staging 13, 13 treatment/prognosis 13 – 4 childhood 1, 12, 13 clear cell carcinoma vs. 760 – 1 cutaneous manifestation, pediatric malignancy 817 testicular metastases, pediatric malignancy 765 Wilms tumor 1 gene (WT1), mesothelioma vs. NSCLC 281 Wiskott – Aldrich syndrome 565 Wnt pathway, medulloblastoma/PNET 696 Women see Female(s) World Health Organization (WHO) classification ameloblastic carcinoma 726 bladder adenocarcinoma 21 brain tumors choroid plexus papilloma/carcinoma 182 germ cell tumors 649, 650 meningeal 638, 639, 639, 640, 646 esophageal carcinosarcoma 339 gestational trophoblastic neoplasia 537, 538 hematopoietic/lymphoid neoplasia 544 cervical lymphoma 514 cutaneous lymphoma 562 leukemia 543 lymphomas, rare 555, 556 lung cancers 298, 313, 315 metaplastic breast carcinoma 181 nasopharyngeal carcinoma 116 ovarian cancer 447 prostate sarcomas 46 small cell carcinomas of the gastrointestinal tract 431 thymic tumors 237, 242 uterine sarcomas 486 Wunderlich syndrome 762 Xeroderma pigmentosum (XP) pediatric basal cell carcinoma 811 pediatric melanoma 810 pediatric squamous cell carcinoma 812 Xp11 translocations, renal tumors 1, 4 – 5 pathology 4 – 5, 5 prognosis 5 Yolk sac tumor see Endodermal sinus tumor Zeiss glands 712 tumors 577 Zeolite, mesothelioma and 280 Zoledronic acid, primary plasma cell leukemia 549 Zollinger – Ellison (ZE) syndrome 374 – 5, 756 gastric carcinoid tumors, type II 358 imaging/diagnosis 374 medical management 374 – 5, 376